<%BANNER%>

Tarascan Kingship

Permanent Link: http://ufdc.ufl.edu/UFE0022858/00001

Material Information

Title: Tarascan Kingship The Production of Hierarchy in the Prehispanic Patzcuaro Basin, Mexico
Physical Description: 1 online resource (365 p.)
Language: english
Creator: Haskell, David
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: agency, hierarchy, kingship, materiality, personhood, tarascan
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The present study investigates the manner in which the king of the prehispanic Tarascan kingdom of West Central Mexico exercised his authority and established his legitimacy over subordinate lords. I argue that this authority and legitimacy was established through the king's ability to encompass the subordinate lords of his realm, in effect making them aspects of his own persona and not autonomous agents. The king's ability to achieve this effect was an inherently intersubjective phenomenon that relied on the perceptions of other social beings, most notably the commoners that the subordinate lords oversaw. By engaging subordinate lords in specific material practices such as unilateral or asymmetric exchange, the king was able to cause commoners to perceive that the king was the primary agent that endowed subordinate lords with the capacity to act meaningfully, and on his behalf. The ability of the subordinate lords to act on their own behalf was severely compromised, because the commoners who witnessed their actions were led to perceive, through the presence of material objects and the social practices behind those objects, the king's agency in the actions of the subordinate lords. I demonstrate that the king was able to encompass the subordinate lords of two sites in the Tarascan core zone, the Lake P?tzcuaro Basin. I achieve this through both ethnohistoric and archaeological investigations. I first analyze how certain discursive practices and narratives, contained in the ethnohistoric record, publicly proclaimed the histories of those social agents and material objects involved in the realization of any act, leading witnesses to perceive that he was the primary motivating and enabling agent in the actions of the subordinate lords. Additionally, I use the ethnohistoric record to identify material correlates of the realization of these material practices. Lastly, I examine the object biographies of those material objects that the king relied upon in his ability to encompass the subordinate lords. Witnesses would have made their inferences according to not only the discursive framing that surrounded particular actions or objects, but also the histories of those objects and whose agencies had gone into the production of their existence in the hands of the subordinate lords.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by David Haskell.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Gillespie, Susan D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022858:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022858/00001

Material Information

Title: Tarascan Kingship The Production of Hierarchy in the Prehispanic Patzcuaro Basin, Mexico
Physical Description: 1 online resource (365 p.)
Language: english
Creator: Haskell, David
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: agency, hierarchy, kingship, materiality, personhood, tarascan
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The present study investigates the manner in which the king of the prehispanic Tarascan kingdom of West Central Mexico exercised his authority and established his legitimacy over subordinate lords. I argue that this authority and legitimacy was established through the king's ability to encompass the subordinate lords of his realm, in effect making them aspects of his own persona and not autonomous agents. The king's ability to achieve this effect was an inherently intersubjective phenomenon that relied on the perceptions of other social beings, most notably the commoners that the subordinate lords oversaw. By engaging subordinate lords in specific material practices such as unilateral or asymmetric exchange, the king was able to cause commoners to perceive that the king was the primary agent that endowed subordinate lords with the capacity to act meaningfully, and on his behalf. The ability of the subordinate lords to act on their own behalf was severely compromised, because the commoners who witnessed their actions were led to perceive, through the presence of material objects and the social practices behind those objects, the king's agency in the actions of the subordinate lords. I demonstrate that the king was able to encompass the subordinate lords of two sites in the Tarascan core zone, the Lake P?tzcuaro Basin. I achieve this through both ethnohistoric and archaeological investigations. I first analyze how certain discursive practices and narratives, contained in the ethnohistoric record, publicly proclaimed the histories of those social agents and material objects involved in the realization of any act, leading witnesses to perceive that he was the primary motivating and enabling agent in the actions of the subordinate lords. Additionally, I use the ethnohistoric record to identify material correlates of the realization of these material practices. Lastly, I examine the object biographies of those material objects that the king relied upon in his ability to encompass the subordinate lords. Witnesses would have made their inferences according to not only the discursive framing that surrounded particular actions or objects, but also the histories of those objects and whose agencies had gone into the production of their existence in the hands of the subordinate lords.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by David Haskell.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Gillespie, Susan D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022858:00001


This item has the following downloads:


Full Text

PAGE 1

TARASCAN KINGSHIP: THE PRODUCTION OF HIERARCHY IN THE PREHISPANIC PATZCUARO BASIN, MEXICO By DAVID LOUIS HASKELL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 David Louis Haskell 2

PAGE 3

To the memory of my mother, Connie. 3

PAGE 4

ACKNOWLEDGMENTS Most prominently I thank a trio of people without whose patience and support I could not have written this dissertation. In keeping with a theme in the dissertation, any one of them could claim to be the primary agent most responsible for this work if they so desired. The first is my advisor and committee chair, Dr. Susan Gillespie, for always taking the time to help me along the way. At every step in my graduate career, Dr. Gillespie had the uncanny ability to lead me to consider topics that I had never considered or did not know how to approach yet had always wanted to tackle. The second is Dr. Helen Pollard, who was gracious enough to introduce me to Michoacn archaeology, provided much needed assistance and advice in the field, and has always made time for me whenever I needed to pick her brain or go through her collection of sources. Surely Dr. Pollard has forgotten more about Michoacn than I will ever know, and I say that with all seriousness. The third is my wife Emmy, whose love and support made it possible for me to pursue graduate school and my doctorate in the first place. I thank the members of my doctoral committee, including Doctors Gillespie and Pollard but also including Dr. David Grove, Dr. Michael Heckenberger, and Dr. Murdo MacLeod. They have my extreme gratitude for their efforts and advice in the production of this dissertation and their constructive criticisms on how to build on it in the future. I also thank my father, who always delighted in the fact that I was a black sheep in a family of engineers, and I am grateful for his help and support as I wrote this dissertation while raising my son. In that regard, the help of my wifes parents is also greatly appreciated. I thank Karin Rebnegger for her tremendous help in the field in 2005 and helping me to understand the significance of the lithic artifacts from Erongarcuaro. She also generously shared some of her data, and I eagerly anticipate her own dissertation on lithic production in the Tarascan core. I also thank Dr. Amy Hirshman for helping me understand the nature of the data relating to 4

PAGE 5

sourcing ceramic pastes in the Ptzcuaro Basin. I thank the Instituto Nacional de Antropologa e Historia for permission to conduct research at Erongarcuaro. I thank the Foundation for the Advancement of Mesoamerican Studies, Inc. for the financial support (Grant number 05036) that made the 2005 field season possible. The Wenner-Gren Foundation also provided financial support (Grant number 7308) that made the radiocarbon dating of samples and chemical characterization of obsidian samples from the 2005 field season possible. In the field, I thank Aida Castilleja of Erongarcuaro and INAH for her help in obtaining permits from INAH and advice concerning the local political landscape. I thank the many people of Erongarcuaro who made my time spent there over the years quite enjoyable, and particularly the workmen of the town who were always happy to work and just as importantly happy to discuss their thoughts on their world. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES .........................................................................................................................10 LIST OF FIGURES .......................................................................................................................11 ABSTRACT ...................................................................................................................................14 CHAPTER 1 INTRODUCTION: POWER AS ENCOMPASSMENT IN THE TARASCAN KINGDOM.............................................................................................................................16 Introduction.............................................................................................................................16 Location and Historical Development of the Tarascan Kingdom..........................................20 Encompassment of Subordinates in the Tarascan Core..........................................................25 Organization of the Presentation............................................................................................27 2 ETHNOHISTORIC AND ARCHAEOLOGICAL SOURCES AND DATA........................36 Introduction.............................................................................................................................36 Ethnohistoric Evidence...........................................................................................................37 Relacin de Michoacn...................................................................................................37 Other Primary Sources....................................................................................................42 Secondary Sources...........................................................................................................43 Archaeological Evidence........................................................................................................46 Urichu..............................................................................................................................46 Erongarcuaro..................................................................................................................50 Ceramic data from Erongarcuaro............................................................................63 Obsidian data from Erongarcuaro...........................................................................67 Mortuary Data from Tzintzuntzan..........................................................................................68 Conclusion..............................................................................................................................69 3 INVESTIGATING AND THEORIZING THE TARASCAN KINGDOM...........................71 Introduction.............................................................................................................................71 Ethnohistoric and Archaeological Investigations of the Tarascan Kingdom.........................72 Kingly Control over People: the Tarascan Bureaucracy.................................................73 State Control Over Resources.........................................................................................77 Archaeological Studies of Kingly Power and Control....................................................79 Models of State Formation: the Origins and Causes of Kingly Control.........................82 Critique of Models of State Formation and Kingly Control in the Tarascan Kingdom.........84 Circularities in Marxist Models.......................................................................................85 Ecological and Settlement Approaches in the Ptzcuaro Basin......................................88 6

PAGE 7

Political Power and the Prestige Reward Model.............................................................91 Legitimacy, Order, and Methodological Relationism.....................................................96 Conclusion..............................................................................................................................97 4 ENCOMPASSING OTHERS: CONSTRUCTING SAMENESS OVER DIFFERENCE THROUGH SOCIAL INTERACTION...............................................................................100 Introduction...........................................................................................................................100 Relationality, Personhood, and Agency................................................................................103 Hierarchy and/as Encompassment........................................................................................106 Dumonts Original Formulation....................................................................................108 Critique and Reformulation...........................................................................................110 Agency, Abduction, and Replication: Encompassment Through Social Action..................119 Gells Definition of Agency and the Role of the Index.................................................121 Primary Agents, Secondary Agents, and Causative Chains..........................................122 Abduction from Indexes and Human Secondary Agents..............................................124 Encompassment Through the Abduction of Replication...............................................131 Encompassment and/as Tarascan Political Theory.......................................................136 Conclusion............................................................................................................................137 5 METHODS FOR THE ETHNOHISTORIC AND ARCHAEOLOGICAL ANALYSIS OF ENCOMPASSMENT.....................................................................................................140 Introduction...........................................................................................................................140 Identifying Meaningful Actions and Objects.......................................................................142 Roles of Wives: Providing Food and Cloth to the Gods and the People.......................143 Objects of Bodily Adornment and the Agencies They Facilitate..................................145 Importance of Feasting in Lordly Authority.................................................................147 Object Biographies, and Objects as Records of Their Own Subject/Object Interactions....149 Acquisition of Raw Materials........................................................................................153 Production......................................................................................................................154 Patronage of Producers..................................................................................................156 Exchange.......................................................................................................................158 Rhetorical Framing and Cosmological Authentication.................................................166 Mortuary Ceremonies....................................................................................................174 Making Abductions From Cross-Cutting Investments in the Past and Present.............178 Conclusion............................................................................................................................181 6 COSMOLOGICAL AUTHENTICATION OF UACUSECHA POWER: ENCOMPASSMENT IN THE ETHNOHISTORIC RECORD...........................................183 Introduction...........................................................................................................................183 Analysis of Narrative............................................................................................................185 Elementary Categories and Their Transformation........................................................186 Syntagmatic Relations, and the Recognition of Continuity and Contrast.....................187 King as Society and Cosmos: the Production of a Sociocosmic Totality.............................190 Construction of Elementary Categories in the RM.......................................................190 7

PAGE 8

Transformation of the Elementary Categories: the Encompassing Uacusecha............193 Subordinate Lords as Chichimecs: The Replication of Legitimate Authority.....................210 Encompassment of Elementary Categories by the Uacusecha.....................................210 Position of the Subordinate Lords.................................................................................213 On Chichimecs and Transformative Capacity...............................................................215 Conclusion: The Ideological Construction of the Relationship between Subordinate Lords and the King............................................................................................................217 7 REPRESENTATIONS OF EXCHANGE AND THE PRACTICE OF ENCOMPASSMENT IN THE RELACIN DE MICHOACN............................................219 Introduction...........................................................................................................................219 Framing Noble Marriage Exchanges in the RM...................................................................220 Brides of Nobles and Wives of Curicaueri: Royal/Noble Intermarriage......................222 Rhetoric and its Indexical Consequences......................................................................225 Investiture of Caciques: the Distributed Person of the King................................................229 Death and Succession in Offices of Local Leadership..................................................229 Insignia and the Agency of Caciques to Represent the King........................................233 Insigniaand Thus Lordsas the Distributed Personhood of the King..........................236 Conclusion............................................................................................................................237 8 ENCOMPASSMENT IN THE TARASCAN CORE: THE CASE STUDIES OF ERONGARCUARO AND URICHU..................................................................................240 Introduction...........................................................................................................................240 Marriage Exchange at Erongarcuaro and Urichu................................................................241 Representations of Relations Between the King and the Lords of Erongarcuaro in the RM.......................................................................................................................241 Affinal relations in the RM....................................................................................242 Military alliance in the RM....................................................................................243 Intermarriage According to other Colonial Documents................................................245 Archaeological Evidence for Royal Marriage into Urichu...................................................248 Mortuary Evidence From Urichu..................................................................................248 Mortuary Ceremonies at Tzintzuntzan..........................................................................252 Production of Lapidary Objects in the Tarascan Core..........................................................258 Lapidary Production at Erongarcuaro and its Implications..........................................258 Wider Contextual Associations of the Lapidary Production at Erongarcuaro.............264 Bezotes, and Their Absence, at Urichu..........................................................................269 Lordly Feasting and Communal Rituals: Patterns of Ceramic Acquisition at Erongarcuaro and Urichu.................................................................................................272 Feasting Ceramic Vessels: Specialized Forms and Conspicuous Decoration...............273 Specialized vessel forms and their role in feasting................................................273 Highly decorated vessels........................................................................................274 Indicators of the Differential Acquisition of Ceramics at Erongarcuaro and Urichu..275 Ceramic Assemblages from Erongarcuaro and Urichu................................................277 Conclusion............................................................................................................................287 8

PAGE 9

9 CONCLUSION: ENCOMPASSMENT, THE TARASCAN CORE, AND IMPLICATIONS FOR THE STUDY OF THE TARASCAN KINGDOM AND ITS DEVELOPMENT.................................................................................................................291 Introduction...........................................................................................................................291 Logic and Claims of Encompassment in the Tarascan Kingdom.........................................292 Material Practices of Encompassment in the Tarascan Core................................................297 Implications for Investigating Tarascan State Formation.....................................................307 Encompassment and Tarascan State Formation............................................................308 Relations in the Core and Tarascan State Formation....................................................314 Encompassment in the Core in Relation to Encompassment in Provincial Areas........317 Guilt By Association and the Relations Between Lords in the Core and Provincial Areas..........................................................................................................................320 Directions for Future Research into the Development of the Tarascan Kingdom...............326 Conclusion............................................................................................................................330 APPENDIX A DATA ON THE OBSIDIAN ARTIFACTS FROM ERONGARCUARO.........................333 Introduction...........................................................................................................................333 B CERAMIC DATA FROM ERONGARCUARO AND URICHU WITH STATISTICAL MEASURES OF DIFFERENCE...............................................................336 Introduction...........................................................................................................................336 Ceramic Data From Erongarcuaro and Urichu....................................................................336 Statistical Tests of Difference...............................................................................................340 LIST OF REFERENCES.............................................................................................................343 BIOGRAPHICAL SKETCH.......................................................................................................365 9

PAGE 10

LIST OF TABLES Table page 2-1. Periods and corresponding local phases and time periods in North-Central Michoacn......49 8-1. Counts and percentage of total sample of obsidian artifacts from Erongarcuaro by according to color............................................................................................................261 8-2. Count and percentages of obsidian sorted by color and source, as determined by INAA and XRF...........................................................................................................................266 8-3. Number and proportion of resist decorated and feasting form sherds from excavated areas at Urichu and Erongarcuaro...................................................................................279 A-1. Results of the sourcing by chemical characterization of 50 obsidian samples from Erongarcuaro...................................................................................................................333 A-2. Counts of Tariacuri phase obsidian samples from particular sources sorted by color.......335 B-1. Ceramic Data of the Total Sherd Sample from Urichu Area 1, categorized by paste and decoration.........................................................................................................................336 B-2. Ceramic Data for the Total Sherd Sample from Area 2 of Urichu.....................................337 B-3. Ceramic Data for the Total Sherd Sample from Area 5 of Urichu.....................................337 B-4. Aggregate Ceramic Data from fields ER-01, ER-02, ER-07, ER-08 at Erongarcuaro......338 B-5. Classification of sherds by paste and presence of feasting form, negative decoration, or both from the fields ER-01, ER-02, and ER-07 at Erongarcuaro...................................338 B-6. Classification of sherds by paste and presence of feasting form, negative decoration, or both from Area 1 at Urichu..............................................................................................338 B-7. Classification of sherds by paste and presence of feasting form, negative decoration, or both from Area 2 of Urichu.............................................................................................339 B-8. Classification of sherds by paste and presence of feasting form, negative decoration, or both from Area 5 of Urichu.............................................................................................339 B-9. Counts of pipe fragments classified by paste from ER-02, ER-03, ER-05.........................339 B-10. Counts of discrete pipes (refit sherds counted as one) classified by paste from ER-02, ER-03, and ER-05............................................................................................................339 B-11. Statistical Tests of Difference for Proportions among different sherd samples...............341 10

PAGE 11

LIST OF FIGURES Figure page 1-1. The limits of the Tarascan kingdom at its height, ca. 1500 A.D...........................................20 1-2. Towns of the Lake Ptzcuaro Basin, with the lake shown in gray (at its approximate level in the early sixteenth century)...................................................................................22 2-1. Map of the archaeological site of Urichu, showing the areas of the site and surface architecture.........................................................................................................................48 2-2. Map of the site of Erongarcuaro showing the survey units and the location of the excavation units of the 2001 field season..........................................................................51 2-3. Topographic map showing the locations of the excavation units (small black squares) of the 2005 field season in fields ER-01, ER-02, ER-03, and ER-08................................52 2-4. Plan and profile drawings of the hearth excavated in unit N109E0 in field ER-02, showing the location of the radiocarbon sample................................................................53 2-5. Plan view of the burials excavated in field ER-03................................................................54 2-6. Profile view of the burials excavated in ER-03.....................................................................55 2-7. Drawing of bowl (Catalog #3082) from burial 5 in field ER-03. Agropecuaria White on Red, with motifs similar to Loma Alta ceramics illustrated by Carot (2001)...............59 2-8. Drawing of bowl (Catalog #3081) from burial 5 in field ER-03. Agropecuaria White on Red with Negative, with motifs similar to Loma Alta phase ceramics illustrated by Carot (2001)..................................................................................................................59 2-9. Olla (Catalog #3084) from burial 3 in field ER-03. Erongarcuaro Red on Orange. Nearly identical to ollas found with the burials in field ER-22, burials that dated to the Loma Alta 2 phase (see Pollard, n.d.b)........................................................................60 2-10. Drawings of spindle whorls recovered from the 2005 field season....................................60 2-11. Profiles of pipe bowl rims recovered in the 2005 field season............................................61 2-12. Drawing of an intact pipe bowl recovered in the excavations in ER-02 in the 2005 field season.........................................................................................................................61 2-13. Drawing of a nearly intact pipe recovered in a survey of ER-05 in 2005...........................62 2-14. Profile drawings of rim sherds from mini-bowls and mini-jars (everted). Both are indicators of elite activity. Lines above rims indicate rim diameters...............................62 11

PAGE 12

2-15. Excavation units of ER-02, with the provenience of temporal markers indicated by abbreviations. Archaeological period is indicated by color..............................................64 2-16. Excavation units of ER-03. Temporal markers are again abbreviated and color coded....67 3-1. Examples of the variety of hierarchical relationships of settlements, as well as the officials resident in each, as given in the Caravajal Visita................................................76 3-2. Map of places and towns in the Tarascan kingdom...............................................................80 5-1. Examples of Tariacuri phase ceramic forms suited to feasting. A. Miniature jar. B. Line drawing with profile of the vessel in A. C. Zoomorphic vessel with spout. D. Olla with spout and vertically oriented strap handle. E. Tripod bowl. F. Olla with spout and vertically oriented strap handle. All vessels were recovered at Urichu in Dr. Pollards archaeological excavations. Vessels represented in A, B, D, E, and F also have resist (negative) decoration. All photos and drawings courtesy of Dr. Helen Pollard...................................................................................................................148 6-1. Characters of the legendary history of the RM and their place in the elementary categories of the narrative................................................................................................196 6-2. Map showing the locations of important places in the legendary history of the RM. Not to scale.............................................................................................................................197 6-3. Transformations of the meanings and relations of the elementary categories of the legendary history of the RM............................................................................................212 7-1. Idealized model of marriage exchange between the Tarascan king and noblemen in which the daughters (women are represented by circles) of the noblemen are given to the king in marriage, and vice versa................................................................................224 7-2. Idealized model of the exchanges of the noble economy in which noblewomen played an essential role................................................................................................................227 8-1. Fragments of obsidian drills (reamers) and a hollow drill core that indicate lapidary production at Erongarcuaro. Photograph by Karin Rebnegger and Chris Valvano.......260 8-2. Map of the survey units of Erongarcuaro, indicating those units in which a local informant claimed to have found bezotes in various stages of production (shaded in light gray).........................................................................................................................262 8-3. Percentages of sherds of dominant pastes from excavations at Erongarcuaro and Urichu..............................................................................................................................278 8-4. Percentages of wares within the feasting forms and resist decoration sample and the total sherd sample............................................................................................................281 12

PAGE 13

9-1. Different permutations of the structure of nested relations of encompassment according to the place of the perceiving person within that structure, as specific relations were differentially emphasized by those actors........................................................................305 13

PAGE 14

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TARASCAN KINGSHIP: THE PRODUCTION OF HIERARCHY IN THE PREHISPANIC PATZCUARO BASIN, MEXICO By David Louis Haskell December 2008 Chair: Susan Gillespie Major: Anthropology The present study investigates the manner in which the king of the prehispanic Tarascan kingdom of West Central Mexico exercised his authority and established his legitimacy over subordinate lords. I argue that this authority and legitimacy was established through the kings ability to encompass the subordinate lords of his realm, in effect making them aspects of his own persona and not autonomous agents. The kings ability to achieve this effect was an inherently intersubjective phenomenon that relied on the perceptions of other social beings, most notably the commoners that the subordinate lords oversaw. By engaging subordinate lords in specific material practices such as unilateral or asymmetric exchange, the king was able to cause commoners to perceive that the king was the primary agent that endowed subordinate lords with the capacity to act meaningfully, and on his behalf. The ability of the subordinate lords to act on their own behalf was severely compromised, because the commoners who witnessed their actions were led to perceive, through the presence of material objects and the social practices behind those objects, the kings agency in the actions of the subordinate lords. I demonstrate that the king was able to encompass the subordinate lords of two sites in the Tarascan core zone, the Lake Ptzcuaro Basin. I achieve this through both ethnohistoric and archaeological investigations. I first analyze how certain discursive practices and narratives, 14

PAGE 15

contained in the ethnohistoric record, publicly proclaimed the histories of those social agents and material objects involved in the realization of any act, leading witnesses to perceive that he was the primary motivating and enabling agent in the actions of the subordinate lords. Additionally, I use the ethnohistoric record to identify material correlates of the realization of these material practices. Lastly, I examine the object biographies of those material objects that the king relied upon in his ability to encompass the subordinate lords. Witnesses would have made their inferences according to not only the discursive framing that surrounded particular actions or objects, but also the histories of those objects and whose agencies had gone into the production of their existence in the hands of the subordinate lords. 15

PAGE 16

CHAPTER 1 INTRODUCTION: POWER AS ENCOMPASSMENT IN THE TARASCAN KINGDOM Introduction [T]he [Tarascan] indians would not have killed Christians if the said Cazonci [the Tarascan king] had not ordered it because the entire land does that which he commands. 1 Make food to eat for this lord, and make mantas [reams of cloth], and do not let up; be well married, and if someone enters your house, give them mantas. The king says that what you [both] give, it is really he who gives it. 2 The first quote, from a Spaniard observing the largely intact Tarascan political structure in the earliest years of Spanish colonial rule, describes the power and authority that the Tarascan king exerted over all of the subjects within his realm. This apparently high degree of political control has been the subject of investigation by both archaeologists and ethnohistorians, who have continued to remark at the kings tight grip on power and further seek to study how such a political arrangement developed in West-Central Mexico during the Late Postclassic Period (1350-1520 AD). 3 I argue, however, that these understandings of the Tarascan kings power, and more importantly how that power was produced and reproduced through practices on the ground, are incomplete. In their place, I analyze how the kings power over subordinate lords was produced through discursive and material practices as the kings encompassment of their agencies, as exemplified in the second quote above. 1 The testimony of the Spaniard Juan Lpez Patio as recorded in the Proceso contra Tzintzicha Tangaxoan, written in 1533. This document was the official record of the 1530 trial of the last Tarascan king, Tzintzicha Tangaxoan, produced on behalf of the Spanish colonial authority who presided over the trial, Beltrn Nuo de Guzmn (president of the First Audiencia of New Spain). Cited in Krippner-Martnez (2001, pp. 19). The document has been published by Scholes and Adams (1952); see also commentary in Martnez Baracs (2005, pp. 151-157). 2 Relacin de Michoacn (1956[1541], p. 208). This is part of the speech of a priest officiating the marriage between a subordinate lord and a woman of the royal court. My translation. Emphasis added. 3 Brand (1971); Beltrn (1986); Carrasco (1986); Gorenstein and Pollard (1983); Martnez Baracs (1997, 2003, 2005); Pollard (1980, 1987, 1993, 2003a, 2003b); Stanislawsky (1947). 16

PAGE 17

Especially when compared to other regions and polities of Mesoamerica, very little is known about the Tarascan kingdom. What is known is derived largely from ethnohistoric sources. Archaeological research is only beginning to narrow this gap. Furthermore, almost all of the interpretations of the ethnohistoric and archaeological records have been the result of similar theoretical and methodological programs that have focused on the organization of the bureaucratic (administrative) structure of the kingdom and/or the political economy of the state. These approaches have produced a consistent, if simplistic, argument in favor of viewing the Tarascan kings power as relatively strong and derived from his control over key resources. However, much information in the ethnohistoric, and as I demonstrate the archaeological, records goes unused by such approaches. It is precisely this kind of information, as exemplified by the second quote, that can shed the most light on just how power was conceptualized, how it was in turn made materially and rhetorically manifest, and ultimately how identities (in particular the identities of the lesser nobility) were constituted in the Tarascan kingdom. In short, it is this kind of information that can lead to a greater understanding of how power in the Tarascan kingdom worked on the ground. Additionally, the implications of the failures or risks of those practices by which power and authority were not just exercised but actively produced and simultaneously legitimated allows for the identification and analysis of the limitations or weaknesses in Tarascan kingly authority. By filling in the gaps of the knowledge of the Tarascan kingdom and analyzing the kinds of practices by which the kingdom was produced and reproduced, a more complete picture of the nature of Tarascan political authority can be posited, ultimately leading to the development of better models of how that system of political authority developed historically. This is important for a broader understanding of processes of state formation and the development of different 17

PAGE 18

kinds of states in Mesoamerica. This culture area in particular has led to many important studies of state formation, largely due to the multitude of state forms that developed throughout its geographical and temporal expanse. 4 Of particular theoretical interest is the ways in which the Tarascan kingdom apparently contrasted with the better-known (both archaeologically and historically) Aztec empire. Whereas the Tarascan king seems to have been the unquestioned ruler of his kingdom, the Aztec emperors power was checked by the power of ostensibly subordinate, but still quite powerful, rulers and lords. 5 Exactly in what ways these states and empires differed in their political structure and the exercise of power is an interesting question, particularly because they arose at almost the same time and in similar environments. In order to make this comparison as fruitful as possible, however, information such as how the Tarascan kings power was conceptualized and how this conceptualization was realized through specific practices is essential. It is not simply enough to say that the Tarascan kings power was strong. Instead, it is necessary to ask: In what ways did the Tarascan king exercise his will and assert his authority? How was the relationship between the king and the subordinate lords perceived by other members of society? What were the potential limitations of the kings power and authority? The approach undertaken here can be characterized by an endeavor to take the second quote that introduced the present study seriously, by seeking to investigate just how statements such as this, and the interactions between king and subjects that they represent, would have produced and reproduced the kings power. As such, I answer the questions raised above by 4 Prominent studies of state formation in Mesoamerica include: Blanton et al. (1993), Blanton et al. (1996), Brumfiel (1983); Demarest (1992); Flannery (1972); Flannery and Marcus (1983, 1996); Sanders, Parsons, and Santley (1979); and Sanders and Price (1968). 5 The literature on the nature of the Aztec empire is expansive, but prominent discussions of the weaknesses of the Aztec emperors power include: Berdan et al. (1996); Calnek (1982); Carrasco (1971); Gibson (1971); Hassig (1985, 1988); Hodge (1991, 1994); Smith and Berdan (1992). 18

PAGE 19

examining the kinds of material and discursive practices promoted and employed by the king in order to constitute the agencies of his subordinates as aspects of his own person. Following Dumont (1979, 1980) and others who have adapted his insights (Sahlins, 1985, 1991; also Gillespie, 2009; McKinnon, 1991), I call this transformation intersubjective encompassment. Drawing on ethnohistoric sources I begin by analyzing how the king and upper nobility (i.e., members of the kings kin-group, who could claim descent from the legendary progenitor of the royal line) conceptualized and promulgated this conceptualization of his own power as the transformation of the lesser nobility into reproductions of the king. Secondly I analyze how material practices and the discourses that framed those practices, as represented in ethnohistoric sources, would have achieved the transformation of the agencies and identities of subordinate lords. Finally, I analyze archaeological and ethnohistoric data pertaining to three sites in the Ptzcuaro Basin (Erongarcuaro, Urichu, and Tzintzuntzan) in order to determine the extent to which those material and discursive practices that would have produced the intersubjective encompassment of the local nobility were realized in these local contexts. 19

PAGE 20

Figure 1-1. The limits of the Tarascan kingdom at its height, ca. 1500 A.D. Tenochtitlan was the capital of the contemporaneous Aztec empire, which spanned much of the rest of central and southern Mexico. After Pollard (1993, p. 5). Location and Historical Development of the Tarascan Kingdom At the time of the arrival of the Spanish conquistador Hernando Corts on the shores of modern Mexico, the Tarascan kingdom 6 was a large conquest empire second in size within Mesoamerica only to the Aztec empire. The maximal extent of the Tarascan kingdom was 6 Just like debates over what to call the Aztec empire, there is no agreement over what to call the Tarascan kingdom. Here I adopt the term Tarascan kingdom for a few reasons, both practical and theoretical. The term Tarascan is and has been more widely used in the literature than its alternatives. Recently a handful of scholars have noted the problematic origin of this term, having originated with outsiders (most likely Spaniards as they adopted a bastardized form of the indigenous term for in-law; see Martnez Baracs 2005, pp. 63-77). Due to this origin, some have argued in favor of the indigenous term Purhepecha, which is the name that the dominant ethnic/linguistic group within Michoacn calls themselves. This is also the more or less modern name of the language spoken by these peoples, who were demographically dominant in central Michoacn, at least at the time of the arrival of the Spaniards (West, 1948). However, in the Purhepecha language, the word purhepecha itself means common people (Martnez Baracs, 2005, pp. 77-84) and to label a political entity, ruled by a small minority of persons with access to resources not available to the wider populace, using the term for common people seems a contradiction in terms. This point has been made by Warren (1984, pp. 118). There is no doubt, however, that the dominant class of the Tarascan kingdom did indeed share much in common, e.g. language, with the modern-day Purhepecha people/ethnic group of central Michoacn (Pollard, 1993, pp. 15). With respect to my choice of the term kingdom in place of state or empire, I simply wish to avoid the terminological debate over what exactly constitutes a state or an empire. A cursory reading of the Relacin de Michoacn will, however, convince the reader of the fact that the Tarascan political structure was headed by what in Western political science would be called a king. Therefore I refer to the area, the realm, over which that king ruled as a kingdom. 20

PAGE 21

roughly equivalent to the modern Mexican state of Michoacn in west central Mexico, but also included parts of the modern states of Guanajuato, Quertaro, Jalisco, and possibly Guerrero and Mexico (see figure 1-1). In order to expand and maintain this large conquest empire, the prehispanic rulers of the Tarascan kingdom developed highly organized secular and religious bureaucracies that were responsible for the collection of tribute, the raising of armies, and the governance of the people in general (Beltrn, 1986; Brand, 1971; Carrasco, 1986; Lpez Austin, 1981; Martnez Baracs, 1997, 2003, 2005, chapter 1; Pollard, 1993, 2003a; RM, 1956). Archaeological work focused on the development and functioning of the Tarascan kingdom has been sparse. The existing archaeological research has documented that in the political and demographic core of the state, the Lake Ptzcuaro Basin in central Michoacn (as defined by Gorenstein and Pollard, 1983; Pollard, 1980, 1993, 2003b, 2003d; see figure 1-2), local chiefs or leaders were integrated into the state system and began to share the markers of nobility defined by and perhaps also emanating from the capital city of Tzintzuntzan during the Tariacuri phase (1350-1520 A.D.) (Pollard, 2003c; Pollard and Cahue, 1999). This transition in elite material culture is mirrored (and occurred slightly later) in other areas of Michoacn, as demonstrated by: Michelet (1992, 1995) and Michelet, Ichon, and Migeon (1988) in the Zacapu Basin just to the north of the Ptzcuaro Basin; by Healan and Hernndez (1999) and Macas Goytia (1989, 1990, 1996) in the Cuitzeo Basin in northeastern Michoacn; at the garrison site of Acmbaro (Gorenstein, 1985); and by Acosta and Uruuela (1997) in the Sayula Basin of Jalisco at the western margins of the Tarascan kingdom. Research at the capitals (see below) of Tzintzuntzan (Acosta, 1939; Cabrera Castro, 1977-78, 1987, 1996; Castro Leal, 1986; Gal, 1946; Noguera, 1931; Rubn de la Borbolla, 1939, 1941) and Ihuatzio (Noguera, 1931; Rubn de 21

PAGE 22

la Borbolla, 1939) reveal only shallow cultural deposits in the ceremonial precincts at these sites, indicating a brief era of major construction befitting imperial capitals. Figure 1-2. Towns of the Lake Ptzcuaro Basin, with the lake shown in gray (at its approximate level in the early sixteenth century). The three capitals that were home to resident members of the Uacusecha (the royal dynasty) are indicated with the larger dots and type. Due to the lack of archaeological investigation into the nature and development of the Tarascan kingdom, most reconstructions of its history and functioning have relied on the available ethnohistoric record. First and foremost among these is the Relacin de Michoacn (RM), which was written circa 1540 and contains information about the native secular and religious bureaucracies, marriage practices, warfare, the history of the royal dynasty and how it extended its rule, and the coming of the Spaniards (for more information on the RM, see Chapter 2). Based primarily on this document, and in particular the official state history of the royal dynasty as related by a member of the indigenous priesthood, most if not all interpretations of the processes that led to the formation of the prehispanic Tarascan kingdom hold that the Lake Ptzcuaro Basin was inhabited by various ethnic groups, some having arrived there later than others. These differences are not believed to have constituted significant barriers to interaction and intermarriage between the various groups (Kirchhoff, 1956; Lpez Austin, 1981; Michelet, 22

PAGE 23

1996; Pollard, 1993, 1994). By the time the Spaniards arrived in Tarascan territory, the Lake Ptzcuaro Basin was predominantly populated by Purhepecha (Tarascan) speakers. This same state history contained within the RM indicates that prior to this era of consolidation among the populations of the Lake Ptzcuaro Basin this region was politically fragmented, with multiple polities ruled by autonomous chiefs or leaders, some of which might have exerted some degree of control over other towns or villages (RM, 1956, pp. 14-150; an example of hierarchical relationships is Coringuaros apparent control of Itziparamucu). In addition to being ruled by autonomous lords, the polities or their ruling lineages each had their own patron god that they worshipped (RM, 1956; also Corona Nez, 1957; Pollard, 1991). The RM (1956, pp. 150-155) states that this fragmented political landscape was transformed only a few generations prior to the arrival of the Spaniards. The descendants of a group of newcomers to the Ptzcuaro Basin, the Uacusecha, managed to establish themselves as the dominant political force in the area (RM 1956, pp. 14-150; also Kirchhoff, 1956; Lpez Austin, 1981; Pollard, 1993, pp. 87-92). The Uacusecha were the ancestors of the king in this official history but in contemporary society (i.e., at the time of the production of the document, were those members of the kings kin-group, i.e. the upper nobility. Uacusecha means eagle in the Purhepecha language, the dominant language in central Michoacn at the time of Spanish contact. According to the legendary history in the RM (1956, pp. 14-150) Uacusecha members ultimately settled at the three towns of Tzintzuntzan, Ihuatzio, and Ptzcuaro (see figure 1-2). The Uacusecha leaders then embarked upon a program of expansion and conquest outside the Lake Ptzcuaro Basin core and came to subordinate populations of various other ethnic or linguistic groups such as Nahuatl speakers (Nahuatl was the language of the Aztecs of the Basin of Mexico at the time of Spanish contact), Otomis, Matlatzincas, and Tecos (Brand, 1943; 23

PAGE 24

Pollard, 1993, pp. 92-105; Stanislawsky, 1947). The three Uacusecha settlements became the capitals of the Tarascan kingdom and primary residences of the upper nobility. In time the faction that resided at Tzintzuntzan achieved primacy and became the preeminent capital of the kingdom. Here I use the terms Uacusecha, royal dynasty, and upper nobility interchangeably to refer to the royal and sub-royal lineages that resided in Tzintzuntzan, Ihuatzio, and Ptzcuaro and which the RM claims had descended from a single progenitor (RM, 1956, pp. 14-150). By the time the Tarascan kingdom was intruded upon by Spanish forces, a large and apparently extremely loyal bureaucracy carried out the wishes and directives of the Tarascan king. In addition to state level officials such as the governor or second-in-command, the war captain, and regional governors, there was apparently a contingent of courtesans (the achaecha). The king also controlled access to local offices, and local leaders or caciques were only recognized as legitimate if they had been chosen by the king and endowed with the insignia of office by the king (Pollard, 1993, 2003a; following the description in the RM, 1956, pp. 203). 7 As evidenced in early colonial documents (see Gorenstein, 1993, pp. xiv-xv; Krippner-Martnez, 2001, pp. 18-35; Warren, 1985, pp. 226-227), these local level lords were extremely loyal to the king even in the uncertain political atmosphere of competition for supremacy between the Tarascan king and Spanish colonial authorities that existed until the execution of the king at the hands of Spanish colonial authorities in 1530. There were also a number of officials responsible for monitoring the collection and storage of tribute of different kinds of goods, as well as 7 In the text of the RM, only some offices are given names in Purepecha, and this has led to some confusion and disagreements over the indigenous names for these offices. Most relevant to this study of kingly control of subordinate local level lords is the fact that these lords are called caciques in the RM. Cacique is a name for leader or petty king that was adopted from the indigenous inhabitants of Hispaniola, brought to New Spain, and applied to a wide range of lords and rulers (Gibson 1964, p. 36). There is some disagreement over which indigenous name applied to the officePollard (2003a, pp. 50-51) believes these local lords were the angmencha, while Aguilar Gonzalez (2006) contends that they were the caracha-capacha. Given such disagreements, I will either refer to the nobles that held this office either as local level lords or caciques. 24

PAGE 25

officials who oversaw craft specialists, at least in the capital (RM, 1956, pp. 174-178; also Pollard, 1993, pp. 45). The state bureaucracy also included a well-developed priesthood, with specified roles and responsibilities for each rank (RM, 1956, pp. 181-182). The picture of the Tarascan kingdom that has emerged from investigations of the ethnohistoric record therefore is one of a highly centralized state in which the king and his representatives directly oversaw a wide range of aspects of daily life that affected the maintenance of the state, such as agricultural production, craft production, personal and group identity, and tribute collection. For these reasons, the Tarascan kingdom is generally contrasted with the more loosely integrated Aztec empire in terms of the extent of direct control at the local level by the state (see Chapter 3). Encompassment of Subordinates in the Tarascan Core In order to discuss whether, how, and to what extent the agencies and personae of specific subordinates were encompassed I present and analyze ethnohistoric as well as archaeological data. Different ethnohistoric sources provide different pictures or viewpoints from which to understand the workings of the Tarascan kingdom and the kings power within that political structure. I rely heavily on one document in particular, the RM, in order to analyze how the king and the upper nobility (the Uacusecha) represented the power of the king and his relation to subordinate lords to the populace at large as well as Spanish colonial authorities. Through the construction and proclamation of history and by rhetorically framing (Keane, 1997, pp. 17, 26-27; see Chapter 5) exchange relationships by attempting to influence how those relations and the objects they involved should be perceived, the king sought to set the terms by which status as a lord entailed a derived status in relation to the king (i.e. as the inferior reproduction of the king). The representations of the RM also present a picture of how the king sought to encompass subordinate lords in practice, at least from the perspective of the king and upper nobility. One of 25

PAGE 26

the prominent means of encompassing lords, at least according to these representations, was through the largely unilateral establishment of gift relations that included the transfer of royal brides and insignia of office. As I discuss in greater detail in chapters 5 and 7, such gifts were essential to the execution of the lords duties, and if they derived from the king these gifts would have indexed the superior agency of the king within the agentic capacities of the subordinate lords. I also suggest that, due to the importance of providing food to commoners as represented in the RM (1956, p. 206), the material objects necessary for food distribution (ceramic vessels) would have been potential targets of exclusive production and gift-giving on the part of the king (following Bray, 2003; Dietler, 2001, 2003). Data within the RM also indicate that such representations themselves, made publicly and made often, were part of the process of encompassment itself, by discursively and rhetorically framing the nature of the gift relation between king and subordinate lord and ultimately constructing the actions of the subordinate lords as the agency of the king. These representations of relations and the practices by which they were realized are then evaluated with reference to the practical realization of relations of power and subordination between the king and specific subordinate lords in the Tarascan core. For this, information from more locally oriented ethnohistoric sources is utilized in order to determine whether or not such practices would have been realized. More critically, archaeological evidence from two sites that were home to subordinate lords, Erongarcuaro and Urichu, is analyzed in order to determine the extent to which those practices that should have been essential to the encompassment of subordinate lords were actually realized. As discussed below, there is evidence that Urichu was in some way subordinate to Erongarcuaro and that Erongarcuaro was superordinate in relation to other lordly centers, and so this case study also examines the evidence for the relations 26

PAGE 27

between the lord of Erongarcuaro and other subordinate lords. The potential for superior/subordinate relations between lords themselves could have had important consequences for the nature of the relationship between the king and subordinate lords, and I examine the impact of relations among lords on the relation between the king and those lords. Organization of the Presentation I have thus far introduced the basic problem of power in the Tarascan kingdom and why exactly the Tarascan kingdom should matter in studies of state formation in Mesoamerica and the world more generally. I present the ethnohistoric and archaeological data sets in Chapter 2. With regard to the ethnohistoric sources, I discuss the nature of the sources, the context of their production where relevant (particularly in the case of the RM, as I rely heavily on the representations contained within that document), and how they are used in this dissertation. I also discuss the archaeological research carried out at the sites of Erongarcuaro and Urichu, including the nature of the investigations carried out at the sites (e.g. the extent of surface surveys and excavations) and some of the findings and data that these investigations produced. Preliminary discussions of the analyses of the artifacts recovered at the site, such as sourcing by chemical characterization, are also included in Chapter 2. In Chapter 3 I first summarize existing models of the Tarascan kingdom and the processes that led to its formation. While of course varying in certain particulars and emphases, these models focus largely on the composition (formal characteristics) of the state bureaucracy, the role of state control over status markers in holding it together, governmental control of the land-tenure system and control over other important resources, and the interrelationship between all three. Pollard has been at the forefront of theorizing and modeling the rise of the Tarascan kingdom, which she believes occurred due primarily to differential access to key resources in the 27

PAGE 28

context of ecological change. All of the investigations and theorizations of power in the Tarascan kingdom are systemic and view macrophenomena as determinative of the nature of interactions between actors. As noted above, they have produced a largely coherent picture of the Tarascan kingdom as a well-integrated political structure in which the king benefited from this high level of organizational integration and consequently wielded unquestioned power. I go on to argue in Chapter 3, however, that the existing models and the underlying assumptions of their theoretical foundations have great difficulties investigating how legitimacy and order were produced by the royal dynasty. I discuss in this chapter why the failure to address the legitimacy of the king is analytically and theoretically problematic. Furthermore, models that are determinative fail to fully utilize the rich descriptions of king/subordinate interactions contained within the ethnohistoric record. Therefore I also argue that the adoption of a relational methodology can investigate how legitimacy and order were produced in the Tarascan kingdom, in part by analyzing the representations of interactions in the ethnohistoric record. As discussed in Chapter 4, then, the adoption of a relational ontology (Gosden 1994, 1999, pp. 119-120) allows for the analysis of the actors that composed the bureaucratic organizational structure of the Tarascan kingdom not as bounded and unchanging entities but as entities that were continually being encompassed within the personhood of the king. More specifically, Dumonts (1979, 1980) concept of hierarchy as the relation of the part to the whole is adopted in order to analyze how subordinate lords were incorporated within the kings persona and why this should have resulted in the superior valuation of the king. I modify Dumonts model, which is largely static, by incorporating Turners (1984) critique and reformulation of Dumonts definition of a hierarchical relation. Turners key insight is that the syntagmatic relation by which the 28

PAGE 29

dominant or encompassing term transforms the subordinate term into a replica of itself is a defining feature of the relation of encompassment. Concomitantly, the dominant term must also establish itself as the paradigmatic instantiation of that syntagmatic relation of replication and substitution. This reformulation merges with the distinction drawn between primary and secondary agency by Gell (1998, pp. 17-21, 35-38, 44) as well as the discussions of representation of Gell (1998, p. 98) and Keane (1997, pp. 7-8, 230). Primary agents are those agents that make manifest some secondary agent that, acting as a representative of the primary agent, brings about some event that the primary agent had willed and set in motion by creating the secondary agent and sending it out into the world. In such arrangements, the proximate effects of the secondary agent, and thus the very agency of the secondary agent, are encompassed within the greater agency of the primary agent, who had willed those effects to fruition. In Turners terms, the primary agent exists as the syntagmatic principle by which the secondary agent is made manifest and endowed with a capacity for the execution of the will of the primary agent. All of this requires the abduction of agency, a process in which the capacity and intent to achieve the realization of some act or material effect is attributed to one agent and not another (Gell, 1998, pp. 13-16). Finally, I discuss how within any relation of encompassment, depending as it does on replication as a tension between sameness and difference, there can be different ways of interpreting the same relation or relations due to the recognition of the differential importance of sameness or difference in relation to the other. Chapter 5 presents methods for analyzing both ethnohistoric and archaeological evidence in order to examine how and through what practices the encompassment of subordinate lords was produced and reproduced. Just as importantly, this chapter discusses how the two sources of 29

PAGE 30

information can be integrated, in a manner that subordinates neither source to the other. In attempting to integrate the ethnohistoric and archaeological data, I assume that the discursive practices as presented in the RM and as presented to subjects of the Tarascan kingdom in the past would have been persuasive and likely went a long way in constituting relations of intersubjective encompassment, but that they could not do so without material objects. The fact that some of those representations of encompassment contained within the RM explicitly involve the transferal of material objects indicates the necessity of the material world. More generally, it should not be taken for granted or unquestioningly assumed due to the manner of their representation in the RM that these discursive practices were realized in all times and in all places. Here archaeological data can be used to at least determine whether or not the kinds of scenes in which those discursive practices could be brought to fruition were produced. Furthermore, because encompassment is an intersubjective phenomenon that relies on participants and witnesses perceiving the roles of the participants, those discursive practices as represented in the RM would not have been effective if they ran counter to the material conditions of the actions that they were framing. Indeed, as Keane (1997, pp. 17, 20-25) discusses, radical dissonance between the two would invite open reflection on the roles of the participants, which would have damning consequences for projects of encompassment. Therefore both the discursive and material practices in which intersubjective encompassment would have been achieved should have been realized in the local contexts of Erongarcuaro and Urichu, and these practices should be evident in the archaeological record. Chapter 5 discusses the interpretive framework in which artifact source data, distributional data, and mortuary data can be interpreted in order to reveal whether or not the king unilaterally endowed subordinate lords with, or otherwise preserved an indexical association with, those objects necessary to the 30

PAGE 31

fulfillment of their roles and responsibilities (thereby encompassing the agencies of the subordinate lords). Chapter 6 investigates how relations between the king and subordinate lords were represented as the encompassment of the latter by the former in narratives concerning the actions of the kings ancestors in the past. First, these narratives constructed the synthetic and encompassing nature of the kingship through the ability of the kings ancestors to encompass what Sahlins (1985a, pp. 77, 89) calls elementary categoriescategories of ecological, social, and supernatural import so fundamental to existence that together they represented the king(ship) as the synthesis of the entire sociocosmic universe (including, of course, all of society and its individual members). More specifically related to the encompassment of the lesser nobility was the claim in the narratives and speeches contained within the RM that offices of local lordship originated with foundational actions by the kings ancestors. The origination of the local office with the kings agency was paired with the categorical replication of these officeholders as sharing the identity of the king (as Chichimecs). Because the offices that supposedly defined the identity, that is the rights and responsibilities of the subordinate lords, were created by the kings ancestors the promulgation of this discourse on history was one way that the king actively constituted the agencies of the subordinate lords as extensions of his own agency and person. In sum, then, I discuss how the lesser nobility was encompassed in the arena of rhetorical discourse as they were merged within the kings personhood as secondary and derived reproductions of the king. In contrast, Chapter 7 examines how the upper nobility claimed in the text of the RM that the king encompassed the lesser nobility in practice through ongoing exchange relations that were foisted upon the subordinate lords and unilaterally constituted by the king. Through this 31

PAGE 32

unilateral constitution, the king was able ensure that the objects so transferred, and the very act of transferal, would continue to be abducted as manifestations of his agency, and his agency alone. As revealed in the Gilberti dictionary, furthermore, the insignia of office (particularly the bezotes) were particularly defining in constituting the officeholder precisely as an officeholder. To continue this line of reasoning, I discuss how the material associations of the objects and royal women so crucial to a lords successful execution of his duties reveals how the claims that these women and insignia derived from, and continued to be associated with, the king would have encompassed the lord and his agencies. Additionally, I analyze the speeches made at such exchange events in order to reveal how they would have facilitated this process of ensuring the intersubjective abduction of the primary (exclusive) agency of the king and the encompassing nature of the king in relation to the subordinate lord. While the promulgation of a typified (Parmentier 1987, p. 115) view of these exchange practices in both local and kingdom-wide contexts represents one method whereby significatory relations between the king and objects so exchanged could have been preserved, it cannot be assumed that these representations were actually presented to local audiences. Nor can it be assumed that such discursive practices could have overridden the ability of witnesses to make their own abductions of agency with respect to their own perceptions of the material contexts that constituted the nature of the exchange relation between the king and the lord. Therefore in Chapter 8 I evaluate the material evidence at Erongarcuaro and Urichu concerning both the production of scenes at which agents of the state would have discursively framed the exchange acts and the objects so exchanged, as well as the regimes of production, acquisition and exchange that would have constituted the capacities for participation in those exchange relations with the king. Furthermore, the relationship between the two lords of Erongarcuaro and Urichu 32

PAGE 33

is examined in order to determine whether it was constituted through the same practices as the relationship these lords had with the kings. Ethnohistoric information concerning intermarriage with the royal dynasty and mortuary data from Urichu is also analyzed in order to determine whether or not lords at these sites were participating in bride exchange with the king or royal dynasty and the extent to which in-marrying women were still being represented as associated with the king. Here the mortuary data from Urichu, in which there is a small sample but which nonetheless presents a gulf between the wealth of female burials in comparison to male burials, is compared to similar mortuary data from the capital of Tzintzuntzan in order to evaluate how males and females at both sites are being presented and represented in terms of their personhood and the relations that composed their identities. Lastly, I examine the evidence for lapidary production at Erongarcuaro and discuss the implications of its presence at this subordinate site when there appear to have been no real transportational advantages to the state in allowing it there. I conclude, therefore, that such production, especially the production of bezotes, was used by the lords of Erongarcuaro in actions in which they sought to elevate their own status in relation to other lords. This potentially took many forms, which I discuss. However, chemical sourcing data on the obsidian used in the lapidary enterprise at this site reveal that the obsidian was likely obtained from the king, and so the kings role and agency were still manifest in the locally produced lapidary objects. It was the potential for the differential emphasis in investing agency (Gell, 1998, 66; see Chapter 4) into, and abducting agency from, the objects so produced that allowed the lords of Erongarcuaro to present their role in the state system to themselves and others in as self-serving a manner as possible while still acknowledging the encompassing nature of the king. 33

PAGE 34

Chapter 9 concludes the present study with a summary and discussion of the evidence, both ethnohistoric and archaeological, for the encompassment of subordinates by the Tarascan king. I conclude that the Tarascan king was able to directly engage subordinates in asymmetrical exchange relationships, thus reconstituting the agencies of those subordinates as aspects of his own agency. As revealed primarily through the archaeological evidence, however, those practices by which the encompassment of subordinate lords was realized were not without some potential significatory ambiguity, which those subordinate lords could, in their own way, exploit. Therefore I discuss this relation between the encompassing nature of the king and the subordinates ability to enhance their own status. This potentially contested relation also has important implications concerning the myriad options open to both the king and subordinate lords for subverting each others power. I go on to discuss the wider implications of a focus on the construction and practice of social relations predicated on the logic of encompassment for the investigation of the processes that gave rise to the Tarascan kingdom. Potentially the most important implication of the logic of encompassment is the recognition that to the extent that everybody knew the business of everyone else, the total field of relations in which the kings encompassing nature was made manifest would have impacted the perceived value of the kings person and likely tip the balance to the king in what might otherwise be symmetric relations between the king and a local lord. Through this discussion, a more complete and detailed picture of the ways in which the Tarascan king was powerful and the manner in which this power was materially and pragmatically produced and reproduced can be developed. This more detailed picture is essential to any model of the Tarascan kingdom that hopes to incorporate concepts such as legitimacy, authority, and ideology; in short, it is essential to the position that there is more to state 34

PAGE 35

formation than the differential accumulation of resources and the defense of that resource gap through unabashed violence (whether physical or symbolic; Bourdieu, 1977, pp. 190-197). 35

PAGE 36

CHAPTER 2 ETHNOHISTORIC AND ARCHAEOLOGICAL SOURCES AND DATA Introduction In order to investigate the manner in which the king claimed to have encompassed subordinate lords in the Tarascan core and evaluate the material realization of relations of encompassment, I use both ethnohistoric and archaeological data. I analyze the representations of the Relacin de Michoacn (hereafter abbreviated as RM), the preeminent documentary source on the prehispanic Tarascan kingdom, in order to examine how the upper nobility explained the kings relationship with society. In turn, this emic view of how those relationships should have worked in practice is used in order to formulate an archaeological approach to relations of encompassment. In this chapter I discuss the RM, its contents, and the context of its production. This introduction to the RM helps introduce the basic interpretive approach that is the basis of the analyses of the representations of the kings power in the RM throughout the present study. I also describe other ethnohistoric sources that are utilized in the present study. These sources provide data that are essential to relating the representations of the RM to the material objects that the actions represented in the RM would have required, to providing background information concerning the two archaeological sites used in the present study, and to confirm the realization of actions similar to the potentially idealized representations of the RM. I also discuss the archaeological sites of Erongarcuaro and Urichu, which are located in the Lake Ptzcuaro Basin. Data from these two sites are the foundation of my analysis of the realization of material and discursive practices through which the king would have encompassed the lords of these two sites. Therefore I discuss the nature of the archaeological research and the data sets that I use in this dissertation. The site of Urichu and the work carried out there has been described at length elsewhere (Pollard, n.d.a; Pollard and Cahue, 1999), and so I only summarize 36

PAGE 37

the research and results of that research. The research at Erongarcuaro, including my own, has not yet been completely published, and so I describe the results of the research at that site at greater length below. Ethnohistoric Evidence Multiple ethnohistoric sources are used as sources of information on the Tarascan kingdom, although I have consulted only a limited number of primary sources. Foremost among these is the Relacin de Michoacn. I also utilize published versions of other documents that describe certain aspects of the state structure or relate information concerning specific towns or locales and their histories. In my use of other ethnohistoric sources I rely on the partial publication of those sources or of the information they contain within secondary sources, along with discussions and background information contained within those secondary sources. Here I detail information concerning those documents and publications that I utilize in the ethnohistoric investigations that are contained in the present study. Relacin de Michoacn The Relacin de Michoacn 8 is now housed in the Escorial library in Madrid, Spain. It consists of 140 leaves (folios) and contains 44 color illustrations. It has been published on numerous occasions, the first being in 1869. In the present study I cite the 1956 edition. My reasons for this are twofold. In my experience this edition is the most complete edition commonly found in university libraries in the United States. I also cite this edition in order to facilitate comparison and cross-referencing citations with earlier scholarship that has utilized this publication. The recent excellent edition of the RM (Alcal 2000), along with accompanying studies, published by the Colegio de Michoacn deserves mention as well. Also worth noting is 8 The full title is Relacin de las ceremonas y ritos y poblacin y gobierno de los indios de la provincia de Michoacn. 37

PAGE 38

the online publication of the entire text of the RM (but not the illustrations) by the Colegio de Michoacn, together with indices and a function that cross-references page numbers of the folios of the original document in several of the published Spanish language editions. An English translation of the RM by Eugene Craine and Reginald Reindorp was published in 1970, although due to certain deficiencies compared to the many (especially more recent) Spanish editions I do not cite this edition. Although the exact date that the original document was written is not known, it must have been written between 1538 and 1541 (Stone, 2004, p. 7). It was therefore produced less than 20 years after the arrival of the first Spaniards to the region and the peaceful submission of the last native king (Warren, 1985). Furthermore, this time span was less than 11 years after that king was killed at the hands of Beltrn Nuo de Guzman (President of the Second Audiencia of New Spain), the king having been charged with conducting pagan religious practices and interfering with the Spaniards ability to exploit indigenous labor (Krippner-Martnez, 2000; Martnez Baracs, 2005; Warren, 1985). The authorship of the document is not definitively known, although Warrens (1970-71) suggestion that the Franciscan friar Jernimo de Alcal wrote the document has been largely accepted by other modern scholars, even to the extent that the recent edition published by the Colegio de Michoacn (Alcal, 2000) credits Alcal as the author. The document was composed with the help of several members of the indigenous nobility, most prominent of which was Don Pedro Cuinierngari, the adopted brother of the last Tarascan king (Kirchhoff, 1956; Krippner Martnez, 2001, chapter 3; Stone, 2004). Don Pedro Cuinierangari had been functioning as the indigenous governor of Michoacn since the death of the last king (Martnez Baracs, 2005, pp. 298-299). Not properly of the royal dynasty, he was 38

PAGE 39

adopted as a brother by the last king upon the arrival of the Spaniards and the confusion and turmoil that ensued. Originally the document consisted of three main parts. The first, describing the practices and deities of the indigenous religion, has unfortunately been lost save for one folio (Kirchhoff, 1956). The second section contains the legendary history of the ancestors of the king, as related by a member of the indigenous priesthood whose responsibility it was to tell this story at a yearly festival (Kirchhoff, 1956, p. xx). The third part contains ethnographic elements such as descriptions of the structure of the state bureaucracy and priesthood, the marriage practices among the different classes of society, the investiture ceremonies of local lords, making war and burying the war dead, etc. This section also contains an account of the arrival of the Spaniards and other events that followed until the death of the last native king in 1530. The RM was produced in an era rife with conflict and intrigue, as both the Spaniards and the Tarascan king and his subjects sought to negotiate, influence, or dictate the terms of their coexistence. The early colonial period saw the rapid splintering of the indigenous nobility as they were often forced to choose sides among the various Spanish civil and religious authorities (Martnez Baracs, 2005; Warren, 1985). Vasco de Quiroga, oidor of the Second Audiencia of New Spain (1530-1538) and first bishop of Michoacn (1538-1565), allied himself with members of the indigenous nobility in their fight against the cruel encomendero Juan Infante, a fight which spanned decades (Martnez Baracs, 2005). This alliance was tested, however, when Quiroga moved the seat of the bishopric as well as the civil government from Tzintzuntzan, the capital of the prehispanic Tarascan kingdom, to Ptzcuaro in 1538, a move that some members of the indigenous nobility opposed (Martnez Baracs, 2005, chap. 6). Spanish interests were hardly monolithic, and most members of the Franciscan order appear to have supported these 39

PAGE 40

members of the indigenous nobility against Quiroga (Martnez Baracs, 2005, p. 261). Clearly the production of the RM and the testimony of the indigenous informants that made this production possible took place in a highly contested political environment. The fact that Tzintzuntzan is referred to as Mechuacan or the Ciudad de Mechuacan, a title the settlement could no longer claim after Quirogas move to Ptzcuaro, reveals some level of bias in the RM in favor of the indigenous nobility from the former capital (Martnez Baracs, 2005, p. 305). The document is undoubtedly a product of the many different viewpoints of those who participated in its production. Other scholars have detailed the ways in which those viewpoints provide a selective or biased account of indigenous society, the relation of certain members of the indigenous nobility to Spanish colonial authorities both civil and religious, and the relations amongst the indigenous nobility themselves (Krippner Martnez, 2001, chapter 3; Roskamp, 1998; Stone, 2004). The playing out and furthering of different and competing viewpoints and agendas in and through the production of the document are too numerous to mention and beyond the scope of this dissertation. It is important, however, to point out that the document contains a view of indigenous society held mainly by members of the upper nobility, as they were the main, if not only, collaborators in its production (Kirchhoff, 1956; Krippner Martnez, 2001, chapter 3; Roskamp, 1998; Stone, 2004). These members of the upper nobility likely sought to secure a strong position for themselves in the new colonial order, and presenting the kings power in an all-encompassing light would have furthered that agenda. The representations of the practices and performances of the state within the RM could also have been presented in a nostalgic light, in other words as idealized remembrances of a past that never actually existed. It is also possible to read some of the speeches given by priests and other actors within representations of such 40

PAGE 41

events as a call to arms for the lords and the populace of the Tarascan kingdom in the colonial context in which these representations were being produced, as discussed by Stone (2004). These are all important points to consider when analyzing the RM, but I interpret the document in terms of how a model of kingly power and legitimacy can be developed and used as a base for further ethnohistoric and archaeological investigation. It is clear that the indigenous informants, priest included, would have had motivations to make their claims as explicitly as possible and had to customize them to Spanish as well as Tarascan audiences (Stone, 2004). With regard to the indigenous Tarascan audience, I suggest that the arguments put forth in favor of the upper nobilitys privileged place in indigenous society should still have been made according to underlying principles of indigenous logic and reasoning (Haskell, 2008). That is, the defense of the upper nobilitys position in indigenous society and in the context of the colonial encounter should have been produced to satisfy members of indigenous society, even if Spanish authorities would have also been a target of persuasion. The scant amount of time between Spanish intrusion into Tarascan territory and the production of the RM would not have been enough to disrupt and destroy deeply held views and beliefs among members of indigenous society (particularly the upper nobility) about how and why the king should rule. I have made these points at greater length elsewhere (Haskell, 2003, 2008). Therefore I examine the representations in the RM of the legendary history of the Tarascan royal dynasty and the exchanges by which the relationships between king and subordinate lords were supposedly consummated in order to analyze the belief of the nobility that the king did encompass subordinate lords, but also how that encompassment could have been achieved. If the RM is a claim made for strong kingly power and the kings encompassment of society, including subordinate lords, then the representations of practices (king/subordinate lord interactions) 41

PAGE 42

should have been formulated in accordance with that principle of kingly encompassment. That is to say, the upper nobility who wished to portray the king as an encompassing being should, as evidence of that encompassing nature, have chosen to represent interactions between the king and subordinate lords as evidence of, or in support of, the kings encompassing nature. The actions and interactions that were represented should have been, therefore, interactions that would have produced relations of encompassment. The representations of the RM can be analyzed from an emic point of view, but this emic viewpoint can lead to a model of how encompassment would have, from an emic and indigenous point of view, been produced in social action. Other Primary Sources I also utilize one sixteenth century dictionary of the Tarascan (Purhepecha) language. This dictionary, originally entitled Vocabulario en lengua de Mechuacan, was written by Friar Maturino Gilberti and first printed in 1559. Gilberti was a Franciscan friar who was active in Michoacn in the middle part of the sixteenth century and wrote numerous works on the Tarascan language and translations of Christian lessons into that language. The Vocabulario has since been published as the Diccionario de la Lengua Tarasca o de Michoacn (1962), with a preliminary note by Jos Bravo Ugarte from which the above information is drawn. This edition is a reproduction of an earlier one published by Peafiel in 1901. Another important document for the case study at hand is the Caravajal Visita. The document describes an inspection made by Antonio de Caravajal in late 1523 and early 1524. Five fragments of the document are known, each pertaining to a major cabecera (head-town or local primary center) within the Tarascan kingdom and the towns subject to it. These fragments were found in disparate locations in the Archivo General de Indias in Seville, Spain, by J. Benedict Warren (Warren, 1963). The fragments record the names of the towns, whether or not 42

PAGE 43

a town has a lord or a calpisque (an adaptation of calpixque, the Nahuatl name of a bureaucratic office that is frequently translated into English as steward; see Gibson, 1971), what town each town is subject to, the names of any nearby bodies or sources of water, and the names of any nearby mountains. The text of the fragment pertaining to Uruapan is published in Warren (1963), the complete text of all five fragments is published in Warren (1977), and a summary of the data of all five fragments is published in Warren (1985). The context of the production of the document and the purpose of Caravajals survey are discussed in Warren (1963). Essentially, the survey was made in order to provide information to serve as the basis of granting of the first encomiendas (territories in which the grantee could have the land worked by resident indigenes as a money-making enterprise; see Simpson, 1950; Warren, 1985, chapter 3) in Michoacn. Of particular relevance to the present work is the fact that one fragment lists Erongarcuaro as a cabecera, states that it was directly subject to the king (in Tzintzuntzan), and furthermore lists Urichu as one of its subjects (Warren, 1977, pp. 404-408). Secondary Sources I also refer to documents or data within documents that have been published, described, and discussed by other scholars. Because I often rely to some extent on the analyses and discussions of these other scholars, I have chosen to simply cite the secondary sources rather than the primary sources directly. Both Garca Alcaraz (1976) and Pollard (1994) analyze native land titles and the Relaciones Geogrficas. The land titles that these authors discuss were produced in different contexts and at various points in history as petitions and claims brought before Spanish colonial authorities required some form of documentation. The Relaciones Geogrficas, on the other hand, were produced between 1579 and 1585 at the order of the king of Spain. Each individual Relacin Geogrfica answered a standardized set of questions concerning an individual town. In 43

PAGE 44

the present study I discuss the analyses of the Relaciones Geogrficas and land titles by Garca Alcaraz (1976) and Pollard (1994) in order to characterize the nature of the Tarascan kings power (see Chapter 3). In order to examine the role of material objects in abductions (inferences; see Chapter 4) concerning the relationship between the king and agents speaking on behalf of the king I also use Stones (2004, pp. 268n14) discussion of the testimony of natives according to the Relacin de Cuiseo de la Laguna (one of the Relaciones Geogrficas) that when judges held staffs, they were believed to be speaking on behalf of the king. I also make use of data from a set of documents from the town of Carapan in the Sierra Tarasca to the west of the Ptzcuaro Basin. These documents, which Roskamp analyzes and discusses, together compose what Roskamp (1998) characterizes as a corpus of interrelated documents. The corpus is composed of the Cdice de Carapan, the Cdice Plancarte, the Genealoga de los caciques de Carapan, the Lienzo de Carapan, and the Lienzo de Ptzcuaro. 9 As Roskamp (2003) discusses, these documents are interrelated, as they refer to many of the same events and some were used as the basis for the production of others. Together, the documents describe events ranging in time from the prehispanic era (including the legendary history of the founding of Carapan) to the end of the sixteenth century. They were used in a series of land disputes by the townspeople of Carapan in the seventeenth and eighteenth centuries (Roskamp, 2003, p. 306). These documents are used in particular to expand on the practice, revealed in the RM, of prehispanic visitations in towns of the kingdom by high ranking indigenous officials, for the purpose of linking archaeological data to ethnohistorically described 9 Roskamp (2003, pp. 307n9) notes that the Cdice de Carapan was called the Cdice de Carapan II or Genealoga de los caciques de Carapan by Mateos Higuera (1948) in the first published reference to this document, who erroneously thought it was the genealogy published by the Tarascan scholar Nicols Leon. Furthermore, the original of the Genealoga de los caciques de Carapan has since been lost (Roskamp, 2003, p. 309). The location of the original of the Cdice Plancarte is also currently unknown (Roskamp, 2003, p. 308). Roskamp also states that he has found two pages that are the last two known pages of the Cdice Plancarte in the Archivo Parroquial de Chilchota. 44

PAGE 45

practices of royal or noble visitation. The documents have been published in isolation at various times and in various forms (Roskamp, [2003] conveniently summarizes all prior publications of these documents). For the sake of simplicity, however, I refer to the descriptions and conclusions of Roskamp (2003). Lpez Sarrelangue (1965, pp. 241, 245) also gives information from primary sources that the daughter of the lord of Erongarcuaro married a lord from Ihuatzio in 1579. As noted in Chapter 1, Ihuatzio was one of the three capitals of the Tarascan kingdom in which members of the royal dynasty resided at the time of Spanish contact. Two different documents indicate the union of the noblewoman from Erongarcuaro with the lord from Ihuatzio. One document is in the Centro de Documentacin Histrica de Chapultepec while the other is in the Archivo General de la Nacin (see Lpez Sarrelangue [1965, pp. 241, 245] for further information on these documents). Lpez Sarrelangue merely summarizes the information given in these documents. However, for my purposes the mere fact that this marriage took place between a town with which this case study is concerned and a lord from a town where the lords were members of the upper nobility is sufficient. I discuss the significance of this marriage in chapter 7. Finally, I make use of a part of a document pertaining to Erongarcuaro that is published by Martnez Baracs (2005, pp. 362-363). This document, a viceregal mandate from 1563 (located in the Archivo General de la Nacin; see Martnez Baracs [2005, p. 363n16] for full location information), rules on a petition by the widow and children of the recently deceased indigenous governor of Michoacn, Don Antonio Huitzimengari. Don Antonio Huitzimengari was a son of the last indigenous king of the Tarascan kingdom, Tzintzicha Tangaxoan (Martnez Baracs, 2005, p. 298; Lpez Sarrelangue, 1965, p. 169; RM, 1956, p. 166). In the part of the document 45

PAGE 46

contained in Martnez Baracs (2005, pp. 362-362), viceroy Don Luis Velasco states that some residents of Erongarcuaro related to him that they had worked a certain parcel of land within the limits of their town. They did this for the benefit of their community because Don Antonio had claimed a tax on the land from them as tenants. Archaeological Evidence In this section I describe the archaeological sites of Erongarcuaro and Urichu and the research that was carried out at these sites. This background information then provides a backdrop for introducing the data sets that are employed in the course of the dissertation. Because Urichu has been described elsewhere by Pollard (Pollard, n.d.a; Pollard and Cahue, 1999), I briefly discuss the nature of the site and the archaeological materials recovered at that site. I devote more space to a discussion of the site of Erongarcuaro, because it has not been discussed at any great length elsewhere. I also discuss two sets of burial data from the capital of Tzintzuntzan. These data are used as a comparative case for the mortuary data from Urichu, in which there is a marked difference between the treatment of males and females. Urichu The archaeological site of Urichu is located just south of the modern town of San Francisco Uricho (see figure 2-1). Research there was carried out from 1990 to 1996 by Dr. Helen Pollard, consisting of a total surface survey and extensive excavations (Pollard, n.d.a; Pollard and Cahue, 1999). The excavations revealed an occupational history dating back to the Classic period (the Loma Alta 3 period in the regional sequence; see table 2-1). The site was divided into 9 topographical areas, and excavations were undertaken in areas 1, 2, and 5. The excavations in areas 1 and 5 were located near pyramidal structures found within those areas. The Area 1 excavations yielded numerous burials dating to the Tariacuri Phase (1350-1520 AD), the period associated with the Tarascan kingdom and therefore the dominance of the region 46

PAGE 47

by the capital of Tzintzuntzan. These burials were associated with a residential structure, and this is believed to have been the residence of the local noble family. One of the burials contained a bezote (lip-plug), indicating that it was the burial of a lord (Pollard and Cahue, 1999). Taken together, the burials contain numerous Tarascan polychromes and other elite ceramics, copper/bronze ornaments such as bracelets and earrings, spindle whorls for spinning cotton, and other elite paraphernalia. 47

PAGE 48

Figure 2-1. Map of the archaeological site of Urichu, showing the areas of the site and surface architecture. Underneath this Tariacuri phase residential structure were layers of cultural deposits associated with the Loma Alta 3 phase. 48

PAGE 49

Table 2-1. Periods and corresponding local phases and time periods in North-Central Michoacn. Adapted from Pollard 2008. Period Local Phase Dates Late Postclassic Tariacuri 1350-1525 AD Middle Postclassic Late Urichu 1000/1100-1350 AD Early Postclassic Early Urichu 900-1000/1100 AD Epiclassic Lupe 600/700-900 AD Epiclassic Jaracuaro 550-600/700 AD Classic Loma Alta 3 350-550 AD Late Preclassic Loma Alta 1 & 2 150 BC-350 AD Middle Preclassic Chupcuaro 500 BC-150 BC In Area 5, excavations yielded a shallow overlay of Tariacuri phase materials to a depth of roughly 40-50 centimeters. Below this, a tomb that was used predominantly in the Lupe/La Joya and Early Urichu phases (Epi-Classic to Early Postclassic) was found intact. The tomb therefore dates to the era before the rise of the Tarascan kingdom. The individuals interred are also believed to have been elites. They were able to procure goods such as shell from both the Pacific Ocean and the Caribbean Sea through extra-regional trade networks. The ceramic assemblages reflect influence from Central Mexico; some vessel forms are generally similar to those known from Tula, Hidalgo. A spectacular example is the three ceramic flutes, with blue pigment, similar to flutes known from Central Mexico. These ceramics are made of local pastes, however, and therefore appear to be locally made copies in lieu of actual imports. Pollard (n.d.a) interpreted the materials excavated in Area 2 of Urichu as indicative of a commoner residential area. Occupation in this area was apparently limited to the Middle and Late Postclassic periods. No burials were recovered in this area. The ceramic data from Urichu used in this dissertation included those artifacts recovered from the excavations in areas 1, 2, and 5 carried out by Pollard. The ceramic data were restricted to those sherds that were fully analyzed by Dr. Pollard and her graduate students according to paste, surface finish, decoration, and vessel form. Furthermore, I limited the analyses of the 49

PAGE 50

ceramic data to sherds recovered in those levels that Pollard has assigned to the Tariacuri phase according to her report to the Instituto Nacional de Antropologa e Historia (Pollard n.d.a). These data were graciously made available to me for the present study by Dr. Pollard. The data set for Area 1 of Urichu consists of 2839 sherds. The ceramic objects interred with the burials of Area 1 were not included within this sample. The data set from Area 2 of Urichu consists of 1892 sherds. The data set from Area 5 of Urichu consists of 7652 sherds. No obsidian data from Urichu were analyzed for this dissertation. Neither the total surface survey nor the excavations at Urichu recovered any evidence of lapidary production. Erongarcuaro The site of Erongarcuaro is located roughly 2 kilometers north of Urichu and encompasses the modern town of Erongarcuaro as well as a substantial part of the land to the west and north of the modern town (see figure 2-2). Led by Dr. Pollard, a total surface survey was carried out in 2001, and the site was divided into 39 fields, based primarily on the division of the surrounding area into agricultural fields. Excavations in 2001 were carried out in fields 1, 7, and 22. The excavations in fields 1 and 7 revealed shallow cultural deposits limited to the Late Postclassic (Tariacuri phase) (Pollard n.d.b). The excavations in field 22 revealed a significant occupation dating to the Loma Alta period, including the remains of a stone walled structure (Pollard n.d.b). 50

PAGE 51

Figure 2-2. Map of the site of Erongarcuaro showing the survey units and the location of the excavation units of the 2001 field season. 51

PAGE 52

Figure 2-3. Topographic map showing the locations of the excavation units (small black squares) of the 2005 field season in fields ER-01, ER-02, ER-03, and ER-08. In 2005, more excavations led by myself with the assistance of Dr. Pollard were carried out at the site of Erongarcuaro (Haskell and Pollard n.d.; Haskell 2006). Excavations were carried out in fields ER-02, ER-03, ER-01, and ER-08 (see figure 2-3). The excavation units in fields ER-01 and ER-08 (a total of seven 2m by 2m units) again demonstrated shallow cultural deposits, limited to the Tariacuri phase. These excavations reached subsoil at a depth of 20 to 30cm. The units in field ER-02 (a total of 9 2m by 2m units) revealed predominantly Tariacuri phase deposits, and a hearth was found in the field ER-02 excavations from which a radiocarbon sample was taken at a depth of 50cm below datum (see figure 2-4). This sample was determined 52

PAGE 53

to date to AD 1450-1510 and AD 1600-1620 (calibrated) at a 68% confidence level (one sigma), or to AD 1430-1530 and AD 1550-1630 (calibrated) at a 95% confidence level (two sigma). Another radiocarbon sample was taken from 44 cm below this first sample (94cm below datum) and was determined to date to AD 890-980 (calibrated) at a 68% confidence level (one sigma), or to AD 810-840 and 860-1000 at a 95 % confidence interval (two sigma). Associated with this second radiocarbon sample were several refitted sherds that, based on vessel form, can be attributed to the Lupe phase. The ceramic evidence and radiocarbon evidence support one another in both of these cases. Figure 2-4. Plan and profile drawings of the hearth excavated in unit N109E0 in field ER-02, showing the location of the radiocarbon sample. 53

PAGE 54

Figure 2-5. Plan view of the burials excavated in field ER-03. 54

PAGE 55

Figure 2-6. Profile view of the burials excavated in ER-03. In addition to numerous ceramic and lithic artifacts, a copper bell was found in the excavations in ER-02. Copper bells, and objects made of copper in general, were restricted to members of the nobility and are therefore indicators of elite occupation of an area (Pollard, 1987, 1993). Additionally, numerous ceramic pipes (for drawings of examples see figures 2-11 through 2-13) were recovered in the ER-02 excavations and small spindle whorls (see figure 2-10) for spinning cotton (an indicator of elite females as defined by Pollard and Cahue, 1999), both of which indicate an area occupied by elites and in which ritual and ceremonial activities were carried out (Pollard, 1993; Stawski, 2008). In field ER-03, four Loma Alta period burials were excavated in four contiguous 2m by 2m units located near the center of the field (see figures 2-5 and 2-6). These burials were dated according to the ceramic objects that they contained, all typical of the Loma Alta period, some of which were very similar to ceramic vessels associated with the Loma Alta burials excavated in field ER-22 in 2001 (see figures 2-7 through 2-9), which were well-dated by numerous radiocarbon samples (Pollard, n.d.b). Additionally, these four excavation units contained a significant amount of obsidian prismatic blades made from Pachuca obsidian, identifiable not only due to their unique color (pale yellowish-green) but also because they are thin. Two such pieces of yellowish-green obsidian from the ER-03 excavations were submitted for chemical characterization, and were indeed revealed to be from the Pachuca obsidian source in the Mexican state of Hidalgo. This is also comparable to the excavations of ER-22, in which 55

PAGE 56

numerous Pachuca prismatic blades were recovered. Along with the burials were a few scant pieces of compressed clay, either of a house floor or wall, indicating that some kind of structure was associated with the burials. The levels above these burials yielded mixed cultural deposits spanning the Loma Alta to Tariacuri phases. A radiocarbon sample associated with the burials and recovered at a depth of 71 cm below surface was analyzed, but returned a date of AD 1310 to 1360 and 1390 to 1410 (calibrated) at a 68% confidence level (one sigma) or AD 1300 to 1420 (calibrated) at a 95% confidence level (two sigma). This date was rejected, however, because the ceramic (see figures 2-7 Through 2-9) and lithic evidence associated with the burials (first encountered as a depth of only 40 cm below surface) were clearly related to the Loma Alta phase. Away from this area of the field, other pits were excavated that were close to the hill separating field ER-02 from ER-03 (see below). These pits contained mixed cultural deposits, with a few sherds that were attributed to the Late Urichu Phase (Middle Postclassic) on stylistic grounds, as well as Loma Alta, Lupe, and Tariacuri phase sherds similar to those found in the other units of ER-03. While the evidence of occupation during the phases between the Loma Alta period and the Tariacuri phase is admittedly small, it is nonetheless present and indicates that this area of the site of Erongarcuaro had an apparently unbroken occupational history. In the area between fields ER-02 and ER-03 is a small hill that is very likely the result of human activity (field ER-05 is actually the eastern slope of this hill while the remainder of the hill is covered with overgrowth, and a modern residence or field house sits atop the hill). Due to the elite nature of some of the artifacts recovered in the 2005 excavations surrounding this hill (i.e. in fields ER-02, ER-07, and ER-03), I suggest that this hill was a pyramidal structure that was a focal point of elite/religious activities. As discussed above, the excavations in field ER-02 56

PAGE 57

recovered numerous indications of elite occupation, including the copper bell mentioned above, spindle whorls with small diameters used for spinning cotton), numerous pipes, and finely decorated ceramics and ceramics associated with rituals and/or elite activities such as feasting based on their form (such as mini-bowls: see Pollard, 1993, pp. 203-204; also Stawski, 2008) (see figures 2-10 through 2-14). In the fields surrounding this pyramidal platform, evidence for obsidian lapidary production was recovered in excavations. This evidence consists of numerous pieces of ground obsidian that are similar to ends of the drills used in obsidian lapidary production at the Aztec city-state of Otumba, as described and illustrated by Otis Charlton (1993). A local informant also showed us (the archaeologists who worked at the site in 2005) obsidian bezotes that had been broken and discarded at some point in their manufacture, and he told us that he had found these pieces in fields adjacent to those fields (ER-02 and ER-03) in which the obsidian drill pieces were recovered in archaeological excavations. Finally, the presence of multiple colors of obsidian debitage (see Chapter 7) can be used as an indicator of lapidary production, following Pollards analysis of lithic sites in the capital of Tzintzuntzan (Pollard, 1993, p. 43). A sample of 38 pieces of obsidian from the excavations in ER-02 (as well as two pieces from ER-03) was sourced using both neutron activation and x-ray fluorescence under the supervision of Dr. Michael Glascock at the Missouri University Research Reactor. As I detail in Appendix A, roughly half of obsidian sample analyzed by neutron activation and x-ray fluorescence from the ER-02 excavations came from the Ucareo/Zinapcuaro source complex, over which the Tarascan kingdom apparently exerted some control, and whose presence is largely limited to elite areas at Urichu (see Chapter 3). Furthermore, the green obsidian samples were sourced to obsidian sources outside of the boundaries of the Tarascan kingdom and 57

PAGE 58

therefore would have been subject to control by the kings cadre of long-distance merchants (Gorenstein and Pollard, 1983; Pollard, 1992; see Chapter 3). The small sample of red/black obsidian came from the same Ucareo/Zinapcuaro source complex that the state apparatus appears to have controlled. I discuss the implications for kingly encompassment of the lords of Erongarcuaro in Chapter 8, and so here I will merely point out that the obsidian evidence supports the conclusion that this area of the site was an elite area. This area was obtaining obsidian through the state system of distribution, which only members of the nobility would have participated in. Bezotes were only worn by nobles and members of the nobility likely controlled the distribution of bezotes, as I discuss in Chapters 7 and 8. It would therefore be logical if such production took place among the elites themselves, what Ames (1995) calls embedded production, or was attached (Brumfiel and Earle, 1987; also Costin, 2001) to the elites living in this part of the site. The few sherds from the excavations in field ER-03 that can be attributed to the Late Urichu phase (Pollard, personal communication, June 2006; they also are similar in decoration to the Copujo ceramic group she identified at Urichu: see Pollard, n.d.a) based on stylistic characteristics were found in a unit close to this same hill. Among these sherds is a rim sherd, highly decorated with red and white paint and resist decoration, and sherds that can be refitted into a hollow cylindrical support for a tripod bowl. Another set of refitted sherds, highly decorated with red and white paint and resist decoration, are also part of a tripod bowl. The vessel form and decoration of these sherds, admittedly scant, are evidence that this area of the site was a locus for elite activity prior to the Tariacuri phase, and raise the possibility that the pyramidal structure existed prior to the Tariacuri phase. 58

PAGE 59

Figure 2-7. Drawing of bowl (Catalog #3082) from burial 5 in field ER-03. Agropecuaria White on Red, with motifs similar to Loma Alta ceramics illustrated by Carot (2001). Figure 2-8. Drawing of bowl (Catalog #3081) from burial 5 in field ER-03. Agropecuaria White on Red with Negative, with motifs similar to Loma Alta phase ceramics illustrated by Carot (2001). 59

PAGE 60

Figure 2-9. Olla (Catalog #3084) from burial 3 in field ER-03. Erongarcuaro Red on Orange. Nearly identical to ollas found with the burials in field ER-22, burials that dated to the Loma Alta 2 phase (see Pollard, n.d.b). Figure 2-10. Drawings of spindle whorls recovered from the 2005 field season. 60

PAGE 61

Figure 2-11. Profiles of pipe bowl rims recovered in the 2005 field season. Figure 2-12. Drawing of an intact pipe bowl recovered in the excavations in ER-02 in the 2005 field season. 61

PAGE 62

Figure 2-13. Drawing of a nearly intact pipe recovered in a survey of ER-05 in 2005. Figure 2-14. Profile drawings of rim sherds from mini-bowls and mini-jars (everted). Both are indicators of elite activity. Lines above rims indicate rim diameters. 62

PAGE 63

Ceramic data from Erongarcuaro The artifacts that comprise the data set from Erongarcuaro are those coming from the total surface survey and all excavated levels in fields 01, 07, and 08. In order to limit the study to the Tariacuri phase, the artifacts of some excavated levels are not included because of evidence that they pertain to periods prior to the Tariacuri phase. These excluded levels are limited to level 6 and below in pit N109E0, due to the radiocarbon date from a depth of 94 cm and the presence of ceramics that appear to not be of the Tariacuri phase, instead most likely pertaining to the Epi-Classic or Early Postclassic period (discussed above; see figure 2-11). I determined which phases the materials from the levels of the excavated units in ER-02 and ER-03 were associated with, and therefore which levels should be included in the analysis of Tariacuri phase material relations between the king and the lords of Erongarcuaro. Dr. Pollard provided essential advice, however, on the phase association of specific artifacts. 63

PAGE 64

Figure 2-15. Excavation units of ER-02, with the provenience of temporal markers indicated by abbreviations. Archaeological period is indicated by color. 64

PAGE 65

Due to the short amount of time that could be spent in Erongarcuaro in the summer of 2006 conducting the analysis, only a sample of the ceramics from the excavation units from field ER-02 could be analyzed completely. Complete analysis included: paste determination, classification according to vessel type, determination of surface finish, and identification of decoration type (incising, excising, appliqu) and elaboration (number of decorative techniques used, number of colors of paints used, if any). Again, due to the short amount of time that could be spent in the lab during the summer of 2006, Dr. Pollard was kind enough to help with the analysis and determined the paste of the sherds for the vast majority of the sample that was analyzed completely. This had the benefit of insuring comparability with the ceramic data from Urichu. I performed the remainder of the classification and subsequent quantitative analyses that are included in Chapter 8. The excavation units from which the ceramic artifacts were analyzed completely coincided with those units in which Karin Rebnegger, a graduate student at Michigan State University, is conducting a full analysis of the lithic artifacts. The units chosen for the full lithic analysis were selected based on the presence of bezote fragments in those units. The end result is that all artifacts of 6 out of the 9 excavation units in field ER-02 were fully analyzed, ceramic artifacts included. In addition, all rim sherds of all excavation units were fully analyzed. These sherds were also included in the analysis contained in Chapter 8. Of Tariacuri phase ceramics, only body sherds from the three excavation units not selected were not fully analyzed, although they were classified according to vessel type, surface finish, and decoration (but not paste). Because they were not fully analyzed they have not been included in the analysis. The ceramic data from the 2001 excavations in ER-01 and ER-07 at Erongarcuaro have also been included in the sample used in this analysis. Dr. Pollard determined that all levels in these excavations pertained to the Tariacuri phase (Pollard, n.d.b). During the lab season in 2003 65

PAGE 66

the ceramic artifacts were typed and classified according to the same criteria enumerated above. Dr. Pollard classified all sherds according to paste, and I and another student of Dr. Pollards (Erica Begun, now a graduate student at the University of Iowa) classified the sherds according to the remainder of the criteria. In total the ceramic sample from the elite area of Erongarcuaro (excavated areas in ER-01, ER-02, ER-07) included 5,839 sherds. Due to the mixed nature of the deposits of the excavation units in field ER-03 (figure 2-12), only ceramic artifacts that demonstrably pertain to the Tariacuri phase, namely pipes and miniature vessels (bowls and jars), from this field are included as part of the data set here. These artifacts are included only to demonstrate that a significant proportion of pipes and mini-bowls from this elite area of the site are Tecolote Orange wares, and are not used to measure the proportion of ceramic artifacts of one paste to all ceramic artifacts of the Tariacuri phase. These sherds were not, in other words, included in the total sherd sample. 66

PAGE 67

Figure 2-16. Excavation units of ER-03. Temporal markers are again abbreviated and color coded. Obsidian data from Erongarcuaro The obsidian artifacts from all survey and excavation units are currently being analyzed by Karin Rebnegger, a graduate student under Dr. Pollard at Michigan State University. She is in the process of analyzing the obsidian data from Erongarcuaro, Urichu, and Pollards data from 67

PAGE 68

Tzintzuntzan and writing her dissertation based on these analyses. I discuss the obsidian artifacts only insofar as they relate to the identification of lapidary production in the elite area of Erongarcuaro and the sources of obsidian used in that production. I discuss the artifacts that were directly related to such production either as tools (i.e., the obsidian drills) or as by-products (e.g. discoid beads and hollow drill cores) (see Chapter 8). I also use Karin Rebneggers color classification data because they demonstrate the prominence of green and other colors of obsidian that are indicative of lapidary production (Pollard 1993, p. 43). I do not, however, discuss the form, function, production, and consumption of the other obsidian artifacts recovered at Erongarcuaro. For such analyses and a discussion of her findings, interested readers should see Rebnegger (2008). Mortuary Data from Tzintzuntzan Mortuary data from Tzintzuntzan that are compared with the mortuary data from Urichu are drawn from the published findings of the fifth field season at Tzintzuntzan (Rubn de la Borbolla, 1944) and the unpublished field notes of Rubn Cabrera Castro (Cabrera Castro, 1977-1978) in concert with the description of the tenth field season that has been published (Cabrera Castro, 1987). Rubn de la Borbollas (1944) published findings of the fifth field season at Tzintzuntzan discuss the grave goods associated with approximately 14 individuals (a definitive number could not be ascertained) that were buried next to ycata 5 on the main platform. Nine adult males and their grave goods were buried in a primary multiple interment on the northwest side of the round part of the ycata, and approximately five females were buried in a multiple secondary interment on the southwest side of the ycata. Rubn de la Borbolla (1944, pp. 130-133) does not provide a detailed list of the grave goods associated with the burials, but does discuss in general terms the 68

PAGE 69

kinds of objects associated with the male burials on the one hand and the female burials on the other. Cabrera Castro directed the tenth field season at Tzintzuntzan, during which a number of burials associated with ycata 3 of the main platform were excavated. He described the results and interpretations of the findings of the field season in Cabrera Castro (1987). For data concerning the individual burials and the associated grave goods, however, I have consulted his field notes (Cabrera Castro, 1977-1978). Dr. Helen Pollard was kind enough to provide me with this resource. Both of these sources (Cabrera Castro, 1977-1978; Rubn de la Borbolla, 1944) are appropriate data sources for the present work, in which I compare the kinds of burial goods associated with males and females at Urichu to the burial goods associated with males and females at Tzintzuntzan. Conclusion The data contained within the RM present an emic view of the place of the king in indigenous society. The upper nobility and priests likely sought to describe the kings power in a favorable light as a means to enhance their standing in the eyes of the Spaniards or potentially remind their indigenous subjects of the nature and origins of the kings power, or both. In representing the nature of the kings power from an emic point of view, however, what is important is that they should also have represented specific practices in accordance and support of that view of the nature of the kings power. In other words, the RM presents a model not only of how the upper nobility conceptualized the kings power and his relationships with his subjects, but also how those relationships and concomitantly how the nature of his power worked in practice. 69

PAGE 70

It is these representations of the nature of the kings relationships with his subordinates as well as how those relationships should have worked in practice that can be related to other ethnohistoric evidence (such as land claims and marriages) as well as archaeological evidence. The representations of the RM can be related to archaeological investigations precisely because they were materialthey could not have been realized without certain objects and social interactions with those objects. The results of the research carried out at Urichu and Erongarcuaro have provided data that can be used to evaluate the degree to which those material interactions that should have been integral to constructing the kings encompassing nature were realized in specific contexts. Therefore I have described the nature of these two sites, as well as mortuary evidence from the capital of Tzintzuntzan that I use as a comparative case, and the data that are used to investigate if and to what extent the king managed to encompass the lords of these two sites. 70

PAGE 71

CHAPTER 3 INVESTIGATING AND THEORIZING THE TARASCAN KINGDOM Introduction In this chapter I summarize investigations of the structure of the Tarascan kingdom and the nature of the kings power within that structure. Most investigations of the Tarascan kingdom have concluded, for one reason or another (which I detail in this chapter), that the Tarascan kingdom was highly centralized and the kings power was quite strong and unquestioned. The explanations of how the state apparatus functioned and how the kings power came to be strong and unquestioned, whether incorporating ethnohistoric or archaeological data (or both), have produced what is basically a coherent and logical picture of the Tarascan kingdom in which the kings ability to command people and control resources were intertwined. Even the factors that likely induced the process of political centralization (i.e., state formation) by favoring militarism and the development of resource disparities within the Ptzcuaro Basin have been identified. What is missing in these approaches is a detailed understanding of what the elites actually did, or why they did what they did (e.g., reorganize craft production and monopolize foreign trade). Particularly problematic is the inability to address why the kings power should have been unquestioned and his legitimacy and the order of the state upon which it was founded universally accepted. There are a few reasons why this inability to address problems of legitimacy and order are problematic. First, the extent to which the subordinate lords remained loyal to the king, even in the early years of the colonial encounter, must be satisfactorily explained, and I suggest that the loyalty of the lords was the result of the lords perception of, and participation in, a cultural order that obliged them to follow the king. This cultural order defined who could act in certain ways and have their actions be perceived as legitimate. Second, as Baines and Yoffee (1998, 2000; see also Van Buren and Richards 2000) argue, order and legitimacy are essential aspects of 71

PAGE 72

a comparative framework that can investigate what made states and the processes that produced those states similar or different. Because order and legitimacy were produced in different ways and to varying degrees, they must be included in any studies of states and state formation that claims to be comparative. Additionally, the specific manner in which order and legitimacy were produced focuses attention on how elites interacted both with other members of society and with the material world, as they accumulated and redefined wealth and how it ought to be used. Therefore I delineate why the existing models of Tarascan state formation and reproduction have ultimately failed to address these problems of legitimacy and order. I further suggest that the adoption of a new paradigm can allow for the investigation of how the Tarascan elites actually went about fomenting the perception of an ordered state in which the kings power was unquestioned. I begin, however, by summarizing what is currently known about the Tarascan kingdom, i.e., the results of those previous investigations into the nature and origins of the kings power. Ethnohistoric and Archaeological Investigations of the Tarascan Kingdom Most of what is known about the Tarascan kingdom comes from ethnohistoric sources. Only recently has archaeologically derived knowledge begun to substantially contribute to understandings of the development and reproduction of the Tarascan kingdom. Interpretations of both the ethnohistoric and archaeological data have posited that the Tarascan king enjoyed a tight grip on what was extensive power. The investigations that have made use of the ethnohistoric and archaeological sources of information have emphasized kingly control over two kinds of things. The first was people, as in the people that composed the state bureaucracy, and through them, the populace at large. The second was resources, most prominently land but also other important resources such as obsidian and metal objects. Clearly control over people and control 72

PAGE 73

over resources were interrelated, although they are largely interpreted as reproducing one another unproblematically, as I discuss in what follows. Kingly Control over People: the Tarascan Bureaucracy As revealed primarily in the ethnohistoric record, the Tarascan bureaucracy was extensive and extremely loyal to the king. Numerous investigators (Brand, 1971; Beltrn, 1986; Carrasco, 1986; Martnez Baracs, 1997, 2003; Pollard, 2003a) have discussed the bureaucracy of the Tarascan kingdom, based primarily on the information contained in the RM and supplemented with locally oriented documents and the colonial dictionaries of the Tarascan (Purhepecha) language (e.g., the Gilberti Vocabulario discussed in Chapter 2). Because most of these discussions have relied on the RM, I cite it unless a specific author has addressed an ambiguity or disagreement over the data or unless a specific author makes a certain point that others do not. The RM records numerous secular bureaucratic offices, ranging from a second-in-command down to local level lords and tribute collectors/public works overseers. The local level lords (to which I will refer with either this phrase or the shorter term cacique which is the term most securely applied in the text of the RM to holders of this post) were, according to the RM (1956, p. 156), responsible for having agricultural fields planted for the god Curicaueri, for going to war themselves, and for ensuring that the commoners in their charge also did not neglect their duties in warfare. The military was headed by a state-level military captain, as well as four lords that appear to have functioned as regional governors of the four parts of the kingdom or which were largely military offices concerned with the defense and expansion of the borders of the kingdom (Pollard, 1993, pp. 123, 126). As discussed by Garca Alcaraz (1976, pp. 233-236, 239-241) the caciques of the towns also likely played some role in regulating the commoners access to the lands that they would work for their own livelihood. The RM (1956, pp. 153, 204-205) states that the offices that these 73

PAGE 74

caciques held were invented by the three Uacusecha members that initiated the wars of conquest that created the state, and so the caciques owed their status to the king and the royal dynasty. More directly, and as discussed by Pollard (1993, p. 124, 2003a, pp. 50-51), every holder of this office was beholden to the king because it was the king who had chosen him (even if from a pool of male kinsmen of the deceased lords) and endowed him with the insignia of nobility, at least according to the RM (1956, p. 203). Below the level of the caciques were the ocambecha (sing. ocambeti), the officials who were responsible for gathering the people for public works (including tending to state agricultural fields; see below) and collecting tribute from the people. In the colonial era, it was the responsibility of each such official to keep track of and oversee 25 houses. If the position and duties of these officials in the early colonial era were similar to their role in the prehispanic era, there must have been a large number of these officials throughout the kingdom. In addition to these local or even sub-local level officials, there were state-level officials in charge of collecting and storing almost every category of tributary good imaginable (RM, 1956, pp. 174-178). Not only were there officials in charge of goods in kind, but there were also officials, in the capital at least, who oversaw practitioners of various trades (Pollard, 1993, p. 54; RM, 1956, pp. 174-178). The state bureaucracy was vast, but it was also linear and isomorphic. In other words, all of the chains of command ended with the king. This is revealed in the Caravajal Visita (see Chapter 2), in which subordinate/superior relations between settlements were never redundant, and all settlements were ultimately and unambiguously subject to the king. For example, and of immediate concern to the present study, a small hamlet or settlement was subordinate to the town of Urichu, which was in turn subordinate to Erongarcuaro, which was in turn subordinate to the 74

PAGE 75

king. At each ascending level, furthermore, the settlements oversaw a greater and greater number of subordinate settlements (see figure 3-1). Gorenstein and Pollard (1983, pp. 123-125), using these data and the analogous settlement pattern of the Ptzcuaro Basin, explicitly state that such isomorphic relations were efficient because they regulated the flow of information and the execution of decisions made by the higher level authorities. The isomorphic hierarchical relations between sites and the administrators of those sites were, in other words, important for the coordination of action in a complex social organization. Brand (1971, p. 646), based on his interpretation of the ethnohistoric sources, suggests that the organization of the bureaucracy reveals the nature of the kings power: [t]he Tarascan state was perhaps the most efficient and powerful military entity in Mexico. This derives from the fact that the Tarascan ruler was an autocrat who shared power with no one. Within this linear bureaucracy, everyone apparently knew their roles, was loyal to the king, and executed the responsibilities of their offices faithfully. Gorenstein (1993) discusses how, sadly and ironically, the loyalty of local-level officials to the king in the chaotic first few years of competing indigenous and Spanish colonial governments contributed to the execution of the last king. According to testimony from Spaniards, caciques continued to reside part of the time in the kings court and to mark off fields whose yields was destined to augment the kings coffers (Krippner-Martnez, 2001, pp. 18-20; Warren, 1985, pp. 223-227). Both practices interfered with the ability of Spanish encomenderos to exploit their encomiendas to the fullest, and this, among other charges, sealed the fate of Tzintzicha Tangaxoan, who was executed at the order of Beltrn Nuo de Guzmn in 1530 (Scholes and Adams, 1952; Warren 1985, p. 223-227; see also Krippner-Martnez, 2001). In the prehispanic era the king did interfere in local politics when he felt it necessary, which is interpreted, again, as evidence of his supreme authority, and also 75

PAGE 76

implicitly at least of his vast knowledge of local situations and crises thanks to his expansive bureaucracy: For example, we know that the king sent a judge to settle disputes within the community of Tetlamn, near Tepelcatepec (Pollard, 1993, p. 128, citing Carrasco, 1969, p. 219). Figure 3-1. Examples of the variety of hierarchical relationships of settlements, as well as the officials resident in each, as given in the Caravajal Visita. The presence of lords in Erongarcuaro and Urichu is gleaned from other ethnohistoric sources, including the RM. Calpisque was adopted from the Nahuatl language, and is usually translated as steward (see Chapter 2). See Gorenstein and Pollard (1983) for further documentation. Note how all settlements are subordinate to only one superior settlement. Following conquests that expanded the borders of the Tarascan kingdom, Pollard (1994) believes that a policy of ethnic assimilation (Tarascanization) was important to state efforts to quell rebellions and more generally foment a sense of common identity and interests among newly subordinated lords and the larger populace. Citing mostly linguistic information from the 76

PAGE 77

Relaciones Geogrficas (see Chapter 2), Pollard suggests that the differential percentages of Tarascan speakers in the towns throughout much of the kingdom indicate they were assimilating to the dominant state culture. Material markers of participation in the state elite culture (e.g., metal jewelry and Tarascan style polychrome ceramic vessels) as well as the promulgation of the state religion marked by uniquely shaped temple platforms called ycatas were probably also important to this project of Tarascanization (Pollard, 1994). State Control Over Resources The state bureaucracy was, at least in part, designed and organized to monitor and appropriate certain key resources on behalf of the state (and the king) and was itself underwritten and controlled through those very resources (Pollard, 2003a, p. 51). A prime example of the underwriting of the bureaucracy with material rewards is the prestation of local lords with the insignia of office at their initiation by a priest acting on behalf of the king (see above), and the larger state prestige reward system (Gorenstein and Pollard, 1983, pp. 101-111; Pollard and Cahue, 1999; see below). Within this prestige reward system, Tarascan style polychrome ceramic vessels, metal jewelry, precious stones, and other foreign luxury goods are believed to have constituted the majority of objects that the state controlled access to and rewarded loyal bureaucrats with (Gorenstein and Pollard, 1983, pp. 101-111; Pollard and Cahue, 1999). Metalwork seems to have been a particular target of kingly exploitation and an important category of objects that was given to subordinate lords. Thus Pollard (1987; also Gorenstein and Pollard, 1983, p. 103), based on a document pertaining to Spanish inquiries into the indigenous metallurgical industry, posits that state control of metallurgists was a key source of kingly power and control over his subordinates. The king also likely controlled access to highly valued goods (precious stones, feathers, gold, silver, and shell) originating outside the boundaries of the kingdom through his exclusive cadre of long-distance merchants (Gorenstein and Pollard, 1983, 77

PAGE 78

pp. 103-104; Pollard, 1993, pp. 121-122; RM 1956:171-172, 178). Subordinate lords then must have performed their duties in the interests of the state in order to receive, or hold onto, such prestige objects (Pollard, 1993, p. 124). Furthermore, the local lords could claim access to land and the labor to work it only through their association with the king and participation in the state system of administration. As discussed by Beltrn (1986, 1994), Carrasco (1986) and Pollard (1993, pp. 124-126), ethnohistoric documents including the RM indicate that the lands that were designated for use by the local lords were determined by the state, and their usufruct rights were considered legitimate only if conferred by the state. Lands that could be used by local lords were just one class of land in the state organization of the land-tenure system. The state claimed the best lands for itself, either as fiscal lands whose products were destined to underwrite state projects, such as war-making or religious ceremonies, or as lands that were directly claimed as patrimonial lands of members of the royal dynasty (Beltrn, 1986, 1994; Carrasco, 1986; Garca Alcaraz, 1976; Pollard, 1993, pp. 118-119, 125; RM, 1956, p. 174). On a related point, Garca Alcaraz (1976) and Pollard (1994) have pointed out that local land titles (such as the Titulos de Tocuaro and the Titulos de Jaracuaro within the Ptzcuaro Basin) and many of the Relaciones Geogrficas commonly attribute the demarcation of town lands to the actions of some past king. In the case of the Relacin Geogrfica de Tiripeto, the informants of the town of Tiripeto claimed not only that a captain sent by a past king had founded the town at the kings orders (Relacin Geogrfica de Tiripeto, 2002, p. 15-16), but also that the people of this town claimed fishing rights in Lake Ptzcuaro that the king had granted them even though their town was located outside of the Ptzcuaro Basin (a point made by Pollard, 2003a, p. 51). Therefore not only did the king appropriate the best land for himself and other members of the upper nobility and regulate the 78

PAGE 79

caciques access to land, but at least in theory entire towns and the totality of their claims to land and other rights originated with the king. Archaeological Studies of Kingly Power and Control A major problem in attempts to investigate the nature of the Tarascan kingdom archaeologically is the dearth of archaeological research that has been realized. The largest efforts of archaeological research in the Tarascan kingdom have focused on two of the Uacusecha capitals, Tzintzuntzan and Ihuatzio (Acosta, 1939; Cabrera Castro, 1987, 1996; Castro Leal, 1986; Gal, 1946; Rubn de la Borbolla, 1939, 1941). At these sites, rich burials and monumental architecture were taken at face value as evidence of political power by tying such objects to the descriptions of adornments worn by the nobility and ceremonies in the ethnohistoric sources. By relating the archaeological evidence to the ethnohistoric recordconcerning, in particular, the burial of the king and lords at the capitalthe implicit connection is that at Tzintzuntzan, and to a lesser extent Ihuatzio, there is archaeological confirmation of kings and therefore political power (Acosta, 1939; Rubn de la Borbolla, 1939). The only large-scale excavations of secondary administrative centers in the Tarascan kingdom have been at Urichu (Pollard, n.d.a; Pollard and Cahue, 1999), in the Zacapu Basin at Milpillas (e.g., Arnauld and Faugre-Kalfon, 1998; Michelet, 1992; Michelet, Ichon, and Migeon, 1988; Pereira n.d.), in the Cuitzeo Basin at Cuitzeo and Huandacareo (Macias Goytia, 1989a, 1990, 1996), at the garrison site of Acmbaro (Gorenstein, 1985) and in the Sayula Basin of Jalisco (Acosta and Uruuela, 1997) (see figure 3-2). The regional settlement pattern study of the Lake Ptzcuaro Basin by Gorenstein and Pollard (1983) examines the disparity in population and importance between secondary centers and the capital as evidence of power differentials between the central elites and the lesser nobility. Their study, moreover, had to rely largely on the projection of demographic data from the colonial era to the prehispanic era due to a lack of 79

PAGE 80

archaeological data. Additionally, using colonial era census documents and estimations of land productivity, Pollard and Gorenstein (1980) posited that population pressure would have been a significant factor leading to the centralization of power and military expansion. Only with the survey, led by Pollard, of the southwestern part of the Ptzcuaro Basin (stretching from Pareo to Erongarcuaro) from 1990 to 2001 has actual archaeological data been used to support the conclusion that increasing population in this part of the Ptzcuaro Basin would have been an important factor in processes of Tarascan state formation (Pollard, 2008). Figure 3-2. Map of places and towns in the Tarascan kingdom. The studies of mortuary patterns by Pollard and Cahue (1999) and Pereira (n.d.) document the use of new kinds of grave goods to mark status with the advent of the Tarascan kingdom. Pollard and Cahue (1999) documented the changes in mortuary practices and funerary goods from the EpiClassic/Early Postclassic to the Late Postclassic at Urichu. In the EpiClassic/Early Postclassic, local elites displayed their wealth and status by being interred with foreign-derived wealth objects, such as shell and jadeite, as well as local copies of symbols of power adopted 80

PAGE 81

from elsewhere, such as Toltec-style ceramic flutes (Pollard and Cahue, 1999, pp. 273, 275-277). In the Late Postclassic period, however, the nobles of the site were interred with Tarascan elite status markers derived from the state system of prestige reward (see below), such as locally (within the Ptzcuaro Basin) produced polychrome ceramic vessels and adornment made of precious metals (Pollard and Cahue, 1999, p. 273). This transition is interpreted as the result of local lords adopting, or being forced to adopt, status markers that were produced for, derived from, and defined by the king and central apparatus of the state (Pollard and Cahue, 1999, pp. 276-278; also Pollard,, 1994, 2003b). Pereira (n.d.) takes mortuary evidence from the Zacapu Basin, with a consideration of mortuary contexts from known sites throughout the kingdom, and expands upon the conclusions of Pollard and Cahue (1999) by discussing differences between male and female burials in terms of wealth and the changes in these differences through time. Pereira notes that in the era prior to state formation, males were afforded more wealth objects than female, while in the era of Uacusecha dominance during the Late Postclassic females were buried with more, and more valuable, grave goods. Pereiras interpretation takes that of Pollard and Cahue (1999) a step further by positing that the marking of females as wealthy and highly valued relates to the practice of intermarriage between the Uacusecha and local lords as recorded in the RM (1956, pp. 208-210). Both elite intermarriage as a form of alliance and the use of state-derived status markers to indicate prestige and value are interpreted as important practices in the effort to reinforce the unity of the kingdom and the place of the king and Uacusecha atop that kingdom (Pereira, n.d.). Other archaeological investigations of the states control of resources have focused most intensely on obsidian, or obsidian blade technology, as this is a ubiquitous material that can 81

PAGE 82

easily be sourced through trace element analysis. Pollard and colleagues (Pollard and Vogel, 1994; Pollard, 2003c) contend that the near exclusive distribution of obsidian from the Ucareo/Zinapcuaro source complex of northeastern Michoacn at state administrative centers, and especially its dominance at the capital of Tzintzuntzan, indicate state control of obsidian from this source. Healans (1997) study of obsidian quarrying at the source suggests that, because there is a conspicuous absence of a significant amount of Tarascan polychromes or architectural indicators of state involvement, the obsidian from this source complex was not being directly extracted and worked into transportable blade cores by state-run workshops, but was probably appropriated through the tributary system from local workers. Research into the nature of the Tarascan kingdom and the kings power, whether ethnohistoric or archaeological, has mostly focused on how the Tarascan bureaucracy reflected the kings power or on the states ability to control and appropriate certain key resources. Few investigators have explicitly attempted to explain the causes and processes of Tarascan state formation, and only Pollard, in collaboration with colleagues, has attempted to explain why the resulting state was so centralized. I discuss Pollards model of state formation in the following section. Models of State Formation: the Origins and Causes of Kingly Control The historical development of the Tarascan political system, characterized by tight control over resources, and through them, people, has been explained as the result of the opportunities afforded by environmental change and differential access to prime agricultural lands. This is the model put forth by Pollard (1982, 2008), who argues that certain settlements in the Lake Ptzcuaro Basin saw their access to prime agricultural lands disappearing more rapidly than other settlements, as the lake level rose throughout the Postclassic. Pollards is the most sophisticated and comprehensive explanation of Tarascan state formation yet advanced 82

PAGE 83

(compared to the discussions of Kirchhoff, 1956; Lpez Austin, 1981; and Michelet, 1996, all of which are based on straightforward or literalist interpretations of the legendary history of the Uacusecha contained in the RM). Pollard posits that the three settlements in the eastern part of the basin, which would become the three Uacusecha capitals of the kingdom, would have taken steps to offset their decreasing land through military expeditions and ultimately wars of conquest. Pollards original formulation of the differential diminution of prime agricultural land and skirmishes over what land was left was based mostly on an understanding of the topography of the basin, the location of the three Uacusecha centers of power, and a reading of the legendary history of the royal dynasty (contained in the RM) that interpreted the repeated disagreements between the Uacusecha and other groups as a filtered remembrance of what was rapidly becoming a constant struggle for land. This model is only beginning to be supported by archaeological data (Pollard, 2008). The survey by Pollard of the southwestern part of the Ptzcuaro Basin revealed significant population growth in this part of the basin from the Early Postclassic through the Late Postclassic periods, a growth that likely would have exacerbated the need for good land to support such growing populations (Pollard, 2008). On a related point, the earlier study by Pollard and Gorenstein (1980) that examined the contact period population of the Ptzcuaro Basin and the ecological carrying capacity of the basin indicated the basin would have had to import significant amounts of food in order to sustain that population. This disparity is believed to have constituted one of the reasons for state expansion and consolidation. Also, Fisher (2005; Fisher et al., 2003) has documented the presence of modified agrarian landscapes in the Ptzcuaro Basin and their collapse in the wake of demographic disaster brought on by colonial intrusion and the ravages of epidemic diseases. This demonstrates that larger populations could have been put to use by the 83

PAGE 84

state by improving the agricultural potential of the land through terracing and other efforts of coordinated and communal labor (Fisher, 2005; Fisher et al., 2003). All of these studies posit, at least implicitly, that a well-run state administrative system and the centralization of power in the hands of the Uacusecha were born from ecologically instigated strife and at least partly solved by a strongly centralized and efficient governmental apparatus that included efficient war-making capabilities (Gorenstein and Pollard, 1983, pp. 124-130). Critique of Models of State Formation and Kingly Control in the Tarascan Kingdom All of the investigations and models of the development and functioning of the Tarascan kingdom fit rather well into a coherent picture of why the Tarascan king enjoyed what appears to have been a tight grip on power. In these models and investigations, the king wielded a great deal of power because he controlled access to key resources and prestige objects, and this control had developed due primarily to the ability of the Uacusecha to appropriate land as ecological change caused demographic stresses but also economic opportunities for the accumulation of wealth at the expense of others and political opportunities for centralization. Once the king and royal dynasty were able to appropriate land and enforce this control over land through a prestige reward system, the system of kingly control essentially reproduced itself. I contend that these interpretations of the ethnohistoric and archaeological evidence are not wrong or unimportant (and I refer to many of their conclusions throughout the present study), but that they only partially explain how the Tarascan kingdom developed and was reproduced through certain practices. The existing models and interpretations of state formation and reproduction are grounded in theories that are deterministic, in the sense that macrophenomena (e.g. systems, institutions, environmental restrictions) are the source of explanations regarding microphenomena, including the actions of individual actors (conforming to the paradigm of methodological holism; see e.g. 84

PAGE 85

Giddens, 1984; Ortner, 1984; Ritzer and Gindoff, 1994 on holistic and individualistic methodologies). They either do not account for the actual interactions that produced and reproduced the kings power or do not allow those individual actions to have explanatory power. I believe the exclusion of the actions by which authority and legitimacy were produced is problematic for a number of reasons, which I discuss below. I then suggest that the adoption of a different paradigm, one that is methodologically relational and focuses on how actors and their agencies were defined by their relation to one another and to the material world, can model and investigate the problems of legitimacy and order that have only partially been addresses by existing models of the Tarascan kingdom and its historical development. Circularities in Marxist Models In the Marxist models of the Tarascan kingdom of Beltrn (1986) and Carrasco (1986) (and also, at least implicitly, of Garca Alcaraz, 1976), the kings power was based on control of the land. Once control of the land was established, it unproblematically reproduced kingly control. Regardless of the specific viewpoint of land control espoused by either investigator (Beltrn favors the feudal mode of production, while Carrasco favors the Asiatic mode of production as the most appropriate model of the Tarascan political economy), common to both works is the belief that the king or royal dynasty claimed land and allowed the subordinate lords and the commoners to work some of that land for their own benefit. By controlling access to land, the king ensured the loyalty of the subordinate lords, who must have complied with the wishes of the king in order to preserve what power they had within the political economy. Furthermore, by preserving the most productive lands for itself, the royal dynasty preserved or even expanded the resource disparity between itself and the subordinate lords, who could rely only on their allocated plots to support them. 85

PAGE 86

These Marxist models are not models of state formation but rather models of how the strong Tarascan political structure was characterized by a specific political economy and mode of production. Both the works of Beltrn and Carrasco acknowledge the role of conquest in the ability of the royal dynasty to appropriate land, especially the best land, but the relationships forged in conquest are not the focus of analysis to any great extent. This is a problem when attempting to study political relations in the Ptzcuaro Basin where state formation actually took place, because conquest was not the only or even the typical foundation of political relations within the Ptzcuaro Basin core (RM, 1956, pp. 148; see also Chapter 8). Both models posit Tarascan political control as the result of the ability to appropriate land from subordinate lords and then allocate some of that land to subordinate lords in order to ensure their cooperation and support, and in turn these subordinate lords enabled the perpetuation of the appropriation of land. Without any discussion of why the royal dynasty should have been able to unproblematically appropriate land in the first place, however, the models become exercises in circularity, without beginning or end. Because the states ability to appropriate land is taken for granted (by both the analyst and apparently the king and subordinate lords), or analytically displaced to a relation of conquest that is not problematized but rather assumed to have resulted in the states ability to appropriate the land, the relationship between the king and subordinate lords is simply the result, or a characteristic of, the political economic system as a whole. Political control must have been strong because political control is both the preexisting condition for and the result of the system as a whole. What these models do not address, in other words, is how the states ability to appropriate the land was produced in the first place and subsequently reproduced. There is no real explanation for why subordinate lords would have acquiesced, and continued to acquiesce, to the 86

PAGE 87

states appropriation of what could have otherwise been their land. Because the appropriation and reallocation of land is modeled as a self-reproducing circularity, problems of legitimacy are rendered epiphenomenal in theory and practice. The perception that the states appropriation of resources was appropriate and right (or at least permissible) either was assumed to not be an important analytical problem, or it was assumed that such a perception would have simply been produced by the state where and when it was needed by putting the resource advantage it enjoyed to work to construct legitimacy (e.g., in the state religion and state ritual, creating a super-added ideology that legitimized existing power relations). The latter assumption, that legitimacy simply would have been produced, once again presupposes a resource advantage that is assumed to have existed. Both assumptions, by relegating the production of a perception of legitimacy to epiphenomenal status, explain neither how the system got started in the first place nor what the state actually did in terms of how the king interacted with subordinate lords. In different theories of state formation and reproduction such as the model of Baines and Yoffee (1998, 2000), these interactions are precisely where legitimacy and the order of the state would have been produced. By order I mean the classification and organization of the roles and responsibilities of the members and institutions of society in relation to one another as well as the wider environment, the cosmos, and the supernatural (Baines and Yoffee, 1998, pp. 212-213; Yoffee, 2005, pp. 39-40). Of course it was the responsibility of a new institution (or set of institutions)to classify and organize the persons and institutions and the relations between them (Yoffee, 2005, pp. 100-109). By legitimacy I mean the institutionalization of peoples acceptance of, involvement in, and contribution towards order (Baines and Yoffee 2000:15). Order and legitimacy, in other words, defined of everyones place within the state and propagated the perception that those places were proper or at least acceptable. I return to the 87

PAGE 88

problem of legitimacy and order, and why they should be important aspects of the investigation of Tarascan state formation and reproduction, below. For now, however, I note that the relegation of problems of legitimacy and order to epiphenomenal status is problematic both practically and theoretically. Ecological and Settlement Approaches in the Ptzcuaro Basin Whereas the Marxist models of Beltrn (1986) and Carrasco (1986) are not specifically concerned with processes of state formation, Pollard (1982, 2008) has explicated a coherent model of the causes and pressures that led to state formation in the Ptzcuaro Basin and military expansion beyond it. To briefly summarize, rising lake levels in the Postclassic began to put pressure on the populations of the basin (Pollard and Gorenstein, 1980) while the topography of the basin necessitated that some groups in the basin offset their shrinking land through warfare (Pollard, 1982a, 2008). The settlement, economic, and administrative patterns and networks were structured to maximize administrative efficiency, and the capital of Tzintzuntzan emerged as the clear primate center of the basin and ultimately the region, undercutting the ability of the other settlements of the basin to challenge Tzintzuntzans supremacy (Gorenstein and Pollard, 1983). The need to feed the continually growing population of the basin favored a strong primate center and the administrative efficiency it managed to foster as the resultant efficiency enabled the state to expand militarily beyond the basin and ensure needed goods through a tributary system (Pollard, 1982b. Finally, the work of Gorenstein and Pollard (1983) examined the settlement pattern of the Ptzcuaro Basin as a whole. The settlement pattern of the Late Postclassic reveals a central concern for administrative efficiency, while the lack of other settlements to challenge Tzintzuntzans primacy is characteristic of the strongly centralized political structure. No other settlement would have had the economic, transportational, or religious importance to pose a 88

PAGE 89

threat to the capitals, and the kings, power. The settlement pattern thus enabled, and was a factor in (re)producing, the centralized nature of the Tarascan political structure. This model, if it can be called a single model because the summary outlined above is derived from a number of works, explains in detailed and logical terms why the Tarascan kingdom developed when and where it did. It also views the Tarascan kingdom in systemic and holistic terms. The belief that groups settled in specific areas of the Ptzcuaro Basin that would have been most affected by the rising lake levels views the populations as responding to the ecological system of the basin. The development of winners and losers in the basin as a whole was largely determined by the same ecological system. The ecological change of the Postclassic and its system-wide impact on the populations of the basin selected for increased militarism and differential access to, and control of, prime agricultural land, which led to productive advantages and a resource disparity between the Uacusecha and the other populations of the basin. Centralization and the state inevitably followed. Similarly, the population pressure in relation to the carrying capacity of the Ptzcuaro Basin favored the consolidation of power in the social system, but it is the system that adapted and transformed itself. In this model no account is offered of the social interactions and perceptions that were involved in such a transformation, although perhaps implicitly it is assumed that one way or another, subordinate lords would have perceived the benefits of more centralized power, potentially reaped greater compensation in the prestige rewards system (see below), and thus acquiesced to stronger kingly power. The model of state formation and centralization espoused by Pollard and colleagues does posit why the resource disparity between groups in the Ptzcuaro Basin should have arisen in the first place and thus avoids the circularities of the Marxist models enumerated above. In this model, the social system as a whole is essentially the actor. The social system behaves in 89

PAGE 90

response to primarily ecological factors, transforming itself and tending toward greater centralization and administrative efficiency. The Late Postclassic settlement pattern as a whole reflected these processes and was their result, confirming the capitals unrivaled power and by implication the kings unrivaled power that is portrayed in the ethnohistoric sources. Once militarism and resource disparities had produced the processes of state formation, moreover, the system essentially reproduced itself. The capitals primacy was self-evident, the royal dynastys resource advantage over subordinate lords was undeniable, and so the kings power went unchallenged, allowing him to direct further action that would increase the wealth in the states coffers and ensure the stability and expansion of the system. As noted above in relation to the Marxist models of Beltrn (1986) and Carrasco (1986), what is absent from the system-determinist approach at the heart of Pollard and colleagues model of Tarascan state formation and centralization is what the state actually does. What is left unanswered is, how did its parts (the factions within the state such as the royal dynasty, subordinate lords, and commoners) relate to one another? As Baines and Yoffee (1998, 2000) contend, such a consideration of how the different parts of states socially and materially interact with one another is necessary in order to fruitfully compare different cases of state formation and how and why they would have led to the many different kinds of archaic states found in the archaeological and historical records. This necessity has not been ignored by Pollard and colleagues, however, who have formulated a prestige reward model as a counterbalance to the deterministic approaches just enumerated. The prestige reward model, described further below, is meant to explain how the king and subordinate lords interacted with one another and why those interactions would have produced the kings unquestioned power. 90

PAGE 91

Political Power and the Prestige Reward Model The delineation between market and state economic flows by Gorenstein and Pollard (1983), combined with the study of mortuary patterns at Urichu by Pollard and Cahue (1999), and the brief outline in Pollard (2003b) outline a prestige reward system that kept subordinate lords beholden to the king. In this model the state established a system in which it could give valuable objects to the lords in return for their cooperation and loyalty through its (the states) ability to control access to valuable objects and insignia of rank that the lords would have desired or even required for their own rule. Conversely, the state could punish insubordinate lords by withholding such objects or even stripping lords of objects (particularly the insignia of rank) they had already been given. As Pollard (2003a, p. 51) notes, the states control of the objects that were important markers of status mirrored and signified the states control of access to office. In order to produce a system in which subordinate lords could be bought off with valuable objects and insignia of rank, the state needed to monopolize valuable objects so that the subordinate lords would have been dependent upon it for such objects. The state did this through a number of ways, including controlling all long-distance exchange to conquering metal rich areas surrounding the Ptzcuaro Basin, forcing metal to flow through the tribute system into state coffers, and organizing under its auspices craft specialists that could produce finely crafted objects and status markers from the raw materials obtained by the merchants and through the tribute system (Gorenstein and Pollard, 1983, pp. 98-111; Pollard, 1987, 2003a, 2003d; Pollard and Cahue, 1999). Particularly as explicated by Pollard and Cahue (1999), the state sought not only to monopolize and control the distribution of precious and foreign objects but also to redefine what was perceived as valuable. In the Late Postclassic period, it was also necessary that the local 91

PAGE 92

populations also perceived the possible benefits [of assimilation and participation in the state system], such as the prestige and the power gained by indentifying themselves with the Tarascan social structure. With the expansion of the Tarascan ideology to these zones [zones being ethnically assimilated], whatever access to resources and to prestigious social ranks began to lose legitimacy and relevance if it did not have a Tarascan origin (Pollard, 2003a, p. 57). Tzintzuntzan and the other Uacusecha capitals came to be the primary cosmological centers of the region, and this importance made objects from the capital or copies of such objects indicators of a relation with the new cosmological centers. This was particularly the case with respect to finely crafted ceramic vessels, as a distinctive style of forms and decorative motifs came to characterize Tarascan ceramics (Pollard and Cahue, 1999; Versluis, 1994). Tarascan metal objects also came to be defined by a unique style (Bray, 1984; Pollard, 1987). This model of the state prestige reward system is valuable and important, and provides a necessary backdrop for the investigation of how hierarchy was produced in the Tarascan kingdom. However, it cannot satisfactorily address issues of how the perception of the legitimacy of the king and acquiescence to the order of the state were produced and transformed. The subordinate lords are not modeled as perceiving the legitimacy of Uacusecha rule; they merely act in their own self-interest and choose from a limited number of options with consequences determined by the state. The inability of the model to address issues of legitimacy and the perception of order is problematic for a few reasons. The model of prestige reward does not explain its perpetuation in the face of the chaos and threats to the kings power wrought by the Spaniards in the first decade of colonial rule. If subordinate lords were merely bought off by the Uacusecha, as the model of prestige reward posits, this does not explain why these lords supported the king and continued to perform their 92

PAGE 93

duties for the king in the years leading up to the kings execution (e.g. attending court, marking off and protecting the kings fields, and even being willing to go to war for the king; see discussion in Chapter 1). In the earliest years of the colonial encounter, the king had peacefully acquiesced to the presence (if not necessarily the authority) of the Spaniards, watched as the idols and symbols of his religion had been destroyed and defaced, and been robbed of most if not all of his treasure. He was ultimately imprisoned, tortured, and finally killed by the Spaniards (RM, 1956; Warren, 1985). His body was burned and the ashes cast in a river by the Spaniards to prevent the use of some part of the kings person as a relic and rallying point of an insurgency by loyal followers (RM, 1956, pp. 275-276), a possibility that the Spaniards must have thought likely if they went to such lengths to prevent it. In short, the loyalty of the subordinate lords remained strong in a colonial context in which not only would the king have lost his ability to buy them off with wealth objects, but in which the Spaniards would have been in a stronger position to buy their loyalty with gifts. The kings weakness in relation to the Spaniards had been demonstrated at nearly every turnhe had been humiliated and tormented at their handsand yet still the subordinate lords did not revolt or abandon him. If the promise of material reward was the only thing that kept the subordinate lords subservient to the king, their loyalty in the early colonial period cannot be satisfactorily explained. Second, while the model of prestige reward as formulated by Pollard and colleagues posits that objects from the Uacusecha capitals (including the finely crafted ceramics) possessed a high value because of their cosmological importance, no explanation is offered as to why these objects should have been associated with that cosmological significance. More importantly, no explanation is offered as to how the capitals attained their cosmological significance in the first place. On a related point, Pollards (1991) study of the Tarascan religion concluded that it was 93

PAGE 94

still in the process of consolidation and organization as it had not completely incorporated the local cults into a more unified system of beliefs. While local beliefs and ideologies may have been resistant to state incursions, the view that Tarascan state religion was still in the process of being reworked is odd given the strong political and economic control that the Uacusecha possessed, even to the point that the king maintained power well into the colonial encounter. If the kings power was almost universally recognized and perpetuated even in the chaos of the colonial encounter, it would stand to reason that the belief system that located him at the center of the state and cosmos should have been similarly accepted. Additionally, the prestige reward system does not explain why the state should have needed to reorganize craft production and create uniquely Tarascan status markers such as ceramic vessels and metal objects. If the subordinate lords were merely being bought off and their loyalties were ensured through the control of valuable objects, the Uacusecha could have simply monopolized access to highly valued raw materials and finished goods crafted in foreign areas and then given these objects to the subordinate lords in return for their cooperation and loyalty. Pollard and Cahue (1999) discuss the use of a consistent and locally defined assemblage of prestige goods and status markers to foment a kind of class consciousness to offset preexisting ethnic loyalties and local mentalities, but this is a different issue than that of the lords acquiescing to kingly rule simply because they needed valuable objects that the king had, as the prestige model posits. By suggesting that there were other factors at work in processes of state formation than simply buying off subordinatesnamely, producing legitimacy, class consciousness, and the construction of a cosmological centerPollard and colleagues recognize the importance of these other factors in the process of state formation. All of these discussions are testament to the 94

PAGE 95

desire to consider how the royal dynasty interacted with lords and converted them to dependent subordinates that owed their loyalty to the royal dynasty because that was the natural order of things, how objects were instrumental to that project, how and why certain objects were suitable to that project, and how and why the objects suitable to that project were transformed by the state. In other words, the indications that the Tarascan state constructed ideological and political legitimacy, monopolized foreign trade goods, and attached craft specialists that could transform and produce valuable objects of a distinctive style mesh well with the model of the intertwining relations and functions of wealth, legitimacy, and order in archaic states formulated by Baines and Yoffee (1998, 2000; see also Yoffee, 2005, and Richards and Van Buren, 2000). Their model views the construction of legitimacy and order as a necessity in processes of state formation. By making the production of order and legitimacy a target of investigation, Baines and Yoffee (1998) hope to make the manner in which wealth was deployed in order to establish or transform symbols of order and to realign and transform identities essential aspects of a comparative framework for investigating the specificities of what states (and the elites at the heart of those states) actually do. On a final note, none of the existing models for the development or workings of the state utilize the rich detail of the ethnohistoric sources, particularly the representations of the RM. The Marxist and ecological approaches predominantly utilize the RM and other ethnohistoric sources in order to confirm specific practices with regard to the land tenure system or hypothesize that specific events and practices reveal the nature of the kings power and state apparatus as a whole. On the other hand, the formulations of a prestige reward system includes the representations of the exchange of insignia of rank but does not necessarily examine the rhetoric contained within the representations of such exchanges and why those representations 95

PAGE 96

should be contained within the RM. As I hope to demonstrate in the present study, such representations can aid in the investigation of how the kings legitimacy and the nature of the state were produced in social action. Legitimacy, Order, and Methodological Relationism The theories that have been used to study Tarascan state formation are ill-equipped to address factors such as legitimacy and the perception of order. The determinist theories (Beltrn, 1986; Carrasco, 1986; Gorenstein and Pollard, 1983; Pollard and Gorenstein, 1980; and Pollard, 1982b) include legitimacy as part of the system as already constituted. On the other hand the prestige reward system considers actors as unswayed by anything other than self-interest (see critiques of theories of self-motivated action in Ortner, 1984, 2001; Ritzer and Gindoff, 1994; also Johnson, 1999, p. 89; Meskell, 1996, p. 11). The perception of legitimacy and orderin other words, that some actions by certain actors are proper or should go uncontestedis an inherently intersubjective and relational phenomenon. It is intersubjective because it takes place between two subjects and within the subjective states of the subjects as they perceive and react to each other (Munn, 1986, pp. 14-16). Legitimacy and order are relational because they define the properness of actors and actions not only in relation to one another but also in relation to other actors and actions, both in the past and in the present. Order and legitimacy are the myriad ways of classifying, evaluating, and discerning who can or should do what, when, and why (Baines and Yoffee, 1998, 2000; Bourdieu, 1977, 1984; Sahlins, 1981, 1985a, 1990, 1991). By incorporating intersubjectivity and the relational perceptions of persons and actions, in other words, why the royal dynasty was able to substitute a new value system of objects for an old one, the questions of why exactly subordinate lords should have tied their fortunes to the kings in the colonial encounter, and why subordinate lords should have 96

PAGE 97

acquiesced to their very subordination and even perceived that subordination as the natural order of the cosmos can begin to be addressed. Furthermore, because a relational paradigm incorporates the material world as a constant force in the interactions between and construction of persons (Gell, 1998; Gosden, 1999, pp. 120-122), such a paradigm inherently involves the material world, a fact that makes such problems as legitimacy and order amenable to archaeological investigation. Therefore what is needed if legitimacy and order are important to studies of the Tarascan kingdom, as I have indicated I believe they are, is a paradigm shift in favor of a relationism that views social and material relations as integral aspects of the contextual construction of why certain actions and actors are valued and others are not. Conclusion Most, if not all, of the interpretations of the ethnohistoric and archaeological evidence suggest that the Tarascan king wielded a great deal of power and that his power was largely unquestioned within his realm. There are many good reasons for viewing the kings power as strong and the Tarascan kingdom as highly centralized. The king headed an expansive and loyal bureaucracy. The land tenure system not only filled the states coffers with a great deal of wealth but also at least theoretically attributed all ownership to the king, so that subordinate lords gained access to land through their connection with the state. The king also controlled a number of important other resources through the tribute system and his exclusive cadre of long-distance merchants. Finally, the settlement pattern in the Ptzcuaro Basin during the Late Postclassic was dominated by Tzintzuntzan and organized to facilitate administrative efficiency, thereby enabling the king to direct action in ways that would have benefited the state and enabled him to capture more resources. 97

PAGE 98

The models that have been put forth in order to explain the nature, origins, and reproduction of the kings power have, however, failed to address how exactly the kings legitimacy and the cultural order of the state that defined who should be able to do what was produced and reproduced. This inability to address the perceived legitimacy and order of the Tarascan kingdom is significant, particularly in relation to the events of the Spanish encroachment on the kings power in which the subordinate lords stood by the king and which the models of political control in the Tarascan kingdom cannot satisfactorily explain. I suggest that the inability of existing investigations of Tarascan state formation and reproduction to address the problems of legitimacy and order are due to their theoretical orientations. The deterministic approaches simply assume that legitimacy and order were produced by the system as it was constituted. The existing studies of the Tarascan kingdom do provide a necessary context by elucidating material flows and transformations, relations to the land, and organization of the bureaucracy. They have contributed much to the theorization and investigation of the nature of the Tarascan kingdom and the causes and processes that led to its development. At the same time they have problematically, at best, investigated legitimacy and order in the Tarascan kingdom, the pervasiveness and endurance of which make the Tarascan kingdom stand out among its Late Postclassic Mesoamerican peers. Therefore in contrast to these approaches, I have suggested that a relational methodology can address the problem of the production of order and legitimacy in the Tarascan kingdom. A relational methodology, viewing the perception and construction of actors as a continual phenomenon that incorporates relations with both other actors and the material world, can address why some actors should be perceived to be able to perform certain actions while others cannot. I discuss some of the basic foundations of a 98

PAGE 99

relational ontology in the Chapter 4, and more specifically I formulate a model of interactions between the Tarascan king and subordinate lords that can explain the relational valuation of some actors in contrast to others. 99

PAGE 100

CHAPTER 4 ENCOMPASSING OTHERS: CONSTRUCTING SAMENESS OVER DIFFERENCE THROUGH SOCIAL INTERACTION Introduction As I discuss in Chapter 3, theories of the Tarascan kingdom cannot incorporate how order and legitimacy were produced in the Tarascan kingdom due to their deterministic nature, in which macrophenomena determine microphenomena, including the nature of interactions between the king and subordinate lords. Therefore in order to investigate how order and legitimacy were produced in the Tarascan kingdom, I outline the foundations of a relational methodology in this chapter. Within a relational framework, I follow Sahlinss (1985a) general use of Dumonts concept of encompassment. Dumonts (1979, pp. 806-815; 1980, pp. 239-245) conceptualization of hierarchy as a system of valuation based largely on the encompassment of the contrary enables the investigation of the production and reproduction of an intersubjective and hierarchical order in which the actions of some were perceived to be legitimate while others were not. Louis Dumonts structuralist explication of hierarchical relations and their relation to the process of valuation has been influential in anthropology (Parkin 2003), and has wide ranging implications, as Dumont himself discussed (Dumont, 1979, 1986). While Dumonts definition of hierarchy as encompassment is useful, his structuralist theoretical and methodological foundation insists on a preexisting totality and fixed structural relations with that totality. As such it does not allow much theoretical space for transformation or history, as some of his critics have noted (e.g. McKinnon, 1991, pp. 34; Turner, 1984; Toren, 1999[1994]). It is therefore ill-suited as the sole basis for any theoretical model and methodology that is expected to explain social transformations such as the development of the Tarascan kingdom. 100

PAGE 101

Therefore certain modifications are necessary in order to inject process and history into Dumonts theoretical and methodological framework for investigating hierarchy. The critique and reformulation of Turner (1984) is especially helpful in this regard, as he introduces syntagmatic relationsthe relations by which terms are sequentially combined with one another in the production of meaninginto relations of encompassment. Turners postulate that encompassment involves the ability of the encompassing term to transform the encompassed term into a replication of itself necessarily implies agency (i.e. a causal and transformative capacity). Exactly how the replacement of the subordinate term and the related production of the whole occurs, however, is difficult to grasp in Turners specific case study (the hierarchical relation between moities in Amazonia), and I do not see an easily comprehendible manner in which to apply it to the case study at hand. Therefore I rely on the work of Gell (1998) and Keane (1997) in order to apply Turners insights in a model of encompassment through social interaction. These anthropologists both apply a semiotic approach to social action and meaning. They also investigate, in highly detailed ways, the role of material objects in the production of meaningful consequences, which makes their work useful and readily applicable to archaeological investigations. Gell in particular discusses how objects are abducted (in other words, inferred) to index (signify) one agent over another. I utilize these works in order to discuss how, through materially constituted interaction with others, certain agents can cause the actions of those others to be abducted to be the results of their own will and intentionality. In the end, the actions of others, and therefore the others themselves, are transformed into replicas of a now encompassing agent. 101

PAGE 102

In the end, in this chapter I develop a theoretical model of what I refer to as intersubjective encompassment, or the novel production of a relation of encompassment between two formerly autonomous or at least contrastive persons. All of the theories that I utilize place meaning at the center of their investigations, and in particular the works of Gell and Keane focus on how material objects are integral to meaningful interaction. Hierarchy, manifested as relations of encompassment, is produced only through meaningful interactions. By investigating how objects and the social interactions they facilitate are meaningful, this model of intersubjective encompassment contrasts with the materialist theories that have provided the dominant framework within which the development and reproduction of the Tarascan kingdom have been investigated and explained (see Chapter 3). Furthermore, the commensurate nature of the structuralism of Dumont and Turner and the semiotic approach to social action and the construction of persons of Gell and Keane, as they are all fundamentally theoretical investigations of the different ways in which objects and actions have meaning and consequences, enables their mutual application and complementary nature in relation to one another. It is not my intent to imply through my own application of these theorists and their insights, however, that there was some preexisting linearity or development among these works. What I present here as a rather linear discussion of how the work of one theorist can illuminate how the others might be applied in stepping-stone fashion is merely the result of my own somewhat idiosyncratic reading of their work. What I have deemed to be an important problem in Tarascan political theory, namely the manner and extent to which the king was able to construct subordinates as reflections of himself, has been the constant background to my own readings of the works I discuss below. Therefore while I believe these works are commensurate 102

PAGE 103

at a fundamental level and can be fruitfully used together, I do not mean to imply that the work of one theorist addresses or fixes, in all times and all places, what I perceive to be the shortcomings another. They merely can be brought together in an expedient fashion in order to help elucidate how a specific problem might be analyzed. The development of this model of intersubjective encompassment requires some technical jargon. The technical language of both the field of structuralism and semiotics is largely unavoidable, and I believe it is important to use that technical language to achieve precision in meaning but also so as to avoid muddying the waters with unnecessary additional terms. In an effort to offset the difficult nature of the jargon, however, I have provided examples, most of which are drawn from the scholars whose works I rely upon. At times it is necessary to modify or extend these examples, and I indicate where I have built upon those examples included in the major works cited. Relationality, Personhood, and Agency The foundation of relationality within anthropological theory goes back to Mausss (1954[1925]) classic study of gift exchange. Mausss basic insight was that objects are imbued with the personhood of their producers and exchangers and so persons can implicate themselves in the subjectivities (see below) of others. This line of analysis has been elaborated upon by numerous ethnographers, for example, Battaglia (1990, 1992), Gell (1998, 1999), Mosko (1985, 1992), Strathern (1988), Wagner (1977, 1991) and Weiner (1992). Within a relational ontology of personhood, the concept of personhood is intimately intertwined with the concepts of agency, objectification, and intersubjectivity. Who a person is, that is to say their socially recognized persona, is constituted through a web of social relations and the agencies of persons within those social relations. I use Gells (1998, p. 16) definition of social agency as a causal capacity that is attributable to some agents mind, will, and 103

PAGE 104

intentionality. A person is constituted through the agencies of others with whom he is engaged or related, and he (or some specific aspect of his persona) exists as a sign of those agencies that have gone into his constitution (Battaglia, 1990; Gell, 1999; McKinnon, 1991; Munn, 1986; Strathern, 1988). By the same token, a person is known by others through his own agency and the effects he has upon others or on the material world through engaging them in action (Gell, 1998, 1999; Munn, 1986; Strathern, 1988; Wagner, 1975, 1977). By producing an object or involving the self with some object, the self becomes a subject to both the object and to other subjects who might encounter the object and through it, the self (Miller, 1987, 2005). The agency of any particular person and the ontological construction of that person are indissolubly linked, and the agency of any one person is a result of, and signifier of, the agencies of other persons and objects that have played a role in the constitution of that agency. Therefore through social interaction, people and things form potentially infinite chains or webs of social relations that spread out across space and back in time, and specific actions and the objects they involve are both the means through which those relations are constituted as well as the outward signs of their existence (Gell, 1998, p. 62; Strathern, 1988). Social relations are formed through interactions between persons but are often realized through, and signified by, objects (Battaglia, 1990; Strathern, 1988; also Gosden, 1999, pp. 120-121). Social interactions, therefore, are inherently material. Because they are the results and material manifestation of a social act and/or capacity, objects are also signs (specifically indexes; see below) of that act and/or the capacity for the realization of that act. Objects, because of their materiality, typically endure through time and therefore can take on lives of their own, signifying and gathering together in a single object (Meskell, 2005, p. 4, using the terminology of the phenomenologist philosopher Martin Heidegger) the agencies and personhoods that have played 104

PAGE 105

a role in the production of their object biographies or life histories (Gaitn Amman, 2005; Kopytoff, 1986; J. Thomas, 1999; N. Thomas, 1991). Objects therefore signify and tie together the multiple social relations that have gone into their own production, and even into the production of those other objects and subjects that are directly linked through the object (Gell, 1998, p. 62, 1999; Miller, 1987, 2005; Strathern, 1988). In this sense objects are rather like persons (Gell, 1998, pp. 9, 12-23, 231). Both objects and persons, and the actions that they make manifest, are signs that exist in the real world of the agencies of others due to their very existence and histories (Gell, 1998, pp. 230-232, 1999; Strathern, 1988; Weiner, 1992). Because both material objects and human subjects, and the relations and social interactions that constitute them, are public and exist in the real world, they are intersubjective phenomena (Miller, 1987, 2005; Munn, 1986, pp. 14-16; Richardson, 1982). They exist between perceiving human subjects, and are open to confirmation and evaluation by other human subjects (Richardson, 1982). Who a person is, and what that person can accomplish or make manifest in the material world, are open to the interpretation of others and to a large extent constituted by those evaluations (Keane, 1997, p. 17). Intersubjective actions, such as speeches or gift exchanges, have consequences according to their nature but also to the extent that others perceive them to be efficacious or deserving of their attention and response. Relations are also intersubjective phenomena in another sense, insofar as the relations and social interactions that a human subject engages in will impact not only the way the participating subject is perceived by others, but will also impact the internal subjective state of the participant (Munn, 1986, p. 16). Thus a gift exchanged will cause the giver to be remembered or kept in mind by the recipientin other words, the giver exists in the mind of the recipient and this penetration of one subject by another will likely influence the future actions of the recipient 105

PAGE 106

(Munn, 1986; also Battaglia, 1992). Social interactions therefore have consequences by constituting and/or transforming the affective states (e.g. consent, refusal, happiness, and anger [Munn 1986, p. 16]) and commitments of the participants, in the end impacting the mind, will, and intentionality of the participants as they formulate their own course of action and desire that their actions have certain effects. However, this aspect of intersubjectivity is not solely ideational or mental, but is material as well. The objects possessed or transferred from one subject to another serve as the reminders of those subjects that have played a role in its production and transferal (Battaglia, 1992; Munn, 1986; Strathern, 1988). The objects act as both the material prompts that instigate memory and the transformation of affective states and as the objectifications of those persons who are the subjects of those affective states (Munn, 1986). By adopting a relational ontology of persons in which persons are not reified and bounded entities but are composed of all of the relations that have produced their social identities, it is possible to conceptualize how other persons are present in the actions of some agents. This is precisely the rhetoric that is recorded in the RM, and introduced in Chapter 1, in which the king claimed to have acted through subordinate lords. More precisely, it is possible to analyze how certain agents encompass others and constitute those others as aspects of their own personhood. As I detail in the next section this is the basis of a system of valuation in which the legitimacy of the king was produced through social interaction. Hierarchy and/as Encompassment Louis Dumonts definition of the hierarchical relation of encompassment was a significant contribution to anthropological theory, the ramifications of which are wide-ranging, as Dumont (1979, 1986) himself discussed. A student of Marcel Mauss, Dumont was a prominent French structuralist, although his insights into hierarchy and hierarchical opposition exist in marked contrast to Lvi-Strauss and others who made use of Lvi-Strausss concept of binary opposition 106

PAGE 107

(Parkin, 2003, p. 3). The Maussian tradition of interpreting societies as coherent wholes is well represented in Dumonts work (Parkin, 2003, p. 106), as Dumont posits that within a society certain values will be related hierarchically, and that the society as a whole should adhere to this fundamental value (Dumont, 1979, p. 814). In his specific discussion of hierarchy as encompassment, which is of greatest relevance to the present work, Dumont (1979, pp. 806-815; 1980, pp. 239-245) defines a structural relation of whole to part(s), in which the whole is superior in value to the part(s) that compose it. Essentially, this relation is of great use because it posits that difference and similarity can exist simultaneously, and moreover that difference can be subordinated to similarity. In other words, Dumont offers a basic terminology of how different actors may be perceived as the same, and why exactly this relation should be hierarchical. Dumont has been critiqued by many scholars to degrees ranging from vehement opposition to his entire anthropological program (e.g., Bailey 1991; Berreman 1971; Beteille 1987; Lynch 1977) to disputes with his discussion of hierarchy in India (e.g., Appadurai 1988; Das and Uberoi 1971) to acceptance of some points and refinement of others (e.g., Sahlins 1985a; Toren 1999[1994]; Turner 1984). The first two are not necessarily relevant to the present study, as I believe the usefulness of Dumonts general definition of hierarchy and encompassment are demonstrated in what follows. On the other hand the critiques and refinements, insofar as they demonstrate some of the weaknesses of Dumonts analysis of hierarchy, are essential to adapting his basic insights to the real world of actions and how actions can produce relations of encompassment. In particular, I rely on Turners (1984) elaboration and insistence on syntagmatic relations between the parts within relations of encompassment because it provides, in my mind, the shortest route to a model of the realization of encompassment through social processes. 107

PAGE 108

Dumonts Original Formulation In his classic study of the caste system in India, Dumont (1980) formulated a conception of the differential valuation of entities and which he explicitly contrasts with hierarchy as a rank-order system or a hierarchy of graduated authority in which incremental increases in authority define higher levels against lower levels (Dumont, 1980, p. 65). Dumont (1980, p. 19) contends that power is not the deepest root of the logical basis of a hierarchical system. Rather, values are the basis of hierarchy. In other words, man does not only think, he acts. He has not only ideas, but values. To adopt a value is to introduce hierarchy, and a certain consensus of values, a certain hierarchy of ideas, things and people, is indispensable to social life. This is quite independent of natural inequalities or the distribution of power. No doubt, in the majority of cases, hierarchy will be identified in some way with power, but there is no necessity for this (Dumont, 1980, p. 20) More specifically, at the heart of this process of valuation (at least in Dumonts formulation) is the relation of parts to wholes. So we shall define hierarchy as the principle by which the elements of a whole are ranked in relation to the whole (Dumont, 1980, p. 66, emphasis in original). Wholes are composed of myriad parts and it is because they represent the totality of the parts that wholes assume a greater value. Dumont (1979, pp. 809-812; 1980, pp. 239-245) contends that the basis of all hierarchical relations is the encompassment of the contrary wherein one term is subsumedencompassedwithin the other. The encompassing term is more highly valued because it represents the totality, while the encompassed term represents only a part of that totality. Dumont perhaps explains this relation best, or at least in a familiar example that is more easily apprehended than others, in his example of the story of the creation of Adam and Eve from Biblical Genesis. God creates Adam first, the undifferentiated man, the prototype of mankind. In a second stage, he extracts a different being from this first Adam. In this strange operation, on the 108

PAGE 109

one hand, Adam has changed identity; from being undifferentiated, he has become male. On the other hand, a being has appeared who is both a member of the human species and different from the main representative of the species. In his entirety, Adamor man in our languageis two things in one: the representative of the species mankind and the prototype of the male individuals of this species. This hierarchical relation is, very generally, that between a whole (or a set) and an element of this whole (or set) (Dumont, 1980, pp. 239-240). It is important to accentuate the two meanings that Adam (man) comes to possess. Adam is, at the outset, the representative and prototype of all mankind, as he is the only human. Through the extraction of his rib and the creation of Eve (woman) from that rib, Adam is now also defined as gendered man, in opposition to gendered woman. Through the continuity of the name and the physical process of extraction in which Adam remains and is the origins of Eve, however, a link between the latter meaning, gendered man (man as a part or subset of mankind), and man as a whole (as mankind), is preserved. The fact that Eve is composed of a part of Adam is what preserves this relationship and defines Adam as both a whole and a part. There are thus two contrasts between man and woman. These two contrasts exist on what Dumont calls different levels. On one level is a relation in which man represents all of mankind, and woman is only a subset of that category of mankind. It is essential to note here that woman is also man in the sense that she is also a (hu)man (as implied by Dumont [1979, p. 809] in the context of another example). This will be quite important below, in the discussion of replication. On the other level man is different from woman as they are differentially gendered subsets of mankind. Thus the relation on one level posits similarity, and this is the relation of encompassment, as man stands for all of mankind. The relation on the other level posits difference, as mankind is divided into diametrically opposed categories (Dumont 1979, p. 809). It is unfortunately confusing that the term levels is applied by Dumont and those who follow his writings on encompassment in another, different, arena of analysis when discussing 109

PAGE 110

contextual reversals of the hierarchical relation between the terms (Dumont 1979, pp. 811-812; also de Coppet, 1992; Parkin, 2003, pp. 59-63; Tcherkzoff, 1987, p. 67). For example, a hierarchical opposition, say male over female, that exists at one level (in public) might be reversed at a less important level, say as females are dominant within the house. The different levels at which the hierarchical relations are reversed are, again, different from the separate levels of contrast and identity within the hierarchical relation proper. This concept of reversal as an indication of a transition in levels of the hierarchical system of valuation is not of great importance here, as I am concerned primarily with the relation between the relations of opposition and hierarchical encompassment proper (i.e., the relation of part-to-whole). Because Dumont sees the relations between the levels of similarity and difference (i.e., in the first use of levels above) as existing both in perpetuity and simultaneity (Dumont, 1980, pp. 242-243), he repeatedly interprets their interrelation but apparent mutual incompatibility as a logical scandal that appears when one intermingles the levels (Dumont, 1979, p. 809; 1980, p. 242). When interpreting the relation between the relations, however, Dumont clearly gives primacy to the relation of encompassment (i.e., the relation in which Adam represents mankind), what he calls hierarchy in the strict sense (Dumont 1979, p. 812; this position is critiqued by Turner, 1984, discussed below). The two levels are themselves ranked in relation to one another; the level on which the dominant term (e.g., Adam) encompasses the subordinate term (e.g., Eve) is itself superior to the level on which the terms are opposed to one another (e.g., Adam and Eve as gendered subsets of mankind) (Dumont, 1980, p. 242). Critique and Reformulation As noted above, Dumonts discussion of valuation, hierarchy, and encompassment has been criticized from many angles. Outright rejections of Dumontian sociology (see above) notwithstanding, the insight that relations of similarity and difference can coexist and be ranked 110

PAGE 111

in relation to one another holds great explanatory potential for the study of the Tarascan kingdom. In its essential elements it at least theoretically allows for a (different) social other to be encompassed and made part of ones own self (thereby constructing similarity), which accords well with the kind of rhetoric contained within the RM in which the actions of subordinates are attributed to the king (see Chapter 1). While Dumonts insights therefore hold promise, the critiques and refinements made by others provide a more systematic analysis of some of the Tarascan data while not necessarily adopting the entire program of Dumont. These critiques are made in reference and opposition to some of Dumonts own positions and discussions of how his model of encompassment relates to the dialectic of the German philosopher Georg Hegel (with which Dumont himself contrasted his own formulation of hierarchy; Dumont, 1980, pp. 242-243). Hegelian dialectic is a model of self-production and self-realization in which the self is confronted with an Other (the negation) but manages to overcome the Other by incorporating the knowledge gained from the confrontation with this Other within the self (the negation of the negation)(Gosden, 1994, pp. 65-66; Miller, 1987, 2005). For Hegel this process inherently takes place in historical time, i.e. as a process (Gosden, 1994, pp. 62-68). Dumont asserts that his structure of encompassment is diachronic, in that differentiations are permitted (1980, pp. 242-243). Differentiation, however, posits that transitions in levels (in the second sense discussed above) are inherent in the process of differentiation, i.e. that the differentiation exists at a lower level. This remains, however, a structural relation that in its totality does not change. Dumont (1980, p. 242) readily admitted that in essence, his hierarchical opposition produces nothing. By this he meant that nothing novel is created. Ultimately, Dumont (1980, p. 243) boils the contrast he draws between himself and Hegels dialectic down to a simple dichotomy; for Dumont the whole preexists while for 111

PAGE 112

Hegel a totality without precedent can be produced synthetically, manifested as negation and the negation of the negation. The reason why any application of Dumonts conceptualization of hierarchy to social process, and specifically to the historical production of relations of encompassment in the Tarascan kingdom, must in the end resemble Hegels dialectic deserves some attention. If relations of encompassment were the foundation of the Tarascan kingdom, there must have been some point before the development of the Tarascan kingdom at which relations of encompassment did not exist. In short, the Tarascan king, as an encompassing person, must have been produced, and must not have existed prior to any social interaction, as Dumonts conceptualization of hierarchy as encompassment insists. Therefore what is needed is a more Hegelian model of the historical production, and reproduction, of a relation of encompassment that, if produced and reproduced, includes the basic elements of encompassment (i.e. the part-whole relationship and superior valuation of the whole) as explicated by Dumont (following Torens [1999] critique of Dumont and preference for Hegels theoretical insistence on historical process and the fact that meaning is historically constituted). The work of Terence Turner, an ethnographer working in Amazonia, is essential in this regard. Primarily in the 1970s and 1980s, Turner dedicated himself to reexamining the theoretical foundations of (primarily Lvi-Straussian) structuralism (Turner, 1977, 1985). Those works (Turner, 1977, 1985) in which Turner discusses the necessity of syntax in any properly structuralist methodology are important to understanding Turners reformulation of Dumonts concept of hierarchy and encompassment, as I discuss below. It is also important to note that Turners analysis of hierarchy within moietal organization in Amazonia focuses on specific social processes (which is to say syntagmatic relations). In order to avoid getting bogged down 112

PAGE 113

in the details of Amazonian ethnography upon which Turner basis his analysis, I apply his model to the example of Adam and Eve discussed by Dumont. Within Turners (1984) critique of Dumonts concept of encompassment, Turner accepts the basic premise of Dumonts discussion of hierarchy as fundamentally a relation of whole-to-part and the encompassment of the contrary. However, he criticizes Dumont for postulat[ing] an immediate identity between the totality and the dominant term [e.g. Adam] of the polarity, and thus reduc[ing] the asymmetry in question to a direct manifestation of the single dimension of encompassment (Turner, 1984, pp. 336-337, emphasis added), i.e., the relation of part to whole. Turner (1984, pp. 338-339, 366-369) believes that the immediacy of the identification of the dominant term (e.g., Adam in the example above) with the whole need not pertain, and moreover, that the identification of Adam with that whole must be produced. Therefore encompassment and the identification of the dominant term with the whole must be produced through the transformation of the relation between the opposed terms, because the whole does not yet exist (Turner, 1984, pp. 338-339, 366-369). In his endeavor to explain this aspect of production, Turner introduces syntax to the problem of relating the hierarchically ranked terms. The syntagmatic or syntactic aspect of [linguistic] structure pertains to the combination of elements [as a sequence, i.e. through time] drawn from different paradigmatic classes into various levels of linguistic units (Turner, 1977, p. 121). The production, and interpretation, of meaning requires syntagmatic relations and their investigation (Turner, 1977, p. 108-124). Change is meaningful only in the context of some kind of contiguity, and this insight of structural linguistics necessarily includes contextual juxtaposition, i.e., syntagmatic relations. For example, if a man were to say that he works his own fields on Mondays and then go on to say that he works his lords fields on Tuesdays, the 113

PAGE 114

preservation of the context (i.e., both utterances involve working fields and days of the week) in the syntagmatic combination of the two utterances is what makes the different fields and days that he works meaningful. If he were to say he works his fields on Monday and subsequently that his pants are blue, the second statement appears as a non sequitor (it does not involve either days of the week nor working), and no meaning can be gleaned from the juxtaposition of the two utterances. With the introduction of syntagmatic relations, Turner can more systematically analyze how encompassment is produced as a relation between parts that is in turn transformed into the relation between part and whole that is familiar from Dumonts discussion. I explain Turners model of encompassment as a process in terms of Dumonts own example of the relation of encompassment between Adam and Eve, introduced above. In some regards this is an inappropriate application, because the example of Adam and Eve posits the preexistence of a whole (Adam), which also makes it a perfect illustration of Dumonts very insistence upon the preexisting nature of the whole. At the same time, it is this insistence that is the target of Turners critique and makes the strict application of Dumonts model itself inappropriate to the active production of intersubjective encompassment, as discussed above. With a certain modification of the Adam and Eve myth below and with the above caveat in place, however, I believe I can explain Turners insights while remaining faithful to his insistence on process. Key to the production of a relation of encompassment is the transformation of the relation between the terms in order to produce a new relation that exists alongside the first (Turner, 1984, pp. 364-369). The first relation involves the opposed terms, e.g., Adam and Eve. The second relation is a new relation produced by the replication of the dominant term (Adam). In this new relation the dominant term is preserved but its replica (what I call Adam, or Adam Prime) 114

PAGE 115

takes the place of the subordinate term (Turner, 1984, p. 367). Because the subordinate term (Eve) has been replaced by the replica of the dominant term (Adam), the subordinate term has now dropped out of the equation completely, at least in this relation. Eve is, in other words, not a differently gendered human in this relation, but a replica of Adamshe is a human being just like he is, and it is this identity that is emphasized in this relation between Adam and Adam. The continuity of substance (Eve is made from Adams rib) establishes Eve as the replica of Adam, as she is composed of the same substance as Adam. Finally, the replacement of Eve by Adam entails that this new relation, that between the dominant term and its replica, constitutes the dominant term as the whole (now constituted as Adam and Adam, which both are members of the class of things called Adam) because it is the only term (Turner, 1984, p. 369). In any relation between a prototype and its replica, there is a certain tension between the sameness and difference of the prototype and its replica (Turner, 1985, pp. 66, 82, 91-93). Because the replica is a replica, it is the same kind of thing as its prototypethey are both members of the same class or type. However, even in the case of perfect identity between the prototype and its replica, the two remain different in at least one respect, namely, that they are two separate instances of that same type. They are each discrete objects, which can exist apart from one another in time and space. A temporal relation between the two also inheres in their relation; the prototype by definition exists prior to the replica. In the example of Adam and Eve, the relation between sameness and difference is more complex. Clearly Eve is different from Adam in one respect. She is a female, as opposed to Adam the male, and if she was not female the point of her very creation is nullified. Eve is also like Adam, however, as she is a human. Again, if she were not also human the very point of her creation is nullifiedit would do Adam (and mankind more generally) little good if a female of another species had been created. The 115

PAGE 116

relation between a prototype and a replica does not necessarily posit perfect identity, but the contextually defined perception of similarity. The hierarchical relation of encompassment pertains because the one relation, that between the prototype and replica (Adam and Adam), encompasses the relation of opposition (the relation between Adam and Eve), just as in Dumonts formulation of the encompassment of the relation of difference by the relation of similarity. However, for Turner the encompassment of one relation by the other is not a given as it is in Dumonts discussion of the encompassment of one term by another. Rather, in Turners discussion of the relation between the two relations, the relation of replication is encompassing precisely because it represents the syntagmatic axis of the structure, upon which successive replications of the relationship defined upon the latter [the relation of opposition between parts] are integrated into a common structure (Turner, 1984, p. 365). Put more simply, the dominant term encompasses the subordinate term precisely because the former is the means of its own replication; it is this process of replication (in time and space, i.e., in syntagmatic relations) that achieves the replacement of the subordinate term by the replica of the dominant term and thereby produces the dominant term as the whole (the entire field, now defined as Adam and Adam, as discussed above). Turners discussion of replication and encompassment also necessarily implicates social action in what Munn (1986, pp. 9-11) calls intersubjective spacetime. Intersubjective spacetime is the field of space and time, woven together, that is continually produced (and therefore always potentially transformed) by social beings as they interact with one another across space and through time. In other words, social beings map out social actions and relations in spacetime and they calculate how they themselves might live in the memories, sentiments, and ultimately the actions of others within that field of 116

PAGE 117

spacetime (ultimately aiming to transform, i.e. expand, that field of spacetime in which they are situated). Turners reformulation is quite complex, with a number of relations and moving parts. I believe that the introduction of syntax (requiring action in space in time, and also implicating the material world) and the ability of the dominant term to be the means of its own replication and concomitant transformation into the encompassing term can be explained through the myth of Adam and Eve once again. A certain modification to the myth is needed, however, to illustrate Turners insight that the dominant term should be the means of its own replication and therefore the means by which the dominant term is produced as the whole. Imagine if Adam himself, rather than God, was the one who had removed his own rib and used it to create Eve. Would this have been the case, Adam himself would be the means of his own replication. He would, in technical terms, be the paradigmatic instantiation of the syntagmatic process of replicationof his own transformation from gendered part (subset) of humanity to the whole of humanity (following Turner [1985, pp. 52-53]). Adam would exist not merely as a category (e.g. as mankind or gendered man), but as an active principle of transformation (from the one meaning to the other), or the embodiment of transformation itself. The advantages to, and specificity of, Turners focus on the syntagmatic production of encompassment in comparison with Dumonts explanation of encompassment deserve to be explicitly stated and summarized. Turners reformulation of relations of encompassment introduces syntagmatic relations, i.e. process, whereas Dumonts explanation is inherently static. Dumonts explanation also requires, or is dependent upon, a preexisting whole (i.e., that the dominant term should already exist as the whole). This is perfectly illustrated in the Biblical example of Adam and Eve, in which Adam is, at the beginning of the myth, the only member of 117

PAGE 118

mankind and therefore is the whole of mankind. Turners reformulation, on the other hand, allows for two separate terms, neither of which preexist the other. The relation of encompassment, which is dependent upon the dominant terms (Adams) ability to constitute the whole, is produced through the dominant terms ability to produce a replica of itself. This replica then takes the place of the subordinate term (Eve) in one relation. Therefore the essential features of Turners model of encompassment that are necessary for its application to social action and the production of encompassment are that, first and foremost, Turners model is precisely about the production of encompassment and not a static relation of encompassment, and second that this relation of encompassment is produced out of an original context of two opposed terms. I have not explained, however, exactly why or how the subordinate term (Eve) is replaced by the replica of the dominant term (Adam), precisely because the myth is ill-suited to such an explication (because Eve does not exist at the beginning of the myth to be replaced), and Turner himself demonstrates this replacement with reference to the moietal structure of Amazonian societies. This aspect of how, specifically, the dominant (encompassing) term replicates itself and furthermore replaces the subordinate term with its replica does require further elaboration in order to apply it to encompassment through social interaction. I believe that the discussions of Gell (1998) and Keane (1997), with some modifications, are quite suitable for this task. Both of their works focus on social action and how certain social actors can represent other social actors. The implications of this aspect of representation allow for the elaboration of a model of encompassment that is capable of analyzing how exactly the dominant term replaces the subordinate term and constitutes the whole. 118

PAGE 119

Agency, Abduction, and Replication: Encompassment Through Social Action Turners formulation of syntagmatic relations and replication as inherent in any process of encompassment is quite useful as a general guide to conceiving how encompassment of one subject by another might be realized. In Alfred Gells (1998) magnum opus on art and the agentic relations inherent in artistic production, he explicates many ideas that are relevant to the current problem of encompassment. I do not follow Gells discussions and definitions precisely, because in certain respects I believe that Gell contradicts himself, or at least that certain of his ideas are not followed through in the remainder of this otherwise tremendously rigorous work. These departures become clear as I proceed through a consideration of Gells more important ideas. First and foremost among Gells ideas integral to the present work is his characterization of social agency as inherently relational and context dependent (Gell, 1998, pp. 21-22). Who or what possesses agency is only identifiable in relation to another subject that exists as a patient of the agents actions. Who or what exists in the agent position or the patient position is also subject to change depending on context or point of view. Second is Gells distinction between primary and secondary agents (Gell, 1998, pp. 17-21, 35-38, 44). In Gells definition, primary agents are human subjects that possess intentionality, while secondary agents are the objects (lacking intentionality of their own) through which primary agents realize some action, i.e., their agency is made manifest. Third is Gells (1998, pp. 13-16) discussion of the fact that agency is abducted (inferred by some witness) from an index, which I explain below, but the index is generally an object. This aspect of Gells thinking in particular is central to analyzing agency and the ability of one subject to encompass another. The fact that agency is always abducted, furthermore, leads me to depart from Gells own definitions of primary and secondary agents and to reconceptualize the inherent relationality of agency as defined by Gell. 119

PAGE 120

In relation to Gells discussion of abduction, the work of Keane (1997; also Jones, 2007, pp. 12-19, 24, 38), who discusses the risks of representation and the reception of intended meaning in any purposive social act, is helpful and leads to important insights concerning the nature of encompassment realized through social interaction. Keane employs a Peircean semiotic approach to social action and the material prerequisites of social action in order to analyze how risks of representation (particularly of the ancestors) are instigated by ritual so that they may be overcome, thus revealing the power of both the ritual and the agents that have pulled it off. In the end, I posit that intersubjective encompassment is realized as the transformation of the opposed subject (the social other) into a secondary agent via the abductions made by a community of witnesses in which the encompassed agent is robbed of intentionality and primary causality of his own, thereby transforming him into a secondary agent by definition. If the encompassing (primary) agent can cause a community of witnesses to abduct his exclusive or primary intentionality and causation, the action achieved by the encompassed (secondary) agent is transformed into the replica of the act that the encompassing agent has already willed and set in motion through his interaction with the secondary agent. Those familiar with Gells concept of distributed personhood might find the delayed introduction of this concept conspicuous, given that it has garnered some attention among archaeologists (e.g. Fowler, 2001; Gosden, 2001; Harrison, 2003; Looper, 2003; Robb, 2004). A model of intersubjective encompassment could rather expediently be produced with distributed personhood as the core concept. However, in order to theoretically model, and to subsequently evaluate archaeologically, exactly why some objects should be abducted (inferred, see below) to be the distributed personhood of a particular person in the first place, the same discussion that I embark upon below is unavoidable. Therefore I begin with the foundations of Gells discussion 120

PAGE 121

of agency, the role of objects in the manifestation of social agency, and the integral nature of the process of abduction from indexes and their relation to agentic actions. Gells Definition of Agency and the Role of the Index Among competing definitions of agency, which abound (e.g. Dobres and Robb, 2000) Gells definition of social agency is the most relevant to the present work, because I am most concerned with the intersubjective calculation of which agent (among potentially many) is more responsible for the realization of any particular action. Gell (1998, p. 16) defines social agency as attributable to those persons who/which are seen as initiating causal sequences of a particular type, that is, events caused by acts of mind or will or intention, rather than the mere concatenation of physical events. He goes on to briefly discuss the imposition that agency is concerned with intentionality, in relation to the contention that philosophers (unnamed by Gell) and sociologists (also unnamed by Gell) recognize that the unintended consequences of actions have important consequences. I do not dispute, following Giddens (1984, pp. 9-14), that actions do have unintended consequences and these unintended consequences might be important or even analytically essential to the investigation of the reproduction of structures. But my intention is not to analyze the reproduction of a general structure, along with everything that is subsumed within that structure, but rather to analyze the reproduction of a cultural order in which agency is regularly attributed to one agent over another, or in which one agent counts more than others (following Sahlins, 1985a, 1985b, 1991). In such calculations, intentionality and mind are essential, as I discuss below. As stated above, Gells definition of social agency is inherently relational and context dependent (Gell, 1998, p. 22). This relationality is expressed in his explanation that agents always exist in relation to patients (Gell, 1998, pp. 21-22). For an agent to be an agent, it must have an effect upon some patient. Thus an agent cannot exist without the existence of a patient. 121

PAGE 122

Furthermore, who or what is an agent or a patient is dependent upon context and even perspective within a given context, and is not an immutable characteristic or property of the things or persons involved. What is a patient in one relation is potentially an agent in relation to some other patient. In his theory of art objects, agents manifest their agency in relation to patients through what Gell calls the index. The index is some object that is the result of some agency and that allows some other agency to be realized or made manifest, in the process creating another patient (Gell, 1998, p. 15). The pivot of the art nexus is the index. The index is the disturbance in this causal milieu which reveals, and potentiates, agency exercised and patient-hood suffered on either side of it (Gell, 1998, p. 37). The index is itself an agent (a secondary agent, as I explain below), as it achieves some effect in relation to a patient. The production of a new patient via the index may even be spatially or temporally displaced from the agent who has produced the index in the first place (as in Gells [1998, pp. 20-21] example of land mines). Therefore in Gells conceptualization of agency, material objects (as indexes) are essential elements of agentic actionsthey are the means through which one social agent affects other subjects (patients). Moreover, the nature and capacities that these objects constitute (and which are imputed into the object in the first place by some agent) help determine the nature of the agent and the specific ways in which he will be able to affect other subjects (patients) (Gell, 1998, pp. 20-21). Primary Agents, Secondary Agents, and Causative Chains The role of the index as an agent leads Gell (1998, pp. 17-21, 35-38, 44) to distinguish between primary agents and secondary agents. Primary agents are entities endowed with the capacity to initiate actions/events through will or intention while secondary agents are entities not endowed with will or intention by themselves but essential to the formation, appearance, or 122

PAGE 123

manifestation of intentional actions (Gell, 1998, p. 36, see also p. 44). Therefore secondary agents are often the means by which some act, which is willed into realization by a primary agent and set in motion through some action, is more proximally realized. The primary agent has both willed and initiated a sequence of actions or events that has involved the production of the secondary agent. The production of the index, as the secondary agent, allows for the realization of the ultimate intended goal or aim of the initiating act through its own secondary agency. Primary agents are always human subjects in Gells distinction between primary and secondary agents. According to Gell (1998, pp. 21, 44), only human subjects can possess mind, will, or intentionality. Objects, as indexes, cannot have intentionality and do not possess a mind and a will. Therefore objects as indexes are relegated to the role of secondary agents. The division between primary and secondary agents, based on the possession of mind, will, or intention (or lack thereof), has drawn discussion and critique by archaeologists as they have attempted to integrate Gells specific ideas into the general view that material objects are active and impact social processes (Gosden 2001, pp. 164-165; Robb, 2004, pp. 132-133). These critiques (Gosden, 2001, pp. 164-165; Robb, 2004, pp. 132-133), however, focus on Gells relegation of objects to the status of secondary agents, or as Gosden (2001, p. 164) puts it, of deficient persons. My critique of Gell runs in the opposite direction from that of Gosden and Robb. I suggest that the distinction between primary agents that are always humans and secondary agents that are always objects should be modified so as to allow for the possibility that human subjects might be transformed into secondary agents. This would mean that a human subject as secondary agent is denied intentionality, will, and a mind in specific contexts and events. While this might be offensive on a universal or philosophical level, it is defensible with regard to specific actions and 123

PAGE 124

the differential role that human agents play in the realization of those specific actions. The possibility that human subjects might be rendered secondary agents hinges on the abduction, made by a community of witnesses, as to which human agent(s) involved in the realization of an act have intended that act to be realized and which agent(s) played the primary causative role in its realization. Abduction from Indexes and Human Secondary Agents Gell uses the term index from semiotics as developed by the American philosopher Charles Sanders Peirce. As defined by Peirce, indexes are natural signs insofar as they exist in spatial or temporal contiguity with the signified (Gell, 1998, p. 15; Keane, 1997, p. 19; Jones, 2007, pp. 12-19; Lele, 2006; Preucel, 2006, pp. 56-65; Preucel and Bauer, 2001, p. 89; Sebeok, 2001, pp. 10-11, 84-96). Very often the temporal contiguity between signifier and signified is a relation of causation. The interpretation of smoke as a sign of fire is an example of an index; here the smoke is interpreted to be a sign of the existence of a fire, because a fire would have caused the smoke that is perceived (Gell, 1998, pp. 13-14). Gell uses the term abduction to describe this process of inference in which art objects as indexes signify an agent. Abduction is not concerned with interpreting the meaning of an index according to semiotic conventions (i.e., symbolic relations) nor (universal) laws of nature but lies somewhere in between (Gell, 1998, p. 15). Abduction is neither induction nor deduction (Gell, 1998, p. 13). For example, Gell (1998, p. 13) points out that it is common knowledge that smoke can exist in the absence of fire, and so it is not a universal law that smoke always means fire. In the process of abduction, however, an inference is made that more likely than not smoke means fire, or perhaps because the existence of a fire somewhere has potentially very serious consequences, the perception of smoke leads the perceiver to jump to the conclusion 124

PAGE 125

(abduct) that a fire exists (or might exist) and to act accordingly (e.g., confirm the existence of a fire, call the fire department, or run away). From an analytical perspective, the meaning of an index is inherently bi-directional (Gell, 1998, pp. 35-38; Jones, 2007, pp. 12-19; Keane, 1997, pp. 15, 229-232). An agent causes an index to be made or brought into existence, and also gives the index its form and potentially the contextual clues as to the meaning of the index but that exist outside of the specific index (as I discuss further in Chapter 5) from which other agents will make inferences about the meaning of the index. The actual process of abduction, however, is unavoidably carried out by social others (the patient or a community of witnesses; see below)and this process of abduction is dependent upon the ability of those others to interpret the clues that compose the index and its context. The relation between agency and intentionality imputed into an index and the meaning (specifically the agency) abducted from the index is not so simple, however. Causality is attributed to some agent not only according to the presence and qualities of the index, but in coordination with past experience (Gell, 1998, pp. 13-16). Running away from smoke, for example, requires the knowledge that smoke generally or most consequentially means fire but also that fire is dangerous. The meaning of the index is also determined by its relation to other (indexical) signs, and the singular process of abducting agency from an index renders the whole set of signs, the assemblage (to use a term familiar to archaeologists; following Webmoor, 2007, p. 571) in which the index is related, as the index from which agency is abducted (Gell, 1998, pp. 14-16; Jones, 2007, pp. 12-19; Keane, 1997, pp. 12-15). I explain the integral role of other contextually relevant indexes with an elaboration of the example of smoke and fire below. The fact that agency is abducted from the index and the context in which it is deployed in social action has a singularly important implication that has garnered little attention from 125

PAGE 126

archaeologists who have used and adapted Gells discussion of the agency of objects. This is the implication that, because agency is abducted from indexes, who or what is an agent is subject to the abductions made by some other witnessing subject(s). Therefore agency, and the question of who has agency, is necessarily a triadic relation, involving agent, patient, and the abductor who determines which is which. As discussed above, that abductor will make his abduction with respect to the nature of the interaction between the agent and the patient, particularly as that interaction involves the index and transformations of the index. In certain situations the abductor might be the patient who abducts the index to be the result of the agents agency and therefore suffers patient-hood (Gell, 1998, p. 37), as in many of Gells discussions of art objects. This need not be the case, howeverthe abductor might be situated outside of the agent-patient relationship and the specific abduction he makes as to who is the agent will have an impact on the kinds of actions that those agents and patients can undertake in the future. In the present study in which I focus on the relationship of encompassment between the Tarascan king and subordinate lords, in many contexts the abduction of who is the (encompassing) primary agent and who is the (encompassed) secondary agent is made by a community of witnesses, including the commoners (these commoners are often the target of rhetorical speeches, as I analyze in Chapters 6 and 7) and perhaps other subordinate lords. Therefore I refer to those subjects who are engaged in abducting who is the primary agent and who is the secondary agent as a community of witnesses, composed predominantly of commoners but likely also including other nobles. Of course, the relationships between the agent, patient, and abductors can become quite muddled. This is because by witnessing the index and abducting the primacy of one agent, the witness is acting either as an agent by actively conferring primary or secondary status on any 126

PAGE 127

particular agent or as a patient because this abduction is prompted by the index. This potentially infinite multiplication of agent-patient relationships distracts from the goal of this discussion, however, which is to formulate a model of intersubjective encompassment between the king and subordinate lords that is achieved via the abductions made by a community of witnesses. Instead, I focus on the implications of the existence of a community of witnesses existing outside of any particular agentic act of social interaction, namely the attempt by one agent to achieve some effect through another agent. This community of witnesses, making abductions based on the nature and context of the index, plays a crucial role in deciding which social actor is an agent or a patient. Not only are abductions made as to who or what is an agent in any particular interaction, but also who or what possesses intentionality, mind, and a will (which is for Gell largely the same thing). On this point, again, I depart from Gells delineation between primary and secondary agents, because who or what might possess intentionality is a matter open to interpretation and not the subject of a universal rule. The central issue is not so much whether or not agents are inherently primary or secondary agents, but rather whether or not a community of witnesses abduct that certain agents are primary agents and others are secondary agents. That is, an abduction is made concerning whose intentions, mind, and will played the primary role in the realization of any act and who is acting as the mere means through which that intended act of the primary agent is realized. Therefore I am not positing the negation of the universal fact that all human subjects possess intentionality, will, and mind (particularly in Chapters 8 and 9 I discuss how secondary agents can and do exhibit all of these). I merely submit that in the abductions made by a community of witnesses, some agents will be abducted to play a role in the realization 127

PAGE 128

of specific actions akin to Gells description of primary agents while others will be abducted to play a role akin to Gells description of secondary agents. To illustrate the implications of the centrality of abduction in equations of agency, the issues of agentic causality and context can be incorporated in the example of smoke signifying (indexing) fire, introduced above. As Gell (1998, p. 15) himself explains, if a witness abducts the smoke to be the result of someone burning swiddens in anticipation of planting season, this does imply (social) agency and not the mere concatenation of physical events. Furthermore, I suggest that the matter of where and when the smoke arises as an index of a fire at a specific spatiotemporal locale, in concert with the abductors (presupposed) knowledge of other potential agents whereabouts at that particular moment, will affect the way in which the agency behind the fire behind the smoke is abducted to be the agency of one person and not another. Additionally, if numerous plumes of smoke are rising simultaneously from the same general area, a witness might make the abduction that a coordinated effort to prepare a large plot of land for planting season is taking place, and so the (more socially relevant) agency behind the simultaneously rising plumes might be that of an agent who has coordinated the effort to cut and burn the swiddens but is not necessarily cutting and burning swiddens himself. The abduction that one agent should be a primary agent while another (human subject) should be a secondary agent hinges on the abduction of the effect and/or meaning of the specific act completed by the secondary agent to be that effect intended by the primary agent. In short, the effects caused by the secondary agent (whether an object or subject) must conform to what a witnessing and evaluating audience (following Keane, 1997, p. 17) believes the primary agents intentions to have been in the first place. This will be the case so long as a community of witnesses perceives that a primary agent could have expected that the chain of events would 128

PAGE 129

follow a path that he foresaw, desired, and willed through an act of his own that had set a chain of events in motion (Gell, 1998, pp. 16-17, 21). At every step in the chain of events, the object and the actions that flow from it confront a community of witnesses and lead them to make abductions about the predictability of that sequence and whether or not the actual events have followed the path that the primary agent is abducted to have intended. In complex chains of social interactions involving potentially multiple human subjects (one of which might be primary agents and others as secondary agents), the interactions that lead up to the realization of some specific act will necessarily take place over time. Therefore the witnesses who abduct the primacy of a primary agent and the status of secondary agents as such will make that abduction with reference to the status of the index at any one time but also in relation to their background knowledge involving the past actions in which the index has been engaged (or their abductions made as to the nature of those past actions). In such abductions, it is important to realize that in some ways the qualities and present situation of the index itself might index the indexs involvement in specific social actions, as in Gells (1998, p. 231) example of the patinated surface of a kula shell that indexes the shells long history of exchange. Gell (1998, p. 62) himself describes how the index is the visible knot which ties together an invisible skein of relations, fanning out into social space and social time. This skein is manifested by the indexthe index prompts witnesses to calculate its composition by tracing out those social relations across space and down through time that have brought the index to its present context in its present state. The ancillary abductions as to the composition of that skein of relations unavoidably involve the context of the index, i.e., its relation to other indexical signs, and knowledge of the composition of social relations apart from the index (and in relation to 129

PAGE 130

other indexes), such as the histories of those objects or classes of objects given to subordinate lords in acts of investiture (see Chapters 3 and 7). To illustrate these points I use Gells example of an ambassador representing his country and more specifically his countrys government or head of state Gell (1998, p. 98). Clearly the head of state intends that the ambassador is a secondary agent, i.e., that the ambassador represents him and will achieve some action that he desires, namely that the ambassador will relay his message to some foreign entity. From the other perspective, the audience (whether a foreign head of state or the populace of that country, should the message be made public) abducts that the ambassador is speaking on behalf of that head of state. Such is the case when the ambassador speaks concerning issues of state and more specifically takes positions that are otherwise known to be (from past experience), or abducted to be, the positions of his government. In this example the government of the ambassador has shaped what the ambassadors message will be and the audience of the ambassador perceives the consonance between their expectations and the actual message and therefore abducts the words, and agency behind those words and the fact that the ambassador has spoken those words to them, to be that of the government that the ambassador represents. In those instances in which the index (in this case the ambassadors message) is abducted to be attributable to the primary agents will and intentionality, even as it is situated in the hands of others (or emanates from the mouths of others) within complex chains of social interaction, it becomes what Gell (1998, pp. 21, 96-154) calls the distributed personhood of the primary agent. The index is, in other words, the manifestation of the primary agents will and intentionality. 130

PAGE 131

Encompassment Through the Abduction of Replication In any act of representation, i.e., of acting as a secondary agent on behalf of a primary agent as in the example of the ambassador, the achieved effect is abducted to be a replica. The act realized by the secondary agent is abducted to be the replica of the act that is abducted to have been intended and willed by a primary agent and subsequently put in motion through some prior action by the primary agent (following Keane, 1997, p. 12). Therefore the act that is realized by the secondary agent, an act that would ostensibly or in other circumstances be evidence (an index) of his own agency and personhood, is transformed in the process of abduction by a community of witnesses into an index of the agency of the primary agent. Because the secondary agent becomes the means through which the primary agent achieves his effects, moreover, the secondary agent also becomes the distributed personhood of the primary agent. This transformation takes place through the object that, as an index, has motivated the abduction by a community of witnesses that it is the distributed personhood of the primary agent. The index, abducted to be the distributed personhood of the primary agent, constitutes the capacity of the secondary agent to execute the will that has been abducted to be the will of the primary agent. The index stands not only as an index of the agency of the primary agent in general, but specifically as an index of the will and capacity to endow the secondary agent with the material objects required for the realization of some action. That action is precisely the action that the primary agent is abducted to have willed and set in motion through the production or transferal of the index to the secondary agent in the first place. The will and intentionality behind the act that the secondary agent makes manifest is, in this process of abduction, attributed to the primary agent. The secondary agent becomes (like) an object in relation to the primary agent, an objectification and index of the primary agents capacity to will and intend that some act be realized, and finally an index of the primary agents 131

PAGE 132

capacity to constitute the secondary agent with the ability to execute that will through some relation with the object-as-index. All of this takes place through the primary agents capacity to ensure that his own role in any course of events is abducted to be primary and thus monopolize the possession of will and intentionality in the abductions of a community of witnesses. The secondary agent is himself subsumed within the primary agents personhood because he exists as an index of the primary agents agency and personhood. The secondary agent, in this process of abduction, is replaced by the primary agent, whose agency and will is abducted to be responsible for the action that the secondary agent has achieved. This is exactly what Turner describes in his model of encompassment (see above). One point within this process of abduction bears emphasizing, particularly in its relation to Turners explication of encompassment as a process. If the abduction is made that an agent is secondary in relation to a primary agent, this entails that the primary agent is abducted to be the agent by which the secondary agent comes into being. In other words the primary agent is abducted to have endowed the secondary agent with the ability and capacity to achieve the action he intends (or has intended, or is abducted to have intended) to make manifest (Gell, 1998, pp. 36, 44). Therefore the primary agent exists as the paradigmatic instantiation of a syntagmatic process, or simply put as the means by which the secondary agent replicates (achieves) the action the primary agent has already intended and in the process becomes a replica of the primary agent himself. As discussed above, this is one of the necessary conditions detailed by Turner (1984, pp. 364-367) for a relation of encompassment to be realized. Therefore encompassment can be achieved through the abduction by a community of witnesses that a primary agent has willed and set in motion a chain of events that has produced some action, and that this chain of events is abducted to involve the production or constitution of a secondary agent by that primary agent. 132

PAGE 133

In order to more fully investigate the manner in which attempts to transform social others into secondary agents, Webb Keanes concept of slippage is useful. In his ethnography of ritual action in Indonesia, Keane (1997) analyzes how the production and interpretation of signs are an essential, or even the essential, characteristic of social action (in his case study, ritual action). Keane (1997, pp. 3, 230) points out in his discussion of representation that there is always a potential risk for slippage between the intended meaning and the received meaning (see also Sahlins, 1985a), and the concomitant risk that the intended effect will not be realized. This risk is due to the paired and inescapable facts that representations are things in themselves (Keane, 1997, pp. 8, 230) and that there is some onus on the perceiver to abduct the representation to be an index of the entity being represented (Keane, 1997, pp. 14-15). I suggest, moreover, that the potential for the derailment of the intended effect of the primary agent by a secondary agent is greater in the instances in which the secondary agent is a human subject as opposed to an object. This should be so because human subjects do possess mind and intentionality in a universal or philosophical sense, and so the secondary agent can actively transform the message or index that the primary agent is relying on to induce the abduction of his (as a primary agents) agency, intentionality, and mind by a community of witnesses. The very act of constituting the ability of a human subject as a secondary agent to achieve the desired effects of the primary agent simultaneously imbues that secondary agent with an agency and recognizes an inherent agency within the secondary agent such that the secondary agent is up to the task (following Keane, 1997, p. 231). However, the ability of the secondary agent to alter the index or the message which is intended to achieve some effect, i.e., the risk that some slippage may occur, remains subject to the abductions made by a community of witnesses, 133

PAGE 134

who may still abduct the primacy of the primary agent and hence the secondariness of the secondary agent. Suppose, to continue the example from above, that an ambassador says something patently absurd or alternatively something that differs in certain respects from the message that the ambassadors head of state intended the ambassador to relay to his audience. In the former situation the audience would likely abduct the primary agency to be that of the ambassador, believing there was no possible way that the head-of-state should have intended the ambassador to say such ludicrous things. The latter situation, however, is not so simple. The members of the audience will have to abduct to whom the words spoken by the ambassador rightly belong and who has willed them into existence (following Keane, 1997, pp. 229-232). The community of witnesses will make the abduction with reference to their knowledge of the past positions of the head of state that the ambassador represents in coordination with the material signs that the ambassador represents that head of state, such as official seals, documents, gifts, and even the very fact that the ambassador has appeared in the flesh before them, having apparently been sent by that head of state. Or, to make the same point with reference to the Tarascan kingdom, abductions will be made with reference to the material signs of a relationship between the king and the subordinate lord, such as the insignia of office (see Chapter 7). This raises the possibility that slippage might occur between the received and intended meaning while a relation of encompassment (a relation of primary and secondary agents) might still be achieved within the abductions of a community of witnesses. In such situations it is necessary to differentiate between multiple potential intended effects that the primary agent has willed and set in motion; a primary agent may intend that many things be achieved by the secondary agent or within his relation to the secondary agent. One potential goal, of course, 134

PAGE 135

would be the reception of the message or the realization of the act that the primary agent has intended. Alternately, however, the primary agent might simply and more fundamentally intend that the secondary agent be abducted to be a secondary agent, thus constituting his own role as the primary agent. The abduction that a secondary agent is a secondary agent will involve those material objects as indexes that the primary agent has endowed the secondary agent with as indexes of the fact that the primary agent intends the secondary agent to act on his behalf. The abduction that the secondary agent is a secondary agent will also be made with reference to the actual act or message made manifest by the secondary agent, and more specifically the consonance between the act that the primary agent is abducted to have intended and the perception of the act that is actually brought to fruition by the secondary agent. Finally, it might be abducted that an (primary) agent would likely not send an index (secondary agent) out into the world to achieve some effect unless it had a reasonable chance of achieving that effect and reasonable assurances that the effect (meaning) would be as intended (following Keane, 1997, pp. 8, 12, 14). Returning to the example of the ambassador yet again, if the audience abducts that the secondary agent does indeed represent the primary agent, the actions of the former are abducted to be the intent of the latter. This is to say that even if the actions of the secondary agent do not strictly conform to the actions intended by the primary agent, if that action is abducted to be the faithful execution of the primary agents intent, the act of the ambassador remains, in effect, a replication. It is a replication of the act that the community of witnesses abducts that the head of state has already intended and willed to fruition. The act essentially manifests its own novel prototype, i.e., the message of the primary agent that is abducted to have been the intent of the primary agent. 135

PAGE 136

To the extent that the ultimate intent and meaning of the primary agent was most fundamentally to constitute the secondary agent precisely as a secondary agent, there is the potential that specific meanings may be lost or altered but the fundamental intent of the primary agent to encompass the secondary agent has nonetheless been achieved. Some slippages might be tolerable so long as the fundamental goal, which was to have the secondary agent be abducted as such in the first place, is achieved. This may sound quite unimportant, or even counterintuitive as far as political negotiations go, but its achievement results in a major transformation of human subjects and their agentic capacities. Most fundamentally, the achievement of encompassment entails a superior valuation of the encompassing agent in relation to encompassed agents (following Dumont, as discussed above). This valuation can be brought to bear on other and future relations with the encompassed agents, or even on novel relations with agents that have heretofore been unengaged in social action. Encompassment and/as Tarascan Political Theory I propose that this model of synthetic intersubjective encompassment, in which formerly autonomous and antithetical human subjects are transformed into an extension of the self, can explain how political power in the Tarascan kingdom was both conceptualized and historically produced. Through the encompassment of subordinate lords, the king as primary agent achieves his own expansion in intersubjective spacetime (following Munn, 1986, p. 9-11; see above); he is, in other words, in multiple places and multiple times at once because he is abducted to be the primary constituting force behind the actions of subordinate lords in the past, present, and future. As he lives in the minds (abductions) of commoners and other subordinates, and through their abductions is present in the bodies and actions of his subordinates, the king is everywhere at once. Viewed from the opposite perspective, the kings expansion takes place at the expense of the secondary agents, the subordinate lords, who in a certain respect disappear from the 136

PAGE 137

intersubjective world altogether. Or at least they live in the intersubjective world only as indexes of the kings superior agency and not as autonomous agents in their own right. To bring this discussion back to its own starting point with Dumonts concern with valuation, the effect of intersubjective encompassment is to produce the superior valuation of the primary agent, in other words the legitimacy of the kings place in society. That is, the specific goal of the king in engaging in projects of encompassment is not necessarily (or not only) to achieve effects at a distance and have his exact intentions replicated by subordinates (although this is certainly a perquisite of encompassment), but is the realization of the encompassment of subordinate lords. It is the realization of the encompassing relation that produces his own legitimacy (i.e. his own superior value, in Dumontian terms) and at the same time casts the actions of subordinates as legitimate only if they are abducted to be the faithful replication of the kings wishes. Conclusion In this chapter I have delineated a model in which the encompassment of other actors, realized as the synthetic production of a relation in which a primary agent (following Gell, 1998, pp. 17-21, 35-38, 44) transforms some other into a secondary agent that exists as an extension and replication of his own mind, will, and agency. This preserves Dumonts basic insight that relations of similarity and difference are themselves related hierarchically, and thus allows for the analysis of how subordinate lords might be extensions of the Tarascan kings agency and person. However, in order to develop a processual model of encompassment, which Dumont explicitly argues against, it has been necessary to incorporate Turners insight that syntagmatic replication is a necessary feature of relations of encompassment, in combination with Gells and Keanes discussion of the abduction (recognition) of replication. Gells discussion of the abduction of the existence of primary agents that intend some act to occur and 137

PAGE 138

then set in motion events that end in the acts realization through secondary agents has been recruited to complement and fill in the gaps, so to speak, of Turners explanation that the dominant (encompassing) term should replicate itself and that this replica should replace the subordinate term. It allows, furthermore, for the examination of exactly how primary and secondary agents (as replicas) are produced. For the abduction to be made that primary and secondary agents exist in the realization of any act, the primary agent must be abducted to have been the motivating force behind the secondary agents production of the act. In other words, the primary agent must have willed that act, set it in motion through some relation with the secondary agent (i.e., materially endowing that secondary agent with the capacity to achieve that act), and then relied on the secondary agent to carry out his will as a replication of the act he has already intended. As Keane (1997, pp. 3, 14-16) points out, however, there are always risks that inhere in this process of representation. The risk that the act realized by the secondary agent is not in fact the act intended by the primary agent is potentially nullified by the community of witnesses and the abductions that these witnesses make. If they abduct that the act of the secondary agent is indeed the act intended by the primary agent, then the novel act manifests a novel prototype of the intended action of the primary agent in the minds of those witnesses. The end result is the preservation of the relation between primary and secondary agent, and hence the relation of encompassment between the two. Finally, I have purposefully avoided the question of how exactly a would-be encompassing/primary agent ensures the abduction by a community of witnesses that his intentions, will, and actions have led to the manifestation of some effect by a secondary agent. This is because this discussion is intimately related to the ability of archaeologists to identify and 138

PAGE 139

analyze relations of encompassment in the archaeological record. In order to ensure the abduction of their own primacy via the index, certain actions involving the index but also its contextual associations must be realized. These actions in turn are frequently, if not inherently, materially constitutedthey involve the use, transformation, or exchange of objects. It is these actions, some of which involve the index directly but others of which involve the index indirectly, that the community of witnesses uses as the basis of its abductions. In Chapter 5 I suggest that by analyzing archaeological evidence and ethnohistoric representations, modern investigators can approximate that same level of knowledge possessed by a community of abducting witnesses and insert themselves into that community of witnesses. 139

PAGE 140

CHAPTER 5 METHODS FOR THE ETHNOHISTORIC AND ARCHAEOLOGICAL ANALYSIS OF ENCOMPASSMENT Introduction In the preceding chapter I outline how the practice of encompassment could have been constituted through interactions with other social actors. To summarize briefly, an encompassing person must be abducted by other social beings to be the primary, or even sole, agent through which an encompassed person is endowed with the capacity to carry out the intentions of the encompassing being. In this way the actions of the encompassed being are abducted to be the replications of the intentions of the encompassing being, and thus the encompassed being is abducted to be merely an aspect of the encompassing beings personhood, and a means through which that encompassing being can achieve social effects across time and space. Endowing others with the capacity to act is a process that often, if not always, involves certain material objects and their transferal. This outline of the realization of encompassment in practice possesses certain key features that allow for the delimitation of discursive and archaeological correlates. The first key point is that agency, and thus relations of encompassment, is abducted by witnesses. Thus interpreting whether or not a relation of encompassment existed, for both witnesses in the past and archaeologists in the present, is a process of abduction. The second key point is that practice in general is fundamentally material; this is true of the capacity of any agent to act as well as the process of endowing others with the capacity to act. Therefore agency and relations of encompassment are abducted from material objects. These generalities in turn lead to specific logical conclusions that allow for the investigation of relations of encompassment. First of all, abduction, just as the interpretation of meaning more generally, is not ahistorical but is culturally constituted through previous practices and discourses surrounding 140

PAGE 141

those practices. Put more simply, witnesses will have different ways of seeing and interpreting objects and actions. The fact that the process of abduction is historically and culturally constituted requires that power relations play a key role in that constitution. Certain agents will be able to mold a semiotic ideology (Keane, 2005), or culturally constituted way of seeing, by their actions in the past and present, and actions in the present concerning the past (i.e., cosmological authentication [Weiner, 1992] which I discuss below). Because certain actors can attempt to discursively frame (Keane, 1997; see below) objects and the proper historical and contemporary meanings of those objects, these efforts to frame objects involved in social interaction must be examined. The practice of discursively framing objects and social relations can be analyzed through an investigation of the representations in the ethnohistoric record of those discursive practices that attempted to frame persons, exchanges, and object biographies. Additionally, the actual realization of framing practices likely involved other material objects, which can be recovered archaeologically. The second logical conclusion is that the abduction of the meaning of an object, and its role in the constitution of a relation of encompassment, takes place with reference to a material world that is perceived by the abducting witnesses themselves. This material world consists of the objects themselves (Gells indexes; see Chapter 4) and the witnesses knowledge of, and/or yet further abductions concerning, other prior human-material interactions that have affected the existence and current status of the object. From the point of view of the abducting witnesses, then, this process of abduction becomes a process of reading the agentic investments (following Gell, 1998, p. 68; see below) that have produced the objects existence in which it confronts these witnesses and instigates their abductions. It is, in effect, a process of reading the biography (Gaitn Amman, 2005; Kopytoff, 1986; see below) of the object in question and 141

PAGE 142

abducting the identity of the primary agent responsible for that biography or some particular aspect of that biography. The biography itself will consist of potentially multiple events of acquisition, production, and exchange, each of which require investments from social actors that can then be read and abducted by witnesses, and which can be investigated archaeologically. The task then becomes to put the two general arenas of action, the discursive framing and the material investments throughout an objects biography, together in a manner in which they would lead to witnesses to abduct the primacy of certain agents over others in the constitution of relations of encompassment. This task involves the investigation of how the framing and material investments would have either reinforced or contradicted one another. Because any actions, and the objects that facilitate them, might be the targets of agents that wish to encompass others through them, I first discuss how the analytical focus can be narrowed to those objects that constituted meaningful capacities in Tarascan society. Not all actions or agencies were equally meaningful, and the Relacin de Michoacn (RM) indicates what actions and objects were important to the authority of local lords. I then outline the concept of object biography as a tool to integrates the others concepts (abduction, investments, framing, cosmological authentication) through which encompassment can be analyzed. Identifying Meaningful Actions and Objects The ethnohistoric record is useful because it contains representations of what practices were actually important to the exercise of power by local level lords. These actions should have been targets of efforts by the king to ensure the abduction of his own agency from the objects that their realization required. The representations of the duties and practices of the subordinate lords in the RM helps to narrow the analytical focus to a more limited range of material culture through which the king would have attempted to encompass subordinate lords. This encompassment would have taken place through the construction of indexical relationships that 142

PAGE 143

linked the king with those objects that were necessary to the execution of the subordinate lords duties. The RM suggests that three objects were central to the execution of the duties of the office of subordinate lords. The first was not an object per se, but a wife. As I detail below, wives were important to subordinate lords because their labor produced cloth and food that were integral to lordly demonstrations of generosity and piety. They were also important because of the status and wealth objects that they possessed. The second object essential to subordinate lords efforts to construct their authority were insignia of office. All of the insignia were important to subordinate lords but the bezote, or lip plug, appears to have been the most significant object, as I discuss below. The final set of objects that was necessary to the subordinate lords relationships with the commoners in their charge was ceramic vessels used for feasting. The RM indicates that feasting, or providing food and drink to commoners and noble visitors alike, was an important duty of subordinate lords. Roles of Wives: Providing Food and Cloth to the Gods and the People Prominent among the responsibilities of the nobles was to be generous. Generosity was demonstrated primarily by distributing goods to other nobles and the commoners. Overseeing the care of the god Curicaueri was also an obligation that the nobles were burdened with. These responsibilities required the labor of a woman or many women. In the prologue of the RM the friar who compiled and wrote that document claimed that the only virtue of the people of Michoacn was their generosity, for they held it to be a great offense to be stingy (RM, 1956, p. 4, my translation). This emphasis on generosity is repeated numerous times throughout the rest of the RM, both in the legendary history of the Uacusecha royal dynasty and in the more ethnographic third part that describes life in the Tarascan kingdom. Primary among those items given by the lords was cloth garments. Lords were 143

PAGE 144

expected to give cloth items to anyone who came calling, whether the visitor were some representative of the king or a commoner (RM, 1956, p. 185). Lords were also expected to provide food to those who entered his house (RM, 1956, p. 206). The RM indicates that the labor of women was vital in the production of both cloth and food. In the description of the marriage between a seor and a female relative of the king, the priest officiating the marriage ceremony specifically instructed the woman to make mantas (reams of cotton cloth) and food for the lord (RM, 1956, p. 208). Furthermore, female labor was essential for the care of the god Curicaueri. In the chapter of the RM concerning the selection of new caciques, the priest instructed those present to place women in the new lords house, so that they may make mantas and food for Curicaueri (RM, 1956, p. 206). In the legendary history of the Uacusecha, the women from Naranjan and Coringuaro who marry Uacusecha men are advised by their kinsmen to make food and coverings for the god Curicaueri and for their new husbands (RM, 1956, pp. 16, 65). In addition to the importance of their labor, female relatives of the king that married subordinate lords would have been important sources of legitimacy for the subordinate lords. This legitimacy was embodied by the valuable adornments that these women possessed. The description in the RM (1956, p. 208) of the marriage ceremony of a female relative of the king and a subordinate lord states that the bride was dressed in all of her finery, including jewelry objects made of precious metals and precious stones. As I discuss further in Chapters 7 and 8, these wealth objects, with their historically constituted association to positions of authority, would have been essential aspects in the efforts of subordinate lords efforts to demonstrate their own legitimate authority. This would have been the case especially if the lords themselves 144

PAGE 145

lacked access to the same kinds of valuables, which appears to have been the case (see Chapter 8). Objects of Bodily Adornment and the Agencies They Facilitate The RM also indicates that the possession and display of wealth objects as forms of bodily adornment was an important and defining characteristic of noble status (RM, 1956, pp. 33). The possession and display of such valuables was likely a necessary requirement for members of the nobility to be recognized as nobles and therefore these adornments were essential to their legitimacy. The transfer of such objects from the king to a subordinate lord was, moreover, an act that constituted the investiture of a subordinate lord and his elevation to the office of cacique (RM, 1956, p. 203). Therefore objects of adornment were constitutive of a lords perceived legitimacy and authority. Particularly emblematic of the act of investiture was the transfer of the bezote (Pollard, 2003b, p. 51). The Tarascan word for bezote is angamecua (Gilberti, 1962, p. 15). Pollard (2003b, pp. 50-51) notes that angamencha was a term that delimited people with bezotes, and possibly was a title or name of an office. The bezote therefore seems to have been the sine qua non of legitimate authority and first among other forms of adornment in the constitution of particular persons as possessors of legitimate authority. The indexical relationships established by the bezote, due to its properties and location on the body, are also noteworthy and helps elucidate why exactly bezotes should have been so important to conceptions of authority in Tarascan society. The proximity of the bezote to the mouth naturally linked the object and its properties to the mouth and its functions, the most relevant of which was speaking. Bezotes in Tarascan society were made of ground and polished obsidian and often included precious stones such as turquoise and/or precious metals. In accordance with Seegers (1975, p. 212) point that [t]he ornamentation of an organ may be related to the symbolic meaning of the organ in a 145

PAGE 146

society, an examination of the role of speaking and speech in the Tarascan political structure will prove useful to understanding the meaning of bezotes. As I discuss more fully in chapter 6, public speaking and exhortation were important aspects of the roles of the local nobility. More generally, multiple forms of bodily adornment were important to the identity of the nobles. Aside from bezotes, the RM (1956, p. 203) states that subordinate lords were given earspools and bracelets at their investiture ceremony. These objects are known from elite Tarascan burials as well, for example at Tzintzuntzan (Cabrera Castro, 1977-78; Castro Leal, 1986) and at Huandacareo (Macas Goytia, 1990, 1996). As discussed above, noblewomen were also elaborately adorned, and elaborately adorned females have been recovered archaeologically at numerous sites in addition to the Uacusecha capitals (Macas Goytia, 1990, 1996; Pereira, n.d.; Pollard and Cahue, 1999). The association of elaborate forms of adornment made from precious metals, precious stones, and shell with positions of authority and leadership has a long tradition in Michoacn (Hosler, 1994, 1995; Pereira, n.d.; Pia Chan and O, 1982; Pollard and Cahue, 1999; Oliveros, 1975) and Mesoamerica more generally (e.g., Brumfiel, 2000; Joyce, 2000). Bodily adornment in general has important intersubjective consequences, which the example of bezotes suggests. As discussed by Strathern (1979), Turner (1980), and Gell (1993) a persons outward adornment becomes conflated, in the mind of witnesses and the possessor alike, with that persons innate powers or nature. The abduction and conferral of these powers can amplify the persons ability to act, and to act meaningfully in ways that others will respond to. Therefore objects that adorn the body in meaningful ways should have been important targets of the kings efforts to encompass subordinates by ensuring the abduction of his own agency from those very objects used in adornment. 146

PAGE 147

Importance of Feasting in Lordly Authority Communal religious festivals, at which the lords would have given food and drink to other nobles as well as commoners, were also important to the identity of the Tarascan nobility (Pollard, 1991; RM, 1956, pp. 9, 103-106, 175). At such festivals, nobles reaffirmed their association with the king and paid reverence to the gods, providers of life and prosperity. Certain kinds of ceramic vessels that were suitable for the dispensation of drink and the distribution of food would have been integral to the ability of lords to successfully produce the kinds of feasts and communal religious ceremonies that their office required. Certain vessel forms suitable to dispensing drink and distributing food are known from Tarascan sites and include vessels with spouts, basket handles, miniature vessels, and bowls (figure 4-1; Castro Leal, 1986; Gal, 1946; Macas Goytia, 1990; Pollard, 1993; Pollard and Cahue, 1999; Rubn de la Borbolla, 1941). In particular, vessels with spouts and handles as well as miniature jars likely would have been used to serve pulque (maguey wine), which is mentioned many times in the RM (1956, e.g., pp. 70, 77). Vessels with spouts and handles are, moreover, depicted in plates 9, 37, and 44 of the RM (1956). The identification of these forms as related to feasting is buttressed by the study of Smith (2008; also Smith, Wharton, and Olsen, 2003), which identifies bowls, cups, pitchers, and miniature vessels as vessel forms involved in feasting at Aztec-dominated Late Postclassic period sites in what is now the Mexican state of Morelos. It is therefore possible that if the king wished to encompass the ability of the local lords to feast commoners and be abducted as the means through which that feast was realized, the king could have provisioned subordinate lords with the vessels necessary for those feasts. In order to ensure the abduction of his own role in provisioning these objects, however, the king should have been perceived to be the primary or even exclusive means through which the lords obtained vessels suitable for feasting. If subordinate lords could obtain such vessels on their own account 147

PAGE 148

this would have created significant ambiguity and likely would not have led commoners to abduct the primary role of the king in the production of the feast. A B C D E F Figure 5-1. Examples of Tariacuri phase ceramic forms suited to feasting. A. Miniature jar. B. Line drawing with profile of the vessel in A. C. Zoomorphic vessel with spout. D. Olla with spout and vertically oriented strap handle. E. Tripod bowl. F. Olla with spout and vertically oriented strap handle. All vessels were recovered at Urichu in Dr. Pollards archaeological excavations. Vessels represented in A, B, D, E, and F also have resist (negative) decoration. All photos and drawings courtesy of Dr. Helen Pollard. 148

PAGE 149

The cache of vessels excavated by Pollard (n.d.a) in Area 1 at Urichu is a striking example of the importance of feasting, and feasting vessels, to subordinate lords in general and within the Tarascan core in particular. A total of 18 vessels, broken but reconstructed, were found grouped together and associated with the noble residence excavated in that area of the site. It is possible that these vessels were stored by the noble residents of the structure and kept on hand for feasting occasions (including religious ceremonies) when they would be used. All of these capacities, from providing for the gods and commoners to wearing the proper attire, would have required specific objects for their realization (or in the case of women required for food and cloth production, other subjects). So far, then, I have identified what capacities would have been essential to the construction of an identity as a lord. What remains is a consideration of the potentially multiple significatory relations of those objects as they interacted with social subjects. The multiple significatory relations were constituted over the life history, or biography (Gaitn Amman, 2005; Kopytoff, 1986; also Gosden and Marshall, 1999), of the objects in question. In the next section I discuss this concept and how agency can be invested (Gell, 1998, p. 68; see below) in objects in order to assure the abduction of that agency from those objects. Object Biographies, and Objects as Records of Their Own Subject/Object Interactions In order to ensure that a community of witnesses would abduct his own primary agency, a the king should have attempted to render the indexical relationships signified by the object-as-index as unambiguous as possible. As I explain in Chapter 4, the meaning of the index hinges on both the actions of the agent and the perceiving witness. The agent invests (following Gell, 1998, p. 68; see below) the index with its physical qualities and its current status in the social world (i.e., where in space it is and/or who is wielding it at any given moment). The witnesses then abduct from those investments the agency that has resulted in the indexs existence 149

PAGE 150

according to those qualities and the indexs status in coordination with their own knowledge of past actions, the capacities of others, and relationships in which others are engaged. Finally, in some contexts the witnesses are helped along in the process of abduction by the efforts of the participants at the moment of its use or exchange to discursively frame (Keane, 1997, pp. 17, 26-27) the object by highlighting some aspect of the index or its contextual relations and therefore to explicitly connect that index to a specific agent. In essence, the abducting witnesses should abduct which agent is most responsible for the qualities and status in relation to that indexs life history or biography (Gaitn Amman, 2005; Kopytoff, 1986; also Gosden and Marshall, 1999). The object-as-index stands as a material record and sign of some or potentially all of those agencies that have gone into producing and adding to the biography of the object, i.e., that have constructed the history of the objects associations with social actors. More specifically, some agent (and likely multiple agents) has been responsible for the existence of the object, the procurement of the raw material required for its production, the exchange or transfer of the object from one person to another, any modifications made to the object resulting from its use, and finally the act which has led it to confront an abducting witness (and in the case of the archaeologist as witness, this includes the manner of deposition as well as post-depositional movement and alteration). The physical qualities and physical existence at a certain time and at a certain place are the signs from which witnesses piece together the biography of the object and abduct the agent (or agents) responsible for (i.e., signified by) that index. Within the biographies of object-as-indexes I focus on the intentional actions that alter the qualities or status of the object-as-index and use to word investment (following Gell, 1998, p. 68) to describe this process of intentionally altering the qualities or status of an object-as-index 150

PAGE 151

in an effort to have that investment abducted by others. This highlights both the purposive nature of social action that is integral to efforts to encompass social others (i.e., an act must be intended, and it must be abducted to have been intended; see Chapter 4) and the fact that an agent hopes his investment of producing or altering the qualities of an object-as-index or altering its status vis--vis social actors will pay off in the abductions made by witnesses. Gell (1998, p. 68) uses the term invest in at least one passage, although he does so without explicitly discussing what he means by his use of the word. Because the word does exist in the literature (even if only obscurely), however, and because I believe it captures the intent of the agent that his actions be abducted to have had some effect fairly well (as stated above), I follow Gell and adapt his usage in order to focus on the material investments made by agents in objects-as-indexes. Certain points regarding the theoretical orientation and methodological focus of the present work bear stating explicitly before the reconstruction of an objects biography can be discussed. With respect to the methodological focus, a key to the archaeological study of the practices by which an objects biography is constructed or produced is their continual nature. The continual nature of practices, which is necessarily implied by the fact that a would-be encompassing person must attempt to regulate the abductions of others through repeated social action, allows for the archaeological study of economic processes of human-object relationships, even if the identification of specific events of production or exchange is beyond archaeological feasibility. This point is made particularly well by Barrett (2000, p. 64): Economic processes, for example, are represented archaeologically because the events of production, circulation, and consumption might all have left a material record of certain deposits and certain distribution patterns. Further, the general processes which are represented by events are then actually defined by those events. In other words the kinds of event which occur as a general consequence of the process also define that process. 151

PAGE 152

Insofar as specific and idiosyncratic events might not be especially important to the construction of reliable abductions of the indexical relationships between objects and persons, they are not necessarily important to the present study. With regard to the theoretical point raised by a focus on investments of agency in objects-as-indexes, it is worth stating that while I discuss economic relationships between human subjects and material objects, I am not necessarily concerned with the purely economic consequences of such activities. In other words, I do not focus on relations of production or exchange purely for their own elucidation or the elucidation of the economic base of a society. Rather, I focus on how these economic activities create and reproduce the indexical associations between objects and agents, which is to say how they constitute the meaning of the objects. Through material practices, actors compete over how they and the objects involved will be perceived. This could in many ways be conceptualized as the application of theories relating objects and personhood in gift exchange (Mauss, 1954; also Appadurai, 1986; Gregory, 1982; Kopytoff, 1986; Strathern, 1988), as opposed to commodity exchange, to the whole spectrum of economic activities. This approach follows other archaeologists (e.g., Chapman, 2000; Gaitn Amman, 2005; Helms, 1988, 1993; Thomas, 1999) as well as anthropologists (Thomas, 1991; Weiner, 1992) who have similarly studied how the previous associations of objects with persons or places of origin are integral to understanding their meaning and/or how they were used in social action. In order to analyze the investments made by agents in objects-as-indexes and whether and how they might lead to abductions made by witnesses concerning the agency signified by the object-as-index, it is necessary to reconstruct the biographies of the objects. By reconstructing biographies of objects-as-indexes, it is possible to identify the agents that have made some 152

PAGE 153

investment in that object over the course of its biography, and the likelihood that that investment will be perceived by a perceiving witness. Below I discuss the stages of an objects biography in turn, beginning with the acquisition of the raw materials. Acquisition of Raw Materials The raw materials for any object must first be acquired. A producer might acquire the raw material directly from its natural state or might acquire it from other agents. In the latter context, the agency (or agencies) of the persons who have won the raw material from its natural state (and perhaps have processed it into a usable form, as in the case of smelting metal) will be invested into the raw material. This investment may be preserved even as the raw material is transformed into another object. Indeed, as discussed by Thomas (1999) the point of acquiring and using a certain raw material is to have the significatory relationships of the raw material (in this case between the raw material and specific places) imbue the object produced with that significance. A community of witnesses, however, might not necessarily be aware of where an object or raw material comes from, especially if the raw materials point of origin is far away from the witnesses and/or the witnesses do not have much knowledge of the area or region from which the raw material comes. In such cases, it is very much an agentic act to acquire raw materials from distant and potentially unknown regions either through ones own agency or through personal contacts with people in those areas, as Helms (1988) discusses. In such cases, the agency of the agent that is capable of displacing the raw materials from their point or region of origin to a novel and foreign area will be invested in that object. Analyses that determine the sources of raw materials through chemical characterization are therefore useful in determining where the raw material came from and how producers acquired it. If producers acquired the raw materials necessary for production from other agents, the nature of this interaction (i.e., market based exchange, down-the-line trading, gift exchange; Renfrew, 153

PAGE 154

1975) can be determined through distributional patterns of artifacts attributable to the same source and/or artifacts of the same class from different sources, as I discuss below under the topic of exchange. Furthermore, ethnohistoric information can aid the identification of the networks through which raw materials were acquired. For example, I rely upon Gorenstein and Pollards (1983, pp. 108-111) contention that through his cadre of long-distance merchants (as discussed in the RM), the Tarascan king monopolized the ability to acquire goods from outside of the Tarascan kingdom. Therefore goods that are determined to be from outside of the Tarascan kingdom through chemical characterization are assumed to have flowed to producers via the king. Production Production is inherently an indexical relation between producer and object because it is a relation of causality (see Chapter 4 on causality and indexical relationships) (Gell, 1998, pp. 23, 33; also Wailes, 1996). The object produced would not exist if not for the efforts of the producer. Moreover, the producer determines the nature of the objectits qualities and characteristics. Therefore a producer invests his own agency in the objects existence as well as its specific qualities. Gells (1992) discussion of technologies of enchantment is relevant with regard to the qualities of the object produced. Technologies of enchantment are technologies or productive processes that are difficult or require special knowledge, magic, tremendous skill, and/or large amounts of time (Gell, 1992; see also Helms, 1993; Hosler, 1995). By seeming almost impossible or beyond (normal) human capacities, technologies of enchantment and the objects that they produce immediately confront witnesses to attempt to imagine the excruciatingly difficult process that produced the object, and therefore to imagine the producer and his high level of skill (Gell, 1992). Therefore technologies of enchantment are often effective ways to 154

PAGE 155

invest ones own agency within an object, because the object lends itself to the abduction of its production and not some other aspect of its biography (Gell, 1992). Archaeologically, replication studies that have attempted to produce the same kinds of objects as are found in the archaeological record can reveal the difficult nature of the productive process and the level of skill and devotion required in order to produce such objects and become proficient in their production. With respect to obsidian lapidary objects, I believe that obsidian lapidary production can be classified as a technology of enchantment due to the difficulties inherent in the production of highly polished and intricately shaped lapidary objects from a tremendously brittle and fragile raw material, as implied by Otis Charltons (1993) study. Production can be identified archaeologically through the recovery of debitage and other by-products of the productive process. Here Ingolds (1993) conceptualization of the taskscape aids the interpretation of artifactual indicators of production. The taskscape is the array of related tasks that take place in space and over time, and because the landscape is inherently temporal the taskscape is embodied, or congealed within the landscape (Ingold, 1993, pp. 162). In other words subjects perceive in the landscape the tasks that are congealed within that landscape, due in part to the effects that those task have upon the landscape. Using such an approach, it is helpful to interpret the discard of such objects and their enduring nature within the landscape as the active signification and citation (Butler, 1993; also Joyce and Lopiparo, 2005), or reiteration of the importance of past actions, of the past production of certain kinds of objects at that certain place. By reproducing the productive process and continually depositing the by-products of that productive process, the accumulation of those by-products signifies that the place is a place of production, and that production in the present is proper and 155

PAGE 156

actively cites (Butler, 1993), or reproduces and reiterates the importance of, those past instances of production. By adopting Ingolds (1993) notion of the taskscape, then, the by-products of production signify the presence of that production to modern witnesses (archaeologists) and signified the presence and properness of that production to witnesses in the past as well. Furthermore, to the archaeologist the by-products of a productive process necessarily instigate the abduction that actual people were engaged in a productive process, and the presence and activities of the producers were perhaps open to direct witnessing in the past. In this regard it is interesting to note that in the RM (1956, pp. 39, 50, 65, 122), characters are often encountered by others as they are busy making arrow points, and therefore it is no leap of faith to imagine that the task of working obsidian in general was something that was public and subject to observance by a wide range of witnesses. In addition to identifying production through by-products of the productive process, distributional patterns of the objects produced can be used to identify production locales. This is posited by Stark (1992), who proposes that an unusually high proportion of objects (in Starks example, of ceramic objects of a particular type) at a given locale in relation to the proportions of the same objects at other sites suggests that the objects were produced at the locale exhibiting high proportions of the objects. The suggestion that comparatively high concentrations of a certain object at a specific locale indicate production at that site will be used in order to identify the production of a certain ware of ceramic objects at Erongarcuaro, in much the same fashion as in Starks case study. Patronage of Producers The relationship between a producer and the object produced through his efforts may not necessarily be so straightforward; other agents can intervene in the process of production. An 156

PAGE 157

important relation examined in the present study is the patronage of producers. By patronage I mean that patrons have directly provided the producers with their livelihood, have provided the raw materials necessary for that production, and have some influence over the productive process and the nature of the products of that process. This follows Gells (1998, pp. 24, 33) definition and discussion of patrons, although I dispute Gells (1998, p. 36) position that the artist (producer) remains the primary agent with respect to the product that he has produced. This position contradicts some of Gells other discussions of patronage, in which the patron is the primary agent in the production of an art object because he has willed its existence and willed that it possess certain characteristics (as in the example of patron commissioning a portrait of himself [Gell, 1998, p. 37]). Gells position that the artist remains the primary agent in regard to the object produced also ignores the material support given by patrons to producers that allows the producers to hone their craft and become able to produce the objects that their patrons desire and require, as in Gells (1998, p. 24) own discussion of the craftsmen of Shah Jehan. This is especially the case in crafts that Clark and Parry (1990, p. 293, adopting the term from Malinowski) call hypertrophic and which require greater output than is required by the needs of patrons or society at large in order to maintain virtuosity in the productive process. Such dedication to a craft often requires the direct material support of a patron, and without that support the skill of the producer, and so the object produced, would not exist. This kind of patronage apparently existed in the Tarascan kingdom in the production of metalwork, as discussed by Pollard (1987). The king had a number of metalsmiths who provided metal objects at his instruction or demands. The document that Pollard analyzes does not necessarily indicate that the king directly supported these smiths, but clearly the king intended that these smiths produce objects for him and initiated a relationship with those smiths in which 157

PAGE 158

his will and requirements determined the production of those smiths and the destination of their output (i.e., his own coffers). Furthermore, given the highly difficult nature of producing obsidian lapidary objects discussed above, it is likely that this activity was hypertrophic and required constant practice. Constant practice would have limited the other economic pursuits that producers were able to engage in, and therefore patronage was likely integral to the practice of this craft activity. To that extent, numerous indicators of craft production supported by elites or high status patrons (often called attached specialization; see Brumfiel and Earle, 1987; also Costin, 2001) have been proposed. These indicators include the location of production within or next to a high status residence, structure, or area; use of restricted or controlled materials that only high status persons would have had access to in such production; and finally the limited extent to which the products were distributed among the populace. With respect to the last indicator, the assumption is made that if such goods are only found associated with high status individuals these individuals likely directly appropriated such objects for their own personal use or possibly exchange with other individuals, presumably also of high status. Where production is demonstrably associated with patronage, this can signal the prominence of a significatory relationship between patron and product, over that of the relationship between producer and product (Gell, 1998, p. 24). In other words, the object is invested with the agency of the patron over the agency of the producer. Exchange The establishment of indexical relationships through exchange relationships is more complex than the other relationships examined in the present study. This is because the meaning that is abducted from any one exchanged object will be abducted according to the relationship between the object and exchange participants as well as the other objects previously exchanged 158

PAGE 159

or even yet to be exchanged between those participants. In what follows I discuss an analytical framework that allows for the investigation of this complex web of relationships between objects and persons that is constituted through time and that plays a determinative role in the agencies invested into and then abducted from any single object. Numerous studies of gift exchange relationships have recognized that such relations are constituted through time and are objectified and indexed by the objects that constitute the relationship. 1 Gift objects are objectifications of the agencies of the participants as well as the exchange relationship between those participants. The objects exchanged are therefore indexes of the exchange relationship because their transfer has caused the relationship to exist. The significatory relationship works in the reverse direction as well, as the perceived nature of the exchange relationship as a whole, composed of the history of gifts and counter-gifts that have flowed between the participants determines how any individual gift is evaluated. In other words, the meaning of any gift can only be determined with reference to the other gifts that have been given over the history of the exchange relationship. The exchange relationship has a syntaxa sequence of gift exchangesand this syntax determines the meaning of any of the individual gift actions. For example, the question of how proper a certain gift is to the goal of continuing a relation of alliance between two families will be evaluated in relation to the previous gifts that have established and perpetuated that alliance. If a gift is particularly meager in relation to the previous gifts, this latest gift will likely be perceived as an indication that the giver does not value his relationship with the recipient and therefore as an insult. Conversely a meager gift 1 The vast anthropological literature on exchange, particularly gift exchange, prohibits anything approaching an exhaustive list of citations. I have been primarily influenced by Appadurai (1986); Battaglia (1990, 1992); Bourdieu (1977, pp. 4-15; 194-197); Gell (1998, pp. 106-109); Gregory (1982); Kopytoff (1986); Leach (1965); Mauss (1954[1925]); McKinnon (1991); Munn (1986); Parry (1986); Orenstein (1980); Strathern (1971); Strathern (1988); Wagner (1977); Weiner (1992). 159

PAGE 160

could be interpreted as an indication that the giver does not have the wherewithal to produce a valuable gift and therefore is of little use as an ally. The complex nature of exchange relationships has important implications for the ability of some agents to encompass others. Because the exchange relationship as a whole impacts the nature of the abductions made from the specific gift objects, the encompassing agent should attempt to influence the nature of the exchange relationship, that is, the give-and-take sequence of actions that constitute the relationship and therefore impact the abductions made from the objects involved in that exchange relationship. This is most easily accomplished by choosing exchange partners that are unlikely to be able to constitute the exchange relationship as a symmetric one, i.e., by proffering a gift that is equivalent to the original gift. In other words, if counter-gifts are rendered impossible or are less highly valued than the original gift, this would be evidence of the superior agency of the giver. Particularly in the case of the exclusion of a counter-gift, the giver has taken a primary and exclusive role in the establishment of the relation between giver and recipient. Without the givers intention and actions, the relationship would not exist, and so the relationship should be abducted to be the result of the givers own agency. This abduction is also applied to the object that has constituted the relationship, as it should be abducted to be the means through which the giver has constituted the relationship and therefore as an index of the givers agency and personhood. In other words, agents can invest objects with their own agency by investing in the overall exchange relationship in which it has flowed, thereby constituting that relation in a specific manner. Finally, if the object continues to be abducted to be an index of the giver, and this indexical relationship impacts how the actions that involve that object are perceived, it is questionable whether or not the object has been given at all. Weiners (1992) phrase keeping-while160

PAGE 161

giving captures this paradox wellcertainly such a gift is not given with no strings attached and the act of giving is not an act of kindness and self-sacrifice in which the giver foregoes any future claim on the object (Gregory, 1982; Strathern, 1988; Weiner, 1992), the actions it facilitates (Gell, 1998, Chapter 7), or the material yield that it produces (Bourdieu, 1977, pp. 3-10; Sahlins, 1972). The gift that continues to index the giver and therefore is not really given becomes precisely what Gell (1998, pp. 20-21, 96-154) calls the distributed personhood of the giver. It is the means through which the giver achieves his effects across space and down through time. In the analyses of objects and the gift relationships that involve them in Chapter 7, this question of whether objects (such as insignia or royal brides) are given in the minds of participants and abducting witnesses is vitally important to analyses of the rhetorical framing of exchange relationships and the objects that flow within those relationships. The necessity to examine sequences of exchange acts in any exchange relationship may seem to render archaeological examination of the indexical relationships of objects difficult or even impossible, as the archaeological investigation of singular gifts and counter-gifts is difficult or even impossible. This difficulty can be worked around in two ways, however, which ultimately complement and reinforce one another. The first way around the contingent and transformable meaning of objects in the exchange relationship is to view these multiple exchange events as parts of a larger exchange process (following Barrett, 2000, p. 64), as discussed above. Viewed as separate events, archaeologists might not be able to know the specific indexical status of an object at any point in time, but who is winning the war over indexicality as a whole and through time can be archaeologically analyzed as certain kinds of objects repeatedly flow in one direction in relation to other objects repeatedly flowing in the other direction. It is precisely these repeated and reproducible gift 161

PAGE 162

actions that would have led to a greater likelihood that witnesses would abduct the primacy of one agency over another over time and in relation to the past. The second way around the difficulty of archaeologically analyzing individual exchange acts is that indexical relationships are abducted in the minds of the participants and on-lookers with reference to their knowledge of the material world and the place of the participants within that material world (Gell, 1998, pp. 230-232; Keane, 1997, pp. 232-233; Munn, 1986). Participants in exchange relationships will necessarily formulate strategies based on their knowledge of the others capabilities to harness the necessary resources involved in a counter-gift and the likelihood of such an outcome (Bourdieu, 1977, pp. 3-10; Gell, 1998, pp. 230-232; Keane, 1997, pp. 232-233). On-lookers and witnesses should do the same. The ability of actors and on-lookers to construct internalized models of wider capacities and previous transactions is the result of their being socialized in a communally constructed context that emphasizes not only knowledge of the rules of the game but also how to play the game well (Bourdieu, 1977, pp. 3-10; Gell, 1998, pp. 230-232). Assuming that most members of a society will have some knowledge of the rules of the game and how to play the game well, these members will have the ability to form internalized models of those transactions and capacities that will affect the nature of relationships and the transactions involving other members of society. It is with respect to these models, and the discourses (or should I say gossip) that both aid in their construction and follow from them, that the indexical relationships of the objects concerned will be abducted. In the meantime, or as time passes between individual gift events, witnesses will abduct the agency behind the transfer of the object-as-index to be the more appropriate party based on their knowledge of the capacities of the participants (following Keane, 1997, pp. 232-233). These capacities will be judged according to knowledge of (or yet 162

PAGE 163

further abductions concerning) the participants own ability to produce (in a loose sense, i.e., bring forth) the counter-gift that is called for (as implied by Bourdieus [1977, pp. 11-13] recognition that the initiation of an exchange relationship presupposes some knowledge of the appropriateness of certain potential partners). Such judgments will be made by witnesses in accordance with their knowledge of the places of the participants in a chain of other exchanges, obligations, and productive capacities that could provide the participant with the counter-gift. This knowledge possessed by witnesses, or their abductions concerning the capacities of the participants, is constituted by or made in reference to the material world. Witnesses will abduct the capacity to produce certain objects by the debitage and discards that are active signs (indexes) on the landscape that signify that capacity (see above). The witnesses knowledge of a persons place in a network of exchange will be abducted from those actions that the witnesses have witnessed in the past or from those objects that the participants possess in the present. For example, the fact that a person possesses a highly valued object from far away could be abducted to index that persons importance and connections to a network through which he could obtain more such valuables. The material world that plays a vital role in the abductions made by witnesses in the past is accessible archaeologically. Therefore just as on-lookers can gauge the capacities of the participants to reciprocate objects through an evaluation of economic capacities, archaeologists should be able to gauge the reasonableness of an expectation that a reciprocating gift of equivalent value will or will not be proffered through a consideration of the place of the recipient in wider economic networks of production and exchange. The process of archaeological interpretation then becomes essentially abducting what the abductions made in the past would have been based on the same material world. Through reconstructions of economic networks 163

PAGE 164

and the places of actors within those networks as producers, givers, and recipients, archaeologists can judge whether or not those actors had the wherewithal to answer challenges to reciprocate. In the preceding discussion of objects and the indexical relationships established through exchange, I have explicitly made use of the analyses of gift exchange in contrast to commodity exchange. Therefore in order to analyze how certain objects index social actors it is necessary to investigate whether or not these exchanges took place in contexts that were constructed as socially meaningful and which meaningfully linked the objects given with the social actors exchanging them as in analyses of gift exchange (Appadurai, 1986; Gregory, 1982; Kopytoff, 1986; Mauss, 1954; Munn, 1986; Strathern, 1988) or if they changed hands in less meaningful contexts in which the primary referent of the object was the use value of the object, and the social relationship thus constituted did not extend in any meaningful way beyond the immediate transaction (and thus behaved like commodities and were likely embedded in market exchange; Appadurai, 1986; Gregory, 1982; Kopytoff, 1986). Exchange has become a subject of great interest in archaeological investigations of complex societies. In particular, the identification of market systems has garnered much attention. Methods have been developed that rely on interpreting the distributions of artifacts in order to identify marketplace exchange. These methods can be used to identify market-based exchange and determine which kinds of objects were exchanged in market or in gift transactions. A key feature of market places is that they pool goods together in one place (Feinman, 1985; Feinman, Blanton, and Kowalewski, 1984; Flannery and Pires-Fereira, 1976; Hirth, 1999). Consumers, distributed in space but congregating at the market, then purchase goods from a range of producers and distribute goods from a wide range of producers relatively evenly across the area served by the market place as they take these goods home (Feinman, 164

PAGE 165

1985; Feinman, Blanton, and Kowalewski, 1984; Flannery and Pires-Fereira, 1976; Hirth, 1999; Pires-Fereira, 1976; Smith, 2001, 2003; Winter and Pires-Fereira, 1976). Within market-based exchange, furthermore, the frequencies of goods in any particular context should be skewed toward locally produced goods. This hypothesis is based on a number of assumptions (Feinman, 1985; Feinman, Blanton, and Kowalewski, 1984). The first assumption is that goods brought to the market place from farther away will incorporate the higher cost of transportation into their prices. The second assumption is that the production of goods takes place on an even playing field, meaning that the same amount of labor and resource input is required for the production of a single good. This should be the case so long as all producers are producing their goods using the same methods and with roughly the same amount of technical skill. The last assumption is that consumers are rational actors who will opt to purchase goods at a lower price when possible. In the case of the Tarascan core zone, Hirshman (2003) has demonstrated that with the advent of the Tarascan kingdom there was not a reorganization of ceramic production in favor of more intensive means of production, which indicates that ceramic goods required the same amount of labor input than in previous eras. Moreover, without differential organization of production (Hirshman, 2003), the cost of production per unit produced should have been comparable across wares. Therefore the predicted skewing toward locally produced wares should occur within market-based exchange at Erongarcuaro and Urichu. I assume that generalized acquisition and deposition of ceramic vessels, reflected in the total sherd samples from both sites, should reflect this skewing of wares produced at potentially multiple locales and therefore should reflect market-based exchange (and following the discussions of Pollard [2003b] and Hirshman [2008]). 165

PAGE 166

I also assume that departures from the profile of wares in specific subsets of the total sherd sample (e.g., decorated serving vessels) indicate a departure from this market-based acquisition. Once objects that circulated outside of market-based exchange networks have been identified, sourcing studies and the analysis of distributional patterns can be used to determine where these objects came from and what other agents or locales were involved in their transfer. By identifying production locales, the chemical characteristics of objects produced at any particular locale, and the frequencies of such objects among a network of exchange partners can indicate the participation of exchange partners in similar or the same networks, such as in down the line trading (Renfrew, 1975). Rhetorical Framing and Cosmological Authentication In the biography of any object, at some point it will doubtless be the subject of debate and commentary, as interested parties attempt to sway how one another as well as a community of witnesses will perceive the object and abduct the agency behind the object and the action(s) in which it is involved. Keane (1997, pp. 17, 26-27) calls this discursive process in which participants in any action explicitly comment upon the relationships between the objects and persons involved in that action the framing of the object and/or action. Explicit commentary on the participants past relationships with the object and with one other can provide background information that might potentially alter how the object and the action are perceived and whose agency will be abducted to be primary in relation to other agents. The discursive framing of the object and the action should, moreover, be appropriate to the actors intentionsthe actor doing the talking should emphasize (or perhaps even fabricate) certain past relationships (e.g., my brother so-and-so made this thing with his own hands so that I might give it to you) that help make his point and in the end help the action being realized achieve its intended effect (e.g., the establishment of a debt relation). 166

PAGE 167

Framing an object given can also involve the promotion of the abduction that the return of the object to its rightful owners is a certainty or mere eventuality (Weiner, 1992, pp. 26, 38-42, 60-63). Where such an abduction can be prompted through discursive framing, the indexical link between an object given and the giver remains intact (Weiner, 1992, pp. 61-63). In contexts where such an expectation does not exist, i.e., participants and on-lookers alike do not regard the return of the object to its rightful owners highly likely, the indexical link can be broken or at least superseded by new relationships (Weiner, 1992, p. 64). The abduction of the future realization of an objects return can be prompted through constructing and proclaiming ideal biographies (Gaitn Amman, 2005) of the objects exchanged, such that an object given away should return to the giver after a specific period of time or upon the realization of some specified event. Certain passages in the RM, such as the representation of the investiture ceremonies of local lords, can be analyzed precisely as attempts to actively frame objects by constructing ideal object biographies (see Chapter 7), and I assume that these efforts to frame those objects in the prehispanic era would have included much the same rhetoric as the representations of the RM. I make this assumption because the end goal of the actors doing the framing would have been the sameto explain how and why the insignia of office and their transferal to subordinate lords should have indexed the king. Additionally, Weiner (1992, pp. 4-5) discusses the process of cosmological authentication in which a durable relationship between an object and a kin-group is proclaimed. Cosmological authentication is a process in which myths, beliefs, and histories explain how or through what means the kin-group and the object became historically associated to the extent that they mutually implicate and are constitutive of one another. Cosmological authentication frequently, and most effectively, locates the foundations for this association outside the present, 167

PAGE 168

in the sanctified past (Weiner, 1992, p. 101). Through the construction of an origin outside of the here and now, the process of cosmological authentication lends an air of sacredness to the relationship or, at the very least, casts the relationship between the object and the group it signifies as historically durable and therefore right, proper, and inevitable. Cosmological authentication only works, however, if it is actively and continually proclaimed. Because cosmological authentication involves myth, legends, and rituals in an effort to sanctify the origins of the relationship between an object and a kin-group, it is best investigated through ethnohistoric means, and particularly native historical traditions. In the present study I employ a structuralist approach to the analysis of narratives concerning the past in order to elucidate the manner in which relationships in the present were supposedly the result of the reproduction of relationships that originated in the past. By utilizing a structuralist method of narrative analysis, following most closely the methods of Sahlins (1981, 1985a) and Turner (1977, 1985), the actions, interactions, and relations between characters can be analyzed as the establishment of the categories to which persons in the present belonged (i.e., they were reproductions of those categories, or tokens of a type [Turner, 1985, p. 66, 82, 91-93]) and the proper relationships between those categories. I detail these methods further in Chapter 6, in which I go on to analyze the legendary history of the Uacusecha contained in the RM. The actual realization of discursive practices such as framing and cosmological authentication in the past cannot be regarded as certainties, however. Some evidence that they took place is needed in order to conclude that certain parties were able to frame the object and the action in which it was involved. In order to identify the degree to which discursive framing took place and who participated in that framing, the material correlates of scenes in which certain 168

PAGE 169

actions (in this case, orations and discourse) were appropriate (e.g., Richardson, 1982) must be identified. The ethnohistoric record is again useful in this regard. The RM describes the contexts of the exchanges through which encompassment would have been realized (if the discursive practices represented in the ethnohistoric record represent reality). Within these descriptions there are indications that priests (most likely the traveling priests called curitiecha; see description of these priests in the RM [1956, p. 181) were omnipresent as the officiates of such exchangesthey were the ones who raised nobles to the office of cacique and who conducted the wedding ceremony involving royal brides and local lords (RM, 1956, pp. 203, 208-210). In the description of the ceremony at which the chief priest (the petamuti) related the history of the royal dynasty, the RM (1956, p. 14) also states that priests were sent to all of the towns of the kingdom to relate that same history. In short, the RM indicates that priests traveled from the capital to the provincial areas in order to officiate and rhetorically frame the historical roots of the relation between the king and subordinate lords as well as particular events. As agents of the state, these priests (both the traveling priests and the head priest) carried the words of the king throughout the kingdom. They were directly (proximately) responsible for the actions that would have produced the intersubjective encompassment of the local lords within the kings personhood, but the priests should also have been encompassed themselves as the kings representatives. While their place as encompassed agents deserves investigation in its own right, unfortunately such an investigation is outside of the scope of the present study. Some background information on who these priests were, is appropriate here, as well as a description of why exactly they should have been perceived as representatives of the king (which also 169

PAGE 170

explains why these priests cannot adequately be examined through the archaeological means used in the present study). The petamuti and the traveling priests were likely members of the nobility. The RM (1956, p. 181, my translation) states that all [the priests] were married and came to their offices by their parentage [por linaje]. Pollard (1993, p. 146) believes that the class of full-time priests that included the traveling priests were probably limited to the nobility. The insignia of the priests included the painted and greenstone encrusted gourds (discussed below) as well as the staffs that they carried. The priests staffs are worth discussing because they were important objects that indexed the fact that the priests spoke for the king. Priests are depicted numerous times (including in the illustrations of the priests installing a nobleman as cacique and marrying a nobleman to a woman from the kings palace) in the illustrations of the RM holding staffs with stone points at the tip and feathers hanging from the top. Stone (2004, p. 116) contends that the indigenous word for this staff is most likely given by Gilberti. Gilberti (1962, p. 505) translates yrechequa tsiriquarequa as vara real [royal staff], and Stone (2004, p. 116) hypothesizes that this staff was a symbolic representation of the kingdom [the irechequa]. She interprets the blue precious stone as iconic of Lake Ptzcuaro, the heart of the kingdom, and the multicolored feathers as symbolic of the four directions of the world encompassed within the kingdom (Stone, 2004, p. 116). As a symbolic representation of the kingdom, the staff gives its holder the ability to speak to the kingdom on behalf of its political (and moral/religious) leader. A passage in one of the Relaciones Geogrficas (see Chapter 2) further helps shed light on how this staff was perceived in local contexts. Stone (2004, p. 268n14) notes that in the Relacin de Cuiseo de la Laguna, indigenous informants in Cuitzeo stated that when a judge possessed a staff similar to the priests staff, the staff denoted the power that he possessed from his king [denotaba el poder 170

PAGE 171

que tena de su rey]. Likely not coincidentally, an entry in Gilbertis (1962, p. 505) Vocabulario translates an indigenous term (cataperaqua tsiriquarequa) as vara de justicia [staff of justice]. These staffs, then, appear to have been integral to the perception among the commoners that the judges, and also very likely the priests who carried similar staffs, executed the will of the king and that the words of these officials were the words of the king. While these staffs were integral to the perception of the priests as agents of the state, they are not readily traceable archaeologically, if at all. For this reason, I assume that priests were perceived as acting on behalf of the king and therefore able to put in motion the encompassment of the subordinate lords. What can be examined, however, is the production of scenes at which these priests, traveling on behalf of the king and from the capital, were present and actively framing the relationships being produced through material relations. Smoking tobacco from pipes was a prominent activity within the ceremonies in which the subordinate lords and king interacted and which involved priests acting as royal emissaries. Smoking was highly important at all religious occasions. The chief priest carried a gourd on his back in which he kept tobacco for smoking, and the traveling priests also carried gourds on their backs, although it is not explicitly stated whether or not their gourds held tobacco as well (RM, 1956, pp. 181-182). All of the lords depicted in the illustration of the administration of justice (e.g., executing wrongdoers) that followed the recitation of the legendary history of the royal dynasty by the chief priest are shown smoking pipes (RM, 1956, p.11). Furthermore, all of the scenes of exchange by which the king would have encompassed subordinate lords were replete with religious significance. The royal bride being married to a cacique was instructed to make mantas and food for the god Curicaueri, and in the legendary history of the RM this responsibility on the part of the bride is made explicit in the marriages between Uacusecha men 171

PAGE 172

and women from Naranjan and Coringuaro (RM, 1956, pp. 15-16, 65, 208-210). The raising of noblemen to the office of cacique included the admonishment to make sure that the god Curicaueri was well cared for by having wood brought to the temples and going to war for (and with) him (RM, 1956, pp. 203-204). In addition to the production of scenes at which exchange acts would have been framed, the Tarascan ruling dynasty had a keen interest in regulating the production of local histories and the relations that they would have proclaimed. This regulation is evidenced by certain aspects of the corpus of documents relating to the town of Carapan (in the Sierra Tarasca west of the Ptzcuaro Basin) especially as discussed and analyzed by Roskamp (2003; see chapter 1). Roskamp (2003, pp. 333-337) notes that these documents, particularly the Cdice Plancarte, Lienzo de Carapan, and the Lienzo de Ptzcuaro all indicate that in the early colonial period the indigenous officials Don Antonio Huitzimengari and Don Pablo Cuiru played a prominent role in affirming the rights and boundaries of the town of Carapan, the historical origins of those rights and boundaries, and assuring that these rights were documented and supported by important officials such as themselves (Roskamp, 2003, pp. 309, 333). The case of Don Antonio Huitzimengari is of more direct concern, because he was the indigenous governor of Michoacn from 1545 until his death in 1562. As the indigenous governor of the province and son of the last indigenous king, Tzintzicha Tangaxoan, he certainly qualifies as a colonial era equivalent of an agent of the state/royal dynasty (Don Pablo Cuiru, on the other hand, was the governor of Carapan itself). For my purposes the importance of this action is that it established royal or stately visitations by high ranking members of the indigenous nobility in the early colonial era, and this visitation was explicitly concerned with the supervision of the production of (in this instance) local histories and ensuring that they conformed with the wishes of the higher ranking 172

PAGE 173

elites. This concern with regulating the production of history had obvious parallels to the precolonial era as it is represented in the RM, in which it is claimed that priests traveled throughout the kingdom to relate some form of official state history to the populace (see above). In sum, the presence of ceramic pipes should indicate the production of a scene at which objects and actions would have been framed. Pollard (1993, pp. 203-204) also notes that mini-bowls perhaps had ritual associations, based on the greater frequency of mini-bowls in the ritual areas of Tzintzuntzan. She suggests this possibility cautiously, however, and therefore I use mini-bowls as only a possible indicator of ritual activity and discursive framing. These objects, moreover, can be used to determine the extent to which agents of the central state administration (members of the priesthood or officials sent from one of the Uacusecha capitals) were present in the local contexts of Erongarcuaro and Urichu. Because they are ceramic, the production place of pipes and mini-bowls can be ascertained by using the same studies that have linked the paste classifications in the Ptzcuaro Basin developed by Pollard (1993, Appendix 2) to clay sources in the Ptzcuaro Basin. One paste in particular originated in the south or southeast part of the Ptzcuaro Basin, and Pollard et al. (2005) have suggested that objects composed of this paste were possibly produced at Ihuatzio, one of the three Uacusecha capitals. This possibility is not definite and should be regarded as preliminary, although at present it is the best available means of identifying the presence of state agents who would have framed objects and actions. Given the relatively low costs and presumably low specialized knowledge required to produce pipes locally, I assume that state provisioning of ceramic ritual paraphernalia was unlikely, and therefore the presence of non-locally produced objects was most likely the result of visitation and shared participation in locally performed rituals and ceremonies at which the relationships between the king and subordinate lords would have been framed. 173

PAGE 174

Religious structures could also have been important to the production of a scene in which objects and actions could be framed as an attempt to influence the agency abducted to be behind them. The uniquely shaped ycatas (temple platforms that are composed of a circular pyramid appended to a rectangular pyramid at its midpoint), so prominent within Tarascan territory, were generally associated with the Tarascan state religion and with the patron god Curicaueri and likely also his celestial male brothers (Acosta, 1939; Castro Leal, 1986; Chadwick, 1971; Noguera, 1931; Pollard, 1991, p. 172; Rubn de la Borbolla, 1941). These structures were likely related to the framing of state-local relationships, but not necessarily the specific exchanges that constituted the encompassment of the subordinate lords. The use of these ycatas as stages for rituals of the state cult of the god Curicaueri would also have likely involved state religious and possibly even secular officials. These officials likely would not have missed the opportunity to explain to local congregants how their community and lord were related to the state and royal dynasty, i.e., to enumerate the social and cosmic bonds between the local lord and the king. At the very least these structures were permanent reminders of Curicaueri and the relations between the god, the king, and the local nobility. Therefore the presence of ycatas can be interpreted as material structures that would have facilitated the discursive framing of such (exchange) relationships. Mortuary Ceremonies As explained above, objects were not the only important thing exchanged between the king and subordinate lords. Women as wives were also highly important to the ability of the local lords to perform many of their duties as lords. The RM (1956, pp. 182, 208-210) indicates that women were exchanged in both directions between the king and subordinate lords, however, meaning that the relation of encompassment was at least potentially negated through reciprocation. Therefore it is necessary to investigate the manner in which the exchanged 174

PAGE 175

women were or were not equivalent in status, as well as the way in which their ties to the noblemen and king were materially and publicly proclaimed. The RM (1956, pp. 207-210) states that the king married off women of his house to subordinate lords in order to establish or cement his relationship with those lords. The relationship between the women and the king need not have been severed by the marriage, however. Rather, if the king wished to implicate himself in the ability of subordinate lords to fulfill their duties, through the perpetuation of a relation with the bride, his relative, he should have invested in the public and material proclamation of that relationship. The relation between the king and subordinate lord as constituted through marriage exchange can be analyzed in the same fashion as gift exchange, insofar as the gifts surrounding the marriages themselves serve to sort out the affiliation of the bride and the status of her productive and reproductive capacities as fully or only impartially transferred (Barnes, 1980; Kuper, 1982; McKinnon, 1991; Strathern, 1988). The affiliation of a woman is not only determined at her marriage, however, but is a source of negotiation and potential contention over long spans of time, including at her death (e.g., Battaglia, 1990, 1992; McKinnon, 1991; Mosko, 1985). Therefore mortuary contexts can be examined as material representations of links between the deceased woman and living members of society, namely the king and subordinate lord. The analytical approach to mortuary remains and actions that led to those remains utilized in the present study attempts to build upon the important analysis of Pereira (n.d.) and his discussion of the burials excavated at Urichu by Pollard (Pollard and Cahue, 1999). Pereira concludes that, based on the grave goods and the elaboration of the burial chamber of males and females at the sites of Milpillas, Urichu, and Huandacareo, females were more highly ranked than the local (male) lords and therefore were likely involved in the general practice of marrying 175

PAGE 176

royal brides to local lords in order to cement alliances (see above). While Pereiras research is an important step, it does not take into account the reasons why the wealth of the women should be displayed and subsequently interred with the women as opposed to a range of other possibilities. In short, it falls victim to the same criticisms leveled at the Saxe-Binford program which became focused on the general position that mortuary practices and specifically the grave goods and grave elaboration associated with the deceased were direct reflections of the roles, ranks, and statuses that the deceased had filled in life (Binford, 1971; Saxe, 1971; also Peebles and Kus, 1977; Tainter, 1975, 1978). Criticisms of the Saxe Binford program view the rituals and ceremonies surrounding a death as a chance to rework and resignify the deceased and their relations to the living, rather than assuming that mortuary remains directly reflect society and the relations and roles that composed it (Barrett, 1990; Parker Pearson, 1982, 1999; Shanks and Tilley, 1982; see also discussions in Gillespie, 2001, pp. 76-78; McHugh, 1999, pp. 13-15; Rakita and Buikstra, 2005, pp. 6-9). Parker Pearson (1982, p. 100, emphasis in original) sums up the general thrust of these critiques: The reconstruction of social organisation through the identification of roles (whether in burial, craft specialisation, settlement hierarchies etc.) can be challenged by the theoretical stance that social systems are not constituted of roles but by recurrent social practices. Mortuary ceremonies are, due to their highly charged nature, a key arena for the realization of just these recurrent practices that weave together the social fabric. The mortuary rites, rhetoric, and the interment of objects that composed ceremonies are a commentary on the practices of the deceased during his/her life and how those practices should now be perceived (following Huntington and Metcalf, 1979, p. 93; also Joyce, 2001, p. 13). Mortuary ceremonies and archaeologically recovered mortuary contexts reflect the participation and performances of numerous persons (living, dead, and very recently deceased) as they attempt to sort out their 176

PAGE 177

relations to one another in light of the death of the deceased (Bloch, 1971; Hertz, 1960[1907]; Huntington and Metcalfe, 1979, pp. 61-67). Moreover, and due to the charged and emotional nature of the death and related ceremonies, these events are prime opportunities for the establishment of social memories that will influence how future similar social practices and relations will be perceived (Gillespie, 2001, pp. 92-93; Kchler, 1991; Kuijt, 2001). In the case examined in the present work, furthermore, it is essential to go beyond the recognition that mortuary ceremonies are highly political events concerned with succession and the augmentation of the power of an office, which of course they frequently are (Gillespie, 2001; Huntington and Metcalf, 1979, pp. 121-183; Knsel, 2006; Kus and Raharijaona, 2001). It is also necessary to examine more than simply the relation of the deceased to the entire community of the living or the other ancestors as a whole. Rather, it is necessary to examine how the living relate to one another and demonstrate the nature of that relation through their relations to the deceased, both in life and in death (following Goodys [1962, p. 133] discussion of the reorganization of the relations of the deceased among the Lodagaa, even including the relations of friend and lover). In other words, mortuary ceremonies are also a medium of claiming that the nature of the relation between certain members of the living community, in which the deceased played a central role, should be remembered and continued in a certain way. While this is obviously not the only function or purpose behind the varied rituals and ceremonies that are performed surrounding the death of an individual, a focus on the relations between members of the living community are necessary in order to analyze the relations of encompassment between the king and subordinate lord through the royal bride. Therefore I analyze the interment of goods with the deceased as a means of relating to the deceased and representing that relation to others. If the goods that are interred with the deceased 177

PAGE 178

are a means of relating the deceased to various members of the living community, then the most productive line of analysis is to interpret the manner in which those goods are meaningful not simply because of what they are (Pader, 1980, p. 143) but also because of who (or potentially where) they come from. I therefore analyze how the representation of the differential burial of elite males and females at Urichu reflected the ability of noblemen and noblewomen, as well as certain interlocutors acting on their behalf, to procure certain kinds of goods and to appropriate the mortuary ceremony as a chance to put that procurement to use by interring these goods with the deceased. This is achieved by comparison to the elite burials of Tzintzuntzan (Cabrera Castro, 1977-1978) in order to demonstrate that the differences encountered at Urichu between male and female burials do not represent some other cultural motive for the differential representation of males and females through the interment of certain kinds of grave goods (e.g., the possibility that certain goods or types of goods were male or female objects unsuitable for interment with the opposite gender) (see Chapter 8). Making Abductions From Cross-Cutting Investments in the Past and Present It is unlikely, or even impossible, that the king would have monopolized the investments made in the objects given to subordinate lords, thereby ensuring that no other agency could be abducted from these objects. Rather, other persons must have been involved in his ability to encompass others, first and foremost of which was the subordinate lords themselves. In the face of cross-cutting investments made by multiple agents, the abductions of witnesses would have been anything but assured. This would have been damning to the kings ability to encompass subordinate lordshis encompassing nature depended on the abduction, made by witnesses, that the actions of the subordinate lords were his actions (i.e., that he had willed, intended, and set them in motion through some action of his own such as the transferal of the objects-as-indexes). Furthermore, if the investments themselves cross-cut one another, leaving witnesses to abduct 178

PAGE 179

one agency or another, how can archaeologists in the present be at all confident that the abductions that those witnesses made in the past can be abducted in the present? In order to escape this interpretive dilemma, I make two assumptions, both based on Keanes (1997, pp. 20, 22, 26) position that discursive framing and the objects being framed are indissolubly linked to one another. The first is that the practice of discursively framing the agencies behind any particular investment in the object and its current situation would be more effective than any single material practice that invested an object with some agency. I make this assumption with the benefit of an ethnohistoric record in which the pomp and majesty of the king or his agents were on full display and which likely would have been quite convincing. The production of certain scenes at which the discursive framing of objects took place, tinged with religious significance (see above) and affective impact, would have been powerful and persuasive by their very nature. The words spoken at such scenes therefore would have carried tremendous significance in the minds of the witnesses of such scenes. The very production of such scenes would have been further evidence of the agency of the king, who could (as evidenced by the scene itself) send members of the priesthood or other agents acting on his behalf to towns throughout the kingdom. I also assume, however, that there were limits to the effectiveness of the discursive framing of the objects that were integral to the kings efforts to encompass subordinate lords. Following Keane (1997, pp. 18-22), the potential for dissonance between the discursive framing of an object and the actual material investments made in that object by some agent must be recognized. If an agent of the state, in other words, had proclaimed some relationship between the king and the object (and thus the secondary agent that manipulated that object) that was not otherwise materially constituted, the dissonance between the two messages, one discursive and the other 179

PAGE 180

material, could not have been ignored. Such a dissonance, by insisting too much (Keane, 1997, p. 17) would have occasioned active reflection and the outright questioning of the attempts to frame the relation between the object (and the potential secondary agent) and the king. Therefore I assume that discursive framing would not have been effective in the absence of some other, material, investment that the king has made in the object in question from which abductions will be made by witnesses. In this section I have discussed how the concept of the object biography can be used as an integrative concept in order to abduct the abductions made by witnesses in the past concerning which agent(s) should be regarded as more responsible for the objects existence and use toward some end in any particular action. Both the abductions of those witnesses in the past and the abductions made by myself are made with reference to the material world and the discursive framing practices, that are by definition public, and, if they are to have any great impact, implicate yet further material practices. In sorting out which agency should be abducted to be primary, however, the issue is not to attempt to resolve the contradictions and ambiguities inevitably created between the multiple agentic investments or the agencies supported by discursive framing. As some archaeologists have noted, the meaning of any particular object is almost inevitably ambiguous or subject to change from one context to the nextthat is part of their nature and moreover part of their power (e.g., Hodder, 1986; Tilley, 1991, 1999). However, I follow Bauer and Preucel (2001) in suggesting that the meaning of an object should not have been ambiguous in any particular context but should have been rendered unambiguous in specific scenes of action, particularly if they were the means through which intersubjective encompassment was to be achieved. Therefore the goal in utilizing the concept of object biography is to allow for the examination of how variation in meaning was dependent upon 180

PAGE 181

context, that is, the realization of certain practices and the participation of certain actors and witnesses. It was the different contexts and knowledges, based on different material practices and traces that witnesses were privy to, that would have determined the nature of the abductions made by those witnesses upon whom the realization of intersubjective encompassment was dependent. Conclusion In this chapter, with reference to the theoretical discussions of the preceding chapter, I have outlined the specific methods and frameworks whereby the real world practices that would have led witnesses to abduct the primacy of a would-be encompassing agent in relation to a secondary agent can be identified. In other words, the task for an encompassing person becomes the task of ensuring, through a range of material engagements and social interactions (scenes of framing that involve material objects just the same), that a community of witnesses abducts his own role in endowing that secondary agent with a capacity to achieve some effect that he himself has intended. These abductions are made by a community of witnesses with reference to their knowledge of, or further abductions in relation to, the material world. This material world, in turn, consists of multiple relations between human actors and the object or objects in question, and these relations are produced through time over the course of the objects biography. Over the course of the biography the physical qualities of the specific objects that flow from primary agent to secondary agent are produced and potentially transformed, the object (and other objects) changes hands in acts of exchange, they are subjects of discursive framing, and they are interred in acts of signification. Witnesses abduct the primary agent behind these biographies or certain aspects of those biographies in reference to the material world that has been involved in the production of that biography. This same material world is accessible to archaeologists, and the same abductions can be made by archaeologists 181

PAGE 182

based on regimes of acquisition, production, and exchange, and the place of specific persons within those regimes. More precisely, archaeologists in the present can, based on the same material evidence to which those witnesses in the past had access, abduct what the abductions made by witnesses in the past would have been. However, some significatory ambiguity is largely unavoidableit is highly unlikely that the king was able to completely and unambiguously monopolize all of the indexical relationships (production, exchange, etc.) in reference to which a community of witnesses would make their abductions as to who the primary agent behind any object-as-index should have been. The king, in short, could not have ensured that he was abducted to be the primary agent in all times and in all places and in all relations. An important problem of interpretation becomes immediately relevant: if the significatory relations as revealed through archaeological analysis of production and exchange regimes should cross-cut one another, how can archaeologists in the present hope to reproduce the abductions made by witnesses in the past? In the face of such ambiguity I assume that, so long as there was some material relation linking the king to the objects that constituted the agencies of the subordinate lords, the discursive framing of those relations (if indeed realized) would have sorted out these multiple relations in relation to one another. Framing would have mitigated against ambiguity and reestablished the agentic primacy abducted to have been behind those objects so important to the material production of relations of encompassment. However, the degree to which material practices would have been achieved and subsequently knowable by witnesses in various contexts in the past is a crucial issue in archaeological interpretation. 182

PAGE 183

CHAPTER 6 COSMOLOGICAL AUTHENTICATION OF UACUSECHA POWER: ENCOMPASSMENT IN THE ETHNOHISTORIC RECORD Introduction The legendary history of the Uacusecha (royal dynasty) contained in the RM accomplishes the rhetorical explication of the legitimacy of the Uacusecha in precisely the logic of encompassment as developed by Dumont (see Chapter 4). In its basic elements the legendary history has much in common with narratives from around the world that tell the story of what Sahlins (1985a) calls the Stranger King. Stranger King narratives are told in many societies that exhibit what has been called Divine Kingship going back to Hocarts (1969, 1970) original analysis of Polynesian divine kingship (see also Feely-Harnik, 1985). Such narratives relate the sequence of events by which strangers from foreign lands invade a group of autochthons (original dwellers or land-owners), win a bride from those autochthons and appropriate the indigenous wealth of the land. They then ultimately combine within their persons dual aspects of divine kingship (de Heusch, 1982; Gillespie, 1989; Sahlins, 1985a; Urton, 1990). These studies of Stranger King narratives, particularly that of Sahlins (1985a), influenced my original analysis of the legendary history of the Uacusecha Chichimecs (Haskell, 2003), and much of this chapter is an outgrowth or further development of that analysis. The legendary history of the Uacusecha Chichimecs constructs the encompassing nature of the Uacusecha Chichimecs out of what Sahlins (1985a, p. 89) calls elementary categories. The Uacusecha Chichimecs are one elementary category, and their ability to encompass their antithesis, the Islanders constructs their status as a sociocosmic totality, subsuming within their identity all aspects of the social and cosmological universe. It is this encompassing nature that establishes 183

PAGE 184

their hierarchical superiority over the rest of society, which by definition represent only parts of the sociocosmic totality that is encompassed by the Uacusecha Chichimecs. The divide between the Uacusecha royal dynasty and the rest of society is bridged, however, by the creation of local offices of leadership as extensions of the royal dynasty and which partake of one connotation of Chichimecness, the possession of legitimate authority. While the subordinate lords are thus Chichimecs in one sense, they cannot reproduce their own legitimate authority and rely upon the royal dynasty for their own reproduction. In this chapter, I explain how this indigenous explication, grounded in the cosmologically authenticating legendary history of the Uacusecha royal dynasty, conforms to Dumontian and also Turnerian models of encompassment (see Chapter 4). The royal dynasty claims to encompass the subordinate lords by existing as the principle by which they are transformed into replicas, albeit secondary and derived replicas, of the royal dynasty. By explicitly arguing that the relationship between the king and subordinate lords was one of intersubjective encompassment, the narrative frames the relation not between persons and objects but the relation between the persons directly, and by implication would frame the material objects that constituted that relation. What follows is developed from my more detailed analysis of the legendary history of the Uacusecha royal dynasty (Haskell, 2003) as well as a more recent development of how the structure of the narrative was crucial to the reproduction of identities between the king and subordinate lords and how that structure might have been related to material practices in the Tarascan kingdom (Haskell, 2008). Here I develop an interpretation of the legendary history in order to draw out the explicit parallels between the rhetoric of the legendary history and the model of encompassment developed by Turner (1984; see Chapter 4) and examine how this legendary history was used to frame relations between the king and 184

PAGE 185

subordinate lords. Because the analysis presented below relies heavily on my previous works (especially Haskell, 2003), however, I will begin with a discussion of the methods employed in the analysis of the legendary history in those works as well as the present work. Analysis of Narrative All of those analyses of Stranger-King narratives mentioned above are structuralist interpretations of narratives concerning the nature of kingship and more specifically why kings are superior over society and how they got to be that way (through actions in the past). They employ different specific methods of structuralist analysis, and I have previously relied on Sahlinss (1985a) own analysis and discussion of Polynesian kingship in combination with Turners (1977, 1985) explication of a structuralist method for the analysis of narrative. The methods of Sahlins and Turner are suitable for the analysis of the legendary history of the Uacusecha Chichimecs because they posit that important contrasts and changes will be internal to a single narrative. That is to say, they incorporate syntagmatic relations (see below, also Chapter 3) into a method that analyzes how meaning is produced and how categories can be transformed through the course of the narrative. Particularly in the work of Sahlins (1985a), it is evident that the transformation of meaning, or the semantic content of the categories, is precisely the point of telling the narrativecategories must be transformed if kings are to be produced. In this section I begin with the more general discussion of the methods of Sahlins and the important concept of elementary categories and of their transformation through relations of encompassment. I then discuss Turners more detailed formulation of a method for the analysis of narratives in both their paradigmatic (i.e. the relations between categories that determines their content) and syntagmatic (i.e. the construction of meaning and content as played out over a sequence) aspects. This discussion also allows me to explain how and why I have chosen to 185

PAGE 186

present the aspects of the legendary history that most fully explain the structure of the narrative and the relation between the king and subordinate lords that it claims, but without getting bogged down in the details. Elementary Categories and Their Transformation In Sahlinss (1985a) analysis of the narratives and rituals surrounding Polynesian forms of divine kingship, he discusses how the characters involved in these dramas are representative of larger categories. These elementary categories are: abstract but fundamental conceptions, represented in persons. The alleged actions of these persons display the right relations between the categories, a process of their combination and organization. (Sahlins, 1985a, p. 89) It is perhaps helpful to think of the actions and stated characteristics of the characters in a narrative as what Fogelson (1989) call epitomizing events. Such events may or may not have happened, and their facticity is not the central issue. What is important is that the event or quality being related by a narrative is included because it establishes what kind of being the character is. That is to say, the events of a narrative establish the content of the elementary category of which the character is a representative. Particularly as any one character interacts with one or many other characters, the events of the interactions display the right relations between the categories, as Sahlins states in the quote above. Such interactions provide a condensed and immediate definition of the contrasts or similarities between the categories and therefore sort out exactly why or in what ways they are different from one another. All of this is not to say that the elementary categories do not change in their significance or semantic contentwhat the characters that represent them are, so to speak. Rather, the structure of the relation between the elementary categories changes, and this change is the point of telling the story. This view is inherent in Sahlinss larger insistence on the importance of the processual nature of structure: 186

PAGE 187

Anthropologists call this [the relations between categories in the quote above] a structure, but the term should not be taken for a synchronic scheme of contrasts and correspondences Just as time and sequence are essential to telling the myth or performing the rite, so too the structure is a generative development of the categories and their relationships. In the event, new and synthetic terms are produced, and elementary categories change their values. (Sahlins, 1985a, p. 89). Rather than view change as a breakdown of the structure (which in Lvi-Straussian formulations should remain static), Sahlins believes that the changes that the terms undergo is an essential aspect of the structure itself. It is these changes that reveal the total or global structure (as Sahlins characterizes it), which static correspondences and contrasts could only partially describe, from a particular point of view or at a particular moment in time (Sahlins, 1985a, p. 103). In other words, events in a narrative not only establish elementary categories and the relations between them, but can transform those categories and their relations. This is the heart of Sahlinss (1985a) analysis of the Stranger-King, in which the foreigners are, at the outset, barbarous (anti-social) outsiders. Through certain actions and interactions with the autochthonous peoples, however, they win the sisters or daughters of the autochthons, appropriate the land, and become kings, the very embodiment of society. Through the course of the narrative, and at certain points within the narrative, the king (or the kingship) embodies both the autochthon and the stranger, the sea people and land people, the male and female, etc., because of his ability to change and transform himself. The problem of identifying meaningful change, the establishment versus the transformation of the elementary categories, brings me to Turners reformulation of the structuralist analysis of narrative. Syntagmatic Relations, and the Recognition of Continuity and Contrast In his reformulation of the structuralist analysis of narratives, Turners (1977, 1985) (re)introduction of syntagmatic relations both outlines the necessity of change in the production 187

PAGE 188

of meaning and the way in which meaningful change can be identified. Going back to the work of the Prague school of linguists (most notably Roman Jakobson), Turner (1977, p. 115) discusses how contrast only becomes structurally meaningful in correlation with continuity. Whether or not a meaningful contrast exists can be determined only if the context surrounding any difference is preserved, as I explain in the example of a farmer discussing his labor schedule in Chapter 4. The preservation of context and concomitant change or substitution of a particular term by definition occurs sequentially, i.e. through time. Therefore the sequence of narrative, and the manner in which the same or different categories are engaged in specific activities or interactions with others, is essential to analyzing the meaning of the categories and ultimately the narrative. The manner in which, in Sahlinss terms, one category appropriates the content of the other category and thereby encompasses that other category, is identifiable only in comparison to the manner in which the elementary category is defined or characterized through epitomizing events in the first place. In Turners method of analysis, the basic unit of a change recognizable in coordination with continuity is the episode (Turner, 1977, 1985). For example, in the myth of the origin of fire among the Kayap of South America, one episode relates the vertical movement of one character from down to up while the character remains in the same place in horizontal space (and a companion remains stationary in both vertical and horizontal space) (Turner, 1985, p. 100). These continuities and contrasts establish the paradigmatic associations, the meaning of the terms in relation to one another. In the same episode of the Kayap myth, the two characters are contrasted with one another precisely because they have been displaced in vertical spacethe displacement indicates a paradigmatic contrast, the fact that the two characters are at this point in the myth different kinds of social beings. 188

PAGE 189

In turn, the episodes of a myth are linked together in sequence by what Turner (1985, p. 102) calls operations. Operations are the aspects of narrative structure that regulate how the actions from one episode are transformed in the next. They determine, in other words, the manner in which a change in one relation is related to a change in another relation in subsequent episodes. It is these operations that move the story forward by regulating the nature of the transformations in sequentially contiguous episodes. One of Turners more important insights is that the operations between episodes and the transformations within episodes are themselves related, such that a series of episodes form a larger paradigm, a definable and separable chunk of the narrative sequence. This is what Turner (1977, p. 131) means by the relativization of syntagmatic and paradigmatic relations; they are interdependent. Syntagmatic relations define the contrasts and continuities by which paradigms are definable, but the content and syntagmatically linked transformations of a group of episodes link them together into a larger paradigm, a larger-scale (macro) episode. For example, a series of episodes that together instigate and then resolve a certain problem might combine to form a macro-episode. In Turners analysis of the Kayap myth, the episodes that together relate the displacement of the character upwards in vertical space and the actions and relations by which he achieves movement back to the ground can be said to constitute a larger macro-episode. This brings me to a point regarding what I present below from my analysis of the legendary history. I discuss those larger scale macro-episodes and the major transformations between macro-episodes that together construct the basic meaning of the narrative, which is the explication of the basis of the relationship between the Uacusecha royal dynasty and the rest of society. I outline the construction of the main elementary categories at play in the narrative by describing the characteristics of the categories and how the actions that the characters, as 189

PAGE 190

representatives of those characters, establish the initial content of those categories. I then go on to describe the major interactions and transformations of the characters and thus the elementary categories of the narrative. I generally focus on the marriages and descent of the characters; marriage is an important form of interaction that brings categories together in order to demonstrate their content as well as the proper manner in which the different categories constitute productive unions once combined. Descent is a way of explaining how and why characters that represent a category, and thus the category itself, either remain unchanged or are transformed in interaction, as I discuss throughout what follows. Finally, certain characters join forces and interact toward common goals; these unions or alliances are further instantiations of the combination of characters and the success or failure of such alliances demonstrates the correct relations between the categories. King as Society and Cosmos: the Production of a Sociocosmic Totality The second part of the RM, a narrative of the history of the Uacusecha royal dynasty, presents an argument, framed in indigenous conceptions of hierarchy that accord well with Dumonts model of hierarchy, for the exclusive legitimacy of the Tarascan king. Through heroic deeds the ancestors of the king established the royal dynasty as a microcosm of the social universe, itself a microcosm of the cosmos as a whole. In order to accomplish this rhetorical feat, the narrative first establishes two opposed elementary categories. In turn, the members of the royal dynasty achieve the encompassment of the two opposed but complementary elementary categories at play in the narrative, and thus they embody a sociocosmic totality. It is from this encompassment and their totalizing nature that they draw legitimacy. Construction of Elementary Categories in the RM As discussed by Dumont and refined by Turner and Sahlins (see Chapter 4), the encompassment of the contrary is a process that results in hierarchy between the two categories. 190

PAGE 191

In order for the encompassing term to embody the sociocosmic totality, the elementary categories of which it is composed must represent opposed but complementary categories that together exhaust the content of the entire field of meaning (Dumont, 1980, pp. 239-245; Sahlins, 1985, p. 89). This is accomplished early in the historical narrative in the RM, as elementary categories that possess cultural significance in all aspects of lived experience, from the mundane to the supernatural, are established. In the narrative, the elementary categories are the opposed but complementary categories labeled Chichimecs and Islanders. Hereafter I place these words in quotation marks only when discussing their semantic content as it is constructed and transformed within and as a result of the narrative, and not when simply indicating that certain characters in the narrative are one or the other. The first of the elementary categories is that of the Chichimecs, specifically the Uacusecha Chichimecs. Chichimec is, of course a loaded word in Mesoamerican usage, having many meanings. With respect specifically to the production of the RM, Stone (2004, pp. 257-258n14) nicely summarizes the various meanings of the word and importantly highlights the fact that it was borrowed from Nahuatl (the dominant language in the Basin of Mexico at the time of Spanish conquest) and imposed by Nahuatl and Spanish translators in the colonial encounter. Mesoamerican scholars are familiar with the characterization of Chichimecs as semi-nomadic (and thus uncivilized) peoples living north of the area of civilized Mesoamerica (e.g. Sahagn, 1950-1969, Book 10). Within the legendary history of the Uacusecha Chichimecs in the RM, the Chichimecs characters were, initially at least, quite similar to Sahagns characterization of Chichimecs, as I discuss shortly. Returning to the legendary history of the royal dynasty, the Chichimecs were foreigners in relation to the Lake Ptzcuaro Basin, the eventual seat of the government of the Tarascan 191

PAGE 192

kingdom. They first appeared at a mountain near Zacapu, to the north of the Lake Ptzcuaro Basin. These Chichimecs worshipped the deity Curicaueri, a masculine solar deity (Corona Nez, 1957, p. 13; Pollard, 1991, p. 170). The Chichimecs subsisted mainly by hunting; in certain passages it is evident that they were unaccustomed to the practice of agriculture or of fishing (both integral parts of the Lake Ptzcuaro Basin subsistence economy) (RM, 1956, pp. 27-28). The narrative also explains that the Uacusecha Chichimecs always had wood for making fires, wood coming primarily from the slopes of the hills that ring the lake (RM, 1956, p. 27). In my complete analysis (Haskell, 2003) of the syntagmatic structure of the narrative, it is clear that the proper place of the Chichimecs was these same mountain slopes and not the flat lands and open waters of the interior of the Lake Ptzcuaro Basin. When separated from the Islanders (see below on the characteristics of the Islanders), for example, the Chichimecs return to higher elevations; this occurs numerous times (Haskell, 2003, pp. 51-58). These core aspects of the initial Chichimec identity relate to the entire spectrum of experiential existence, from their way of life or occupational practice (hunting and being relegated to higher elevations) to religious worship and cosmological identity (worshipping a masculine solar deity). Many of the events that happen to the Chichimecs or which the Chichimec instigate themselves are instantiations of a paradigmthey are epitomizing events (Fogelson, 1989) that establish a proper relationship between this social group and their ecological cum cosmological niche. In contrast, the Islanders, with whom the Uacusecha Chichimecs are continually paired in numerous productive activities (cooking and sharing food, worshipping deities, marriage and biological reproduction), are antithetical, but complementary, to the Chichimecs. I use the collective term Islanders to refer to the many autochthonous groups who occupy the islands of Lake Ptzcuaro because this is the term used in the narrative itself. The Islanders of Xaraquaro 192

PAGE 193

worship a feminine deity. They introduce the Uacusecha Chichimecs to fish as food, and they practice agriculture. The chief priest of the goddess Xaratanga, a goddess associated with agricultural and fishing fertility or success (RM, 1956; Corona Nuez, 1957; Pollard, 1991) was himself a native of Xaraquaro. The Islanders occupied the low islands in Lake Ptzcuaro, and therefore are low and inside in relation to the Chichimecs. In the construction of the Islander category, it is contrasted with the identity of the Chichimecs established in every aspect of lived experience, from the mundane to the supernatural. The entire cosmos is divided into two antithetical categories that represent opposed but complementary social, ecological, and supernatural categories. The Chichimecs and Islanders represent, respectively, both the male and female genders (specifically in the nature of their patron deities); hunting in contrast to fishing and agricultural pursuits; occupation of the higher elevations in contrast to the lower elevations (with the associated relationship between outside and inside or periphery and center); and celestial and terrestrial supernatural significance. Considered jointly, these categories represent the sum of all social, ecological, and supernatural categories relevant to Tarascan society within the Lake Ptzcuaro Basin, itself conceived of as a microcosm of the universe (Pollard, 1993, p. 63). In the next section I discuss how these antithetical but complementary categories are unified within a single, encompassing term. Transformation of the Elementary Categories: the Encompassing Uacusecha Through heroic and even semi-miraculous feats, members of the Uacusecha Chichimecs achieve the encompassment of the Islander category. This encompassment takes place as a process, played out in the course of the narrative, as characters interact with one another. Their interactions both reveal their nature as instantiations of the elementary categories just described and transform the meaning of those categories as the Uacusecha Chichimec category encompasses the Islander category. Figure 6-1 illustrates the place of the most important 193

PAGE 194

characters within the elementary categories involved. The darkest column indicates the Chichimec category in its original content: barbarous and uncivilized hunter-gatherers of low status. The white column indicates the Islander category in its original content: civilized autochthons engaged in agricultural and aquatic pursuits who are higher in status than the Chichimecs. Between the two columns is the gray area, the combination of the two categories that the members of the Uacusecha royal dynasty come to occupy through the course of the narrative. In the diagram, double horizontal lines indicate marriage, while single horizontal lines indicate an alliance or the engagement of the characters so linked in certain pursuits such as governance, raids, or co-residence. Vertical and diagonal lines indicate descent. Where a line of descent results from a marriage symbol, the offspring is explicitly said to be the result of that marriage. Where the lines of descent flow from one character to another, no known spouse is given for the father that would have produced the offspring. The dashed line of descent indicates the likelihood of parentage, but it is not made explicit in the text. The double diagonal lines between Sicuirancha and Uapeani/Pauacume indicate the passing of multiple generations. Finally, within the diagram as a whole and within the Uacusecha column, the relative Chichimecness or Islanderness of the characters is indicated by the black and white lettering. The legendary history begins with the progenitor of the royal dynasty, Hireticatame, arriving at Zacapu, to the north and outside of the Lake Ptzcuaro Basin (see figure 6-2). Hireticatame married a woman from Naranjan, a town also north and outside of the Lake Ptzcuaro Basin. The wifes patron deity is named as Uatzoricuare, which means heat (RM, 1956, p. 20n5). The marriage produced a son, Sicuirancha. Trouble ensued promptly (in narrative time, at least), however. The Naranjans butchered a deer that Hireticatame had shot, which he had explicitly told them not to do, and in the dispute between Hireticatame and his in194

PAGE 195

laws over the deer Hireticatame shot and killed one of his brothers-in-law (in a generic sense, i.e. a man from Naranjan). Hireticatame himself soon became a victim of a revenge killing at the hands of the Naranjans (RM, 1956, pp. 14-22). 195

PAGE 196

Figure 6-1. Characters of the legendary history of the RM and their place in the elementary categories of the narrative. 196

PAGE 197

Figure 6-2. Map showing the locations of important places in the legendary history of the RM. Not to scale. The similarity between Hireticatame and his in-laws the Naranjans with respect to their location (outside the Lake Ptzcuaro Basin) and the similarity of their patron deities (related to burning and heat) reveal that they are not complementary opposites. In other words, they are not 197

PAGE 198

contraries, and the encompassment of one by the other would accomplish nothing on a sociocosmic scale. A large amount of the social and cosmological world would remain unencompassed should this union remain permanent, as becomes evident shortly. The failure of the marriage brought about by the homicidal relations between Hireticatame and his in-laws, however, drives the action of the narrative further so that a proper union, one in which a more totalizing encompassment can be achieved, will be realized. Spurred on by his fathers death, Hireticatames son moved into the Lake Ptzcuaro Basin and settled at Uayameo (RM, 1956, p. 23). Generations passed, and the protagonists (the descendants of Hireticatame) are now explicitly called Chichimecs (RM, 1956, p. 24). A bad omen occurred, and the Chichimecs split into five factions. Four of the factions dispersed in various directions. One of these four factions, which is to become relevant shortly, moved to Coringuaro, to the east of the Lake Ptzcuaro Basin. The fifth faction, that of the Uacusecha Chichimecs (the ancestors of the king proper), led by the brothers Uapeani and Pauacume, traveled along the edge of the lake. They met a fisherman from the island of Xaraquaro in their travels (RM, 1956, pp. 26-27). In time, the younger brother, Pauacume, married the daughter of this fisherman (RM, 1956, p. 31). The Islanders do represent the contrary of the Chichimecs. The subsistence economy of the Islanders was based on fishing and agriculture, as opposed to the hunting economy of the Chichimecs (illustrated succinctly in the unfamiliarity of each group with the food of the other in this encounter [RM, 1956, pp. 27-28]). Additionally, through this marriage, the Chichimec brothers gained a modicum of legitimacy, as they were made priests (sacrificers) by the Islanders of Xaraquaro. The Xaraquaros also give the Uacusecha Chichimecs the associated insignia of office and trappings of nobility: bezotes, earrings, and the so-called tweezers (objects made of 198

PAGE 199

precious metal shaped somewhat like two half-moons connected by a strip of metal doubled over on itself so the half-moons met one another) worn about the neck by priests (RM, 1956, p. 32). Unlike the marriage between the Chichimecs and the Naranjans, this marriage did unify fundamentally opposed categories that carried significance in all aspects of life. In essence, their unification produced a totality, and it was foretold that the united Chichimecs and Islanders will rule over the land (RM, 1956, p. 29). Furthermore, the offspring of this union was the culture hero Tariacuri who, it was foretold, will be lord (RM, 1956, p. 28). He was the one figure most responsible for the creation of the Tarascan kingdom. As the product of a Chichimec father and an Islander mother, he unites the two categories within his person. Through the course of the rest of the narrative this synthesis embodied by Tariacuri is elaborated upon and divided among multiple characters, the correct combination of which produce a unified Tarascan kingdom. 1 To return to the narrative, in time the lord of Coringuaro, (the town populated by one of the factions of Chichimecs) intervened in the affairs of the Uacusecha Chichimecs. He ultimately convinced the lord of Xaraquaro to expel the Uacusecha Chichimecs from that island and remove their bezotes and earrings (RM, 1956, pp. 32-33). The Uacusecha Chichimecs went on 1 While my original analysis was written independently of Stones (2004) discussion of the RM (based largely on her 1994 dissertation), I would be remiss if I did not discuss how she explores the importance of the figure of Tariacuri, although I would contend that at times she fails to fully appreciate why the dual nature of Tariacuri and the combination of Chichimec and Islander categories is so important. At any rate, Stone (2004, pp. 83, 258n18) does note that on the drawing of the royal family tree that is included in the RM, Tariacuri is located at the midpoint of the literal oak tree that forms the backdrop for the family tree, and this position is unique, as most other members of the royal dynasty are located on branches, off to the side of the trunk of the tree. Stone contends, probably correctly, that oaks in general and this oak in particular constitute a world tree connecting the vertically differentiated levels of the universe and located at the center of the world. Tariacuri is therefore explicitly located at the center or midpoint. This, I believe, is due to his nature as a figure that paradigmatically stands between and thus uniting two categories. His feat of shooting the hummingbird and gaining the daughters of Zurumban in marriage (discussed below in the text), exists as the central episode in which the qualities of the Uacusecha and the events of the narrative itself pivot (following the discussion of pivoting as an operation in narrative structure of Turner, 1985, p. 83), in effect mirroring the first half of the narrative and transforming those previous transformations (Haskell, 2003). Interestingly, Pollard (1993, p. 57) includes the word teruhuacuri, which translates as the one in the middle in a list of kinship terms that is meant to demonstrate the importance of the elder/younger distinction among siblings. The similarity between this word and the name of Tariacuri is likely more than a coincidence; his name is perhaps a pun or play on words, and is likely further proof of his embodiment of two categories. 199

PAGE 200

to discover a place in Ptzcuaro that had been foretold to them, and they built their temples and settled there. The correctness of the elementary categories involved in the marital union of the Islanders and Uacusecha Chichimecs is indicated by the contrast drawn with Hireticatames marriage to the woman from Naranjan. That marriage fell apart due to strife internal to the relationship; the Naranjans butchered the deer that Hireticatame had shot, something which he had explicitly told them not to do. The marriage between the two Uacusecha leaders and the Islanders of Xaraquaro disintegrates only upon the insistence of the Coringuaros. The Coringuaros interventions in the affairs of others appear throughout the narrative and serve to drive the action forward when an ostensibly proper union of elementary categories, but one lacking in some specific aspect of their combination, is achieved (Haskell, 2003, pp. 53-54). Following the intervention of the Coringuaros, it is evident in the actions of the Uacusecha Chichimecs that the duality achieved by the Uacusecha Chichimecs in their union with the Islanders and embodied singularly in the character of Tariacuri is of only limited efficacy. Tariacuris Islander qualities are manifested only in relation to his own Chichimec cousins Zetaco and Aramen (Haskell, 2003, pp. 59-60). The parentage of Zetaco and Aramen is not explicitly given in the narrative, and they appear to be reproductions of their Chichimec father but without any Islander ancestry. The Chichimecness of Zetaco and Aramen is explicitly marked in the RM, as the priest of Zurumban, whom they have deceived into a trap, remarked that only Chichimecs deceive so (RM, 1956, pp. 56). Tariacuris Chichimec identity in relation to the Islanders was manifested when Tariacuri embarked on a program of marauding and generally causing havoc for the people of the lake (for example, when he lights fires on the hilltops and isolates the Islanders of Xaraquaro; RM, 1956, pp. 45-46). Tariacuris identity as a Chichimec in relation to the Islanders renders this initial duality within the Uacusecha 200

PAGE 201

Chichimec lineage as one of limited efficacy and scale, unable to transform and thus encompass the wider Chichimec and Islander categories in their relation to one another. Tariacuri plays the role of Islander counterpart to his Chichimec cousins, but in interactions with the wider social universe, including the Islanders, he was relegated to playing the role of Chichimec. If he and the Uacusecha were to appropriate the content of the Islander category and encompass it, they should have been able to act as Islanders do within that larger social universe. After Tariacuri has become an adult and following some further disputes with the Coringuaros, the lord of Coringuaro proposed that Tariacuri marry his daughter. Tariacuri agreed to the marriage, but it is apparent that this match was ill-fated from the start, although it does produce a son named Curatame (RM, 1956, p. 66). Soon after Tariacuris wife began to live with him, she committed adultery with her own categorical kinsmen, i.e. men from Coringuaro and a village allied with Coringuaro (Itziparamucu: see RM, 1956, p. 138). Tariacuri was extremely distraught and nearly drove himself to madness roaming the mountains day and night for firewood (RM, 1956, pp. 64-75). Thus in his actions there was an intensifying Chichimecness brought about by the failure of his marriage relation with the former Chichimecs the Coringuaros. Upon examination of the elementary categories involved in this latest union, it is apparent why it should fail. It is not composed of the categories necessary to produce a synthesis of contraries, which, as the narrative has just indicated, are the Chichimecs and Islanders. The Coringuaros are former Chichimecs. Their patron deity is associated with the morning star and is a brother of Curicaueri, the patron deity of the Uacusecha Chichimecs (Corona Nez, 1957, p. 32; Pollard, 1991, p. 171). Coringuaro is not located within the Lake Ptzcuaro Basin but outside (to the east). Therefore both groups are male, outside (in origins if not also in actuality), 201

PAGE 202

worship celestial patron deities, and Chichimecs. In these ways, the two categories are too similar to be contraries; certainly they do not produce the kind of sociocosmic totality achieved with the union between the Chichimecs and Islanders. Furthermore, the offspring of this union (Curatame) was explicitly said to lack the legitimacy required to rule, as Tariacuri declared that he was an upstart from Coringuaro (RM, 1956, p. 102). Similar to Hireticatames marriage to the woman from Naranjan, this union fell apart due to its own internal incompatibilities that resulted from the characters being essentially similar. But again, the events involving the Coringuaros drove the action of the narrative forward and out of this strife a new union was forged. In his distress, Tariacuri sought the help of Zurumban. Zurumban was the chief priest of the goddess Xaratanga, a feminine deity related to agricultural fertility as well as the moon (Corona Nez, 1957, p. 75; Pollard, 1991, p. 170). In fact, he was hand picked by that goddess to serve as her priest (RM, 1956, p. 48). He was also a native of Xaraquaro, indicating that he was categorically an Islander (RM, 1956, p. 48). He therefore represented femininity and agricultural production. In the interaction between Tariacuri and Zurumban, the latter challenged the former to shoot a hummingbird with an arrow. Tariacuri successfully answered the challenge and his skill in shooting caused Zurumban to remark to Tariacuri: certainly you are a Chichimec, for this bird is not big. Was it a task to shoot, as it is so small? How can anyone best you? You never miss and there is no one who can best you at shooting (RM, 1956, p. 76, my translation). Through this feat, Tariacuri gained the daughters of Zurumban in marriage (RM, 1956, pp. 74-81). This marriage essentially reproduced the union between the Chichimecs and the Islanders of Xaraquaro; it therefore also reproduced the synthesis of contraries represented by the 202

PAGE 203

Chichimec and Islander categories. Most importantly, through this marriage the Chichimecs gained the women of the Islanders as well as their (feminine) wealth, in the form of the dowry of the daughters. Because this union and the appropriation remained permanent, the Uacusecha maintained the ability to reproduce the duality that was heretofore embodied solely by Tariacuri, as the narrative demonstrates in the events that follow. In contrast to the marriage between the Uacusecha and the Islanders of Xaracuaro, however, the Uacusecha were in a position of power in this new marriage with the Islanders. The Uacusecha had been priests on the island of Xaracuaro and therefore subservient to the Islander lords of that place. In this latest iteration of the Chichimec and Islander duality, in contrast, Tariacuri has gained the upper hand. Any potential advantage enjoyed by Zurumban over Tariacuri as wife-giver is explicitly negated in the RM; in a speech admonishing Zurumban to quit getting drunk and to go on raids and thus gather wealth for the goddess Xaratanga, Tariacuri stated that he was making Zurumban a lord (RM, 1956, p. 81; even if, ironically, through the winning of the daughters and wealth of Zurumban, Tariacuri was the one who was becoming a lord). With the superiority of Tariacuri marked rhetorically in this passage, and marked as much with moral superiority and piety (directing Zurumban to gather wealth for the goddess Xaratanga) as political supremacy, the ascendance of the Uacusecha is apparent. This is the final marriage recorded in the priests narrative and this fact demonstrates the importance and the productivity of the totality that has been produced. The elementary categories involved are the correct categories, and are combined in a manner in which the Uacusecha are dominant. In fact, the appropriation of the Islander women and wealth, and their preservation by the Uacusecha Chichimecs, indicates that the Uacusecha Chichimecs have appropriated the content of the Islander category. They now embody both categories and can act 203

PAGE 204

as both Islanders and Chichimecs. In other words, the Chichimec category has encompassed the Islander category, in Dumontian terms, and thus represents the sociocosmic totality. In the process, the Uacusecha Chichimec category has become something more than what it was previously; it has been transformed from a subset of the whole sociocosmic universe into an embodiment of that totality. The second half of the narrative is largely an exercise in preserving this duality that the Uacusecha (namely Tariacuri) has achieved. At the same time, the Chichimec/Islander duality embodied solely by Tariacuri is elaborated upon in order to produce a new Uacusecha triumvirate that, ultimately, conquers the land and establishes the Tarascan kingdom. The first demonstration of the preservation of the newfound wealth and duality of the Uacusecha occurred when the lord of Coringuaro demanded that Tariacuri give him riches that he had taken in raids (RM, 1956, pp. 86-89). In reply, Tariacuri gave the emissaries from Coringuaro arrows tipped with stone points of various colors, saying that the green tipped arrows are the green feathers that the lord of Coringuaro had demanded, and so on. The emissaries took the arrows back to Coringuaro and the sons of the lord, who were running the daily operations of the lordship in place of their aging father, broke the arrows and cursed Tariacuri for the ruse he played in switching the points for the riches. When the elderly lord of Coringuaro was informed of his sons actions, he chastised his sons for breaking the arrows, saying that they might have held some divinity (RM, 1956, p. 89). In this negated exchange (or extortion), the preservation of both the Islander wealth and Chichimec hunting/warrior identity of the Uacusecha are preserved. Not only did the Coringuaros fail to gain the riches (symbolic of the feminine wealth of the land, and syntagmatically related to the feminine wealth the Uacusecha had just previously acquired through the marriage to Zurumbans daughters), they had initially set out to 204

PAGE 205

extort, but they also failed to appropriate the arrows (symbolic of the prowess of the Uacusecha in hunting and war) that Tariacuri had given them instead. 2 Following the interaction with the Coringuaros, Tariacuri settled in Ptzcuaro and almost immediately began to search for his long lost nephews, the sons of his cousins Zetaco and Aramen. These nephews, Hiripan and Tangaxoan, had not been discussed at all following the death of their fathers (RM, 1956, p. 91). The impression is that the nephews were the victims of the same fate that had befallen their fathers, who were murdered trying to escape up mountain slopes following various tricks and misdeeds against the Islanders (RM, 1956, pp. 60-63). The parentage of Hiripan and Tangaxoan is not explicitly stated and the narrative defies attempts to delineate the mother and father of each of these characters. For example, neither Hiripan nor Tangaxoan are explicitly linked to either Zetaco or Aramen in a father-son relationship. It might seem that one was the son of either Zetaco or Aramen and the other youth was the son of the other, yet the two youths appear to have the same mother (RM, 1956, p. 91). Rather than view this as problematic, however, I believe the obfuscation is really a device to imply that Hiripan and Tangaxoan should simply be regarded as reproductions of Zetaco and Aramen, their prototypically Chichimec fathers. The Chichimecness of Hiripan and Tangaxoan is confirmed in the course of the narrative, which picks up the trials and tribulations of the two youths as they journeyed with their mother. They are presented as impoverished orphans, who were forced to beg for food scraps in the marketplace of Azajo (RM, 1956, pp. 91-92). Eventually they returned to the Ptzcuaro Basin, and their mother suggested that they live with her brother, the lord of Erongarcuaro. In return for room and board, the youths promised to provide service to the lord (RM, 1956, p. 94). They 2 In numerous passages in the RM (1956, pp. 39, 50, 65), Uacusecha characters are making arrows when encountered by others. 205

PAGE 206

neglected their obligations, however, and were constantly roaming the mountains for firewood for the temple fires. The lord of Erongarcuaro complained that Hiripan and Tangaxoan were crazy Chichimecs and that they had not performed the services they had promised, and therefore he expelled them from his house (RM, 1956, p. 95). The mother of the youths suggested they go to Urichu; the lord there was also her brother. The same sequence of events that transpired at Erongarcuaro was repeated at Urichu, and so the youths and their mother went on to Pareo (RM, 1956, p. 96). There they promised the same services to the lord, but the lord welcomed them and told them that they will perform no such service; they were Chichimec lords and should take wood to the temples, as they had done in Erongarcuaro and Urichu (RM, 1956, p. 97). Eventually Tariacuris advisors encounter the youths at Pareo and bring them to be with their uncle Tariacuri in Ptzcuaro (RM 1956:97-99). The Chichimecness of the youths is made clear through their actions, in both their religious piety and their constant roaming of the mountains. Following their reunion with Tariacuri, their uncle helped to establish Hiripan and Tangaxoan at what was, or would soon become, Ihuatzio (RM, 1956, p. 100). With their uncles blessing and encouragement, the youths started going on raids and they generally prospered at their new settlement. Through their actions, Hiripan and Tangaxoan became the Chichimec counterparts to Tariacuris new, predominantly Islander identity, represented by his largely advisory role and the simple fact that he was not the one who goes on raids anymore because now he had his Chichimec nephews to fill that role. 3 As discussed by Sahlins (1985a, p. 91) in his analyses of Polynesian divine kingship and Stranger-King narratives, once the marauding and barbarous stranger has been domesticated and transformed into the king, these behaviors 3 Curatame also appears unable to fill the role of the raiding and marauding Chichimec, as he spends much of his time getting drunk on pulque (RM, 1956, p. 100). 206

PAGE 207

devolve to junior members of the kings lineage. The new king must shed these behaviors because they are inappropriate actions of a king; their devolution to the junior members of his lineage, however, preserves the duality of the divine kingship within the royal family. This is precisely what was happening in Tariacuris search, recovery, and tutelage of his nephews. The Chichimec/Islander duality was reproduced in another combination when Hiripan and Tangaxoan captured the Islander Zapiuatame. Rather than sacrificing the Islander, however, Zapiuatame joined Hiripan and Tangaxoan, and with their newfound Islander companion the Uacusecha Chichimecs enjoyed much success and prosperity (RM, 1956, pp. 116-121). With the addition of an Islander to the duo of Hiripan and Tangaxoan, the Chichimec/Islander duality established in the person of Tariacuri was reproduced in the descending generation. However, the status of Zapiuatame as an Islander who is an outsider in relation to the Uacusecha threatens the exclusivity of that duality within the Uacusecha. It was, in other words, a problem of the right elementary categories but in the wrong relation to the distinction now being drawn between those inside and outside of the Uacusecha royal dynasty. The inclusion of an Islander from outside of the Uacusecha would threaten the monopoly of power that the Uacusecha were working toward. Furthermore, if the Uacusecha truly was a dually constituted category that included both Chichimecs and Islanders, they should not have needed the aid and support of an outsider Islander. This problem of the right categories but in the wrong relation to the Uacusecha was resolved through, and brought into direct contrast with, Tariacuris suggestion that his son Hiquingaje join Hiripan and Tangaxoan in Ihuatzio (RM, 1956, p. 124). Hiquingaje appeared out of nowhere; he was mentioned only one time previously, when Tariacuri foretold that Hiripan, Tangaxoan, and Hiquingaje would be lords (RM, 1956, p. 107). He is not said to be the 207

PAGE 208

son of Tariacuri by any particular woman, in contrast to the explicit disclosure of Curatames parentage following Tariacuris marriage to the woman from Coringuaro (RM, 1956, p. 66). Given that Hiquingaje was not mentioned following the marriage to the woman from Coringuaro, it is entirely possible, if not implied, that Hiquingaje was the product of Tariacuris marriage to Zurumbans daughters. Therefore he was likely the son of Tariacuri, a dual Chichimec/Islander character, and the Islander daughters of Zurumban. Hiquingajes Islanderness was marked in contrast to the Chichimecness of his cousins Hiripan and Tangaxoan as the three young lords did penance in a cave. While Hiripan and Tangaxoan ate raw plants, they gave Hiquingaje toasted maize to eat (RM, 1956, p. 124). The distinction in the foods eaten by these characters set Hiquingaje apart from his Chichimec cousins. In time, the three young Uacusecha lords established themselves in Ihuatzio. Tariacuri advised them and sometimes harshly instructed them in the proper worship of the gods. The three young lords rise to greatness was instigated by Tariacuris suggestion that they take a fish to his brother-in-law Hivacha, the lord of Zirahuen (Zirahuen is also the name of a small lake south-southwest of Lake Ptzcuaro by which the town of Zirahuen is located). The three did as instructed, but they were insulted by Hivacha when they were not given any food to eat or pulque to drink. The three were enraged at this affront and vowed their revenge on Hivacha. They left, but before they had gotten too far away, the majordomo of Hivacha acknowledged that Hivacha had treated them badly, and he offered them a number of precious feathers. The majordomo asked that they remember him and his family in the event of an attack upon his town. After this exchange Hiripan climbed a tree, but the dead and rotting limbs of the tree could not support him and he fell to earth. Tangaxoan ran around in a craze, fearing that his brother was dead. In a final demonstration of the self-reproducing nature of the Uacusecha duality, Tangaxoan 208

PAGE 209

(categorically Chichimec) and Hiquingaje (categorically Islander) revived their relative through their joint efforts (RM, 1956, pp. 143-147). Having revived Hiripan, the three young Uacusecha lords returned to Ptzcuaro and informed Tariacuri what had happened and that they wished to go to war against Hivacha. Tariacuri listened to them, and then instructed them to meet him the following morning. The next morning Tariacuri told the three that they would be lords in Tzintzuntzan, Ihuatzio, and Ptzcuaro, and that from these towns they would rule over the land. He then told them that he had arranged for the help of Erongarcuaro, Urichu, Pechtaro, and Cumachen in the fighting. With the help of these allies they encircled Hivacha and defeated him. Following the attack Hivacha was sacrificed, but the majordomo reminded the Uacusecha war party of his gift to them, and they remembered him and spared his life (RM, 1956, pp. 147-150). Following the defeat of Hivacha the army went on to conquer numerous towns throughout Michoacn. During these battles, Tariacuri died (RM, 1956, pp. 150-152). Following the conquests, the people of the towns were unsettled and they abandoned the towns, sometimes carrying off the riches of the towns. In order to prevent this chaos (and reclaim the riches that were rightfully theirs), the three lords instituted local offices of leadership. With local lords in place, the people could be pacified, settled, and productive. The gods were provided for, and order reigned (RM, 1956, pp. 152-153). Not only does the story of the origins of the Tarascan kingdom establish the basis of the local offices of leadership with the Uacusecha (thereby cosmologically authenticating their role in constituting the abilities of those who are to hold these offices), but the fundamental exchange between the state and the populace is established through the parable of Hivachas majordomo. In return for their lives (i.e. not being 209

PAGE 210

sacrificed following their defeat in war), the people of the kingdom must sacrifice their wealth (and, judging from the information contained in other parts of the RM, their labor) to the state. Through the course of the legendary history of the Uacusecha Chichimecs, the Uacusecha are transformed from barbarous, hunting outsiders to the kings of a unified kingdom. The Uacusecha conquests of most of Michoacn are the syntagmatically constructed result of the other actions of the narrative. The most important of these actions is the Uacusechas encompassment of the Islander category. Once the duality that the Uacusecha contains within itself was distributed among Chichimec members and Islander members, with no outsiders, the Uacusecha members embark upon a successful program of conquest. Subordinate Lords as Chichimecs: The Replication of Legitimate Authority Using the preceding analysis of the sequence and transformations within the historical narrative of the royal dynasty, the specific nature of the production of hierarchy in the relation between the elementary categories can be discussed. Here I analyze how the Uacusecha Chichimecs are constructed as the encompassing category and furthermore how this encompassing status is rhetorically related to the specific encompassment of the subordinate nobility elsewhere in the RM. Encompassment of Elementary Categories by the Uacusecha Through the course of the priests narrative the Chichimec ancestors of the king accomplish what Dumont calls the encompassment of the contrary: they subsume into their own identity the identity of the Islanders. Their identity as Chichimecs at the beginning of the narrative makes use of actual Chichimecs as uncivilized hunter-gatherers from the north to anchor the identity of the royal dynasty, providing a referent that all listeners will be familiar with. This conflation, combined with the paradigmatic actions of the early ancestors of the king as hunters, constitutes these Chichimecs as one of the elementary categories. Following 210

PAGE 211

Tariacuris marriage to Zurumbans daughters, the Uacusecha gain the ability to act as Islanders, as insurers of fertility and keepers of wealth. Additionally, this marriage endowed the Uacusecha with the ability to reproduce the duality that they have achieved. The encompassment of the Islander category by the Chichimecs that is chronicled in the priests narrative establishes the Uacusecha Chichimecs as that category which takes the higher value in Dumonts terms and produces the synthetic third category, that of the sovereign, in Sahlinss (1985a, p. 90) discussion of Polynesian forms of kingship. Figure 6-3a diagrams the initial opposition of the Chichimecs to the Islanders. Figure 6-3b diagrams the ability of the Uacusecha Chichimec category to encompass, and subsume within itself both the Islander and Chichimec categoriesin other words the two elementary categories that together constitute the sociocosmic totality. This is the logic behind the royal dynastys claims of Chichimec origins: that due to the claim that their ancestors were actual Chichimecs (i.e. barbarian hunters), they had the ability to encompass their contrary, which was specifically accomplished through their hunting prowess. The Uacusechas ability to recreate society and place themselves atop it through hunting, i.e., warfare instantiated as the hunting of men (the metaphorical links between hunting and war are made explicit at the kings coronation [RM, 1956, p. 53]), is a transformation, and logical extension, of that encompassment of elementary categories and the hierarchy it produces. The Uacusechas hunting prowess enabled them to take a higher value than their counterparts the Islanders through their encompassment of the Islander category. This new status structurally demanded, moreover, that this new status be evidenced, or made manifest, in the world of social relations. Therefore the Uacusecha Chichimecs took up arms and recreated society, transforming it into one with a centralized and legitimate authority at the 211

PAGE 212

top of society (see figure 6-3c), which they have both created and embody as a result of their superior valuation. Figure 6-3. Transformations of the meanings and relations of the elementary categories of the legendary history of the RM. As a result of their encompassing superiority and their place atop society, the meanings of the elementary categories are transformed. There are now, therefore, multiple Chichimecs. In addition to the Chichimecs as Chichimecs (i.e. the early ancestors of the king who are paradigmatically equivalent to the Chichimecs that exist beyond the boundaries of Mesoamerican civilization, as in figure 6-3a), there are now Chichimecs as kings, what I label Chichimecs (Prime, following the convention in geometry) (as in figures 6-3b and 6-3c). These are the Chichimecs who have transformed themselves from low status outsiders into kings 212

PAGE 213

through exploits paradigmatic of a Chichimec: hunting and war (figure 6-3c). In the process, the Chichimec category is itself transformed. It attains connotations of high status, actually of the highest status in the society as a wholekings are Chichimecs (Chichimecs, to be precise). With the advent of hierarchy (established by the encompassing actions of the kings ancestors, as in figure 6-3b), the opposition of Chichimec and Islander takes on an additional dimension: kings are Chichimecs, and Islanders are unfit to be lords (RM, 1956, p. 102). Position of the Subordinate Lords In association with, and as a result of, the legitimacy of the Uacusecha royal dynasty due to their status as the encompassing category, is the hierarchy that is created when the royal dynasty institutes offices of local leadership. According to the legendary history of the Uacusecha in the RM, in the wake of the conquests of the Uacusecha triumvirate of Hiripan, Tangaxoan, and Hiquingaje (under the counsel of Tariacuri), there was widespread chaos in conquered towns. The populace had run away from the conquered towns, and had taken the precious metals, stones, and other riches with them. The Uacusecha leaders Hiripan, Tangaxoan, and Hiquingaje recovered these riches from the people and ordered them back to their towns, to live as they had before, tending to the fields. The people complained that they could not settle, however, because they had no lords or leaders. To settle the people, Hiripan, Tangaxoan and Hiquingaje made lords in all of the conquered towns so that the people would be settled, and so they could live as before (RM, 1956, p. 153). Subordinate lords therefore stand as the representatives (replicas) of the royal dynasty; this was the intent of the Uacusecha leaders who instituted the offices in the first place. In the priests same conclusion to his narration (i.e., as he completes the narrative to be written down in the RM), the priest chides those in attendance, asking where more Chichimecs will come from 213

PAGE 214

(RM, 1956, p. 155). The implication is that those in attendance are not the Chichimecs they should be; they have failed the king and failed Curicaueri, and now there is a need for more Chichimecs, as the real Chichimecs who have been faithful in their duties are off at the wars. The responsibilities of the lords also identify them as Chichimecs. The lords are responsible for ensuring that the lands dedicated to Curicaueri and to the wars are planted and harvested, and they are also responsible for ensuring that firewood is brought to the temples for Curicaueri. By doing these things, the lords continue the service of Curicaueri that had been the primary responsibility of the Uacusecha, but now with the advent of newly created offices of leadership (discussed below) the lesser Chichimecs, the caciques, take a role in these responsibilities. The development of the structural opposition between Chichimecs and Islanders therefore continues, and builds upon the connotations of high status (i.e. kingliness, or the possession of legitimate authority) that the Uacusecha Chichimecs have achieved. This further progression manifests itself when the position of the lesser nobility, or those nobles not themselves members of the royal dynasty, is examined. In the same passage in which the priest admonishes the lords for not being the Chichimecs they should be, he also derisively implies that they are Islanders, saying that they are from a god of the Islanders (RM, 1956, pp. 154-155). In this way, the lesser nobility joins the royal dynasty as Chichimecs in their opposition to the commoners, who are Islanders in a metaphorical sense. Figure 6-3d shows graphically how the lower nobility have joined the royal dynasty as Chichimecs in opposition to the commoners, as indicated by the lines and shading of the three categories (the two categories of Chichimec nobility and the one category of Islander commoners). 214

PAGE 215

The royal dynasty institutes these offices of local leadership following the conquests that created a unified kingdom. As a result of these conquests and the legitimacy established by the encompassment achieved by the Uacusecha, legitimacy is the sole property of the Uacusecha, and it is given out to the holders of these local offices. (The exchanges that signify this relationship in the ongoing social relations that constituted the Tarascan kingdom are discussed in the next chapter.) The subordinate lords did not achieve legitimacy via their own encompassing actions, but they come to possess it through their relationships with the Uacusecha royal dynasty. As a result of that relationship, they are transformed into high status individuals with the legitimacy to rule their towns, but not the ability to be kings. They are high status individuals and therefore Chichimecs, but not Chichimecs in the same way that the Uacsuecha royal dynasty is Chichimec; they are not Chichimec by origin as the Uacusecha claim to be, nor are they encompassing, kingly Chichimecs (Chichimecs prime) as the Uacusecha claims to be. To indicate the similarity and difference being established, I call the subordinate lords Chichimecs (double prime) in order to indicate their status as one that is derived from the royal dynasty and the original Chichimecness of the royal dynasty. The subordinate lords are derivative reflections of the royal dynasty, similar in some aspects, but not the real thing. On Chichimecs and Transformative Capacity Therefore, according to the manner in which the relationship between the royal dynasty and subordinate lords is presented in the RM, there is a fundamental division between the king (or the royal dynasty) and all other nobles. The lords of Tzintzuntzan, Ihuatzio, and Ptzcuaro were members of the Uacusecha; they had inherited their positions from Hiripan, Tangaxoan, and Hiquingaje (if not in a strict line of succession then by right of being descended from some 215

PAGE 216

Uacusecha founder). These leaders had the legitimacy to rule by the fact that they were members of the category (the Uacusecha Chichimecs) that embodied the sociocosmic totality through their historical encompassment of their contrary (the Islanders). The lords of subordinate towns could claim no ancestry of legitimate lordship independent from the actions of the Uacusecha because legitimate lordship originated with the Uacusecha. The claim by the royal dynasty to have instituted local offices of lordship is at the same time a claim to have instituted the very conditions through which the lesser nobility can become Chichimecs (i.e. legitimate lords). Without the Uacusecha this transformation would have been impossible. To make the point explicitly, the narrative is the cosmological authentication (following Weiner, 1992; see Chapter 5) of the encompassment of the lesser nobility because it is the Uacusecha that, due to the actions of the Uacusecha ancestors in the past, exists as the principle by which the lesser nobility is reproduced as replications of the Uacusecha (following Turners examination of relations of encompassment; see Chapter 4). This replication is made rhetorically explicit through the extension of the name Chichimec to the member of the lower nobility by the royal dynasty through the narrative. In sum, what the legendary history reveals Chichimecness to be is a principle of transformation itself. It is the innate abilities of the Uacusecha Chichimecs as warriors and hunters that allows (or compels) the surface events as transformations to take place. The first target of the Chichimecs transformative abilities is themselvesthey transform themselves into holders of exclusive legitimacy by virtue of their own categorical transformation, i.e. their encompassment of the Islanders. The proposition that the Uacusecha Chichimecs ought to be a transformative principle reveals further why the initial union of the Uacusecha Chichimecs and the Islanders of Xaraquaro is ultimately doomed to failure, even if this union does produce 216

PAGE 217

Tariacuri, the first dual or synthetic being and the one who is more responsible for the transformation of the Uacusecha Chichimec category than any other. This first union is proposed by the Xaraquaros, and they are the ones who attempt to transform the Uacusecha Chichimecs into priests. The status of Uacusecha protagonists as subjects of an Islander transformative capacity simply cannot pertain, and the union is torn asunder by the efforts of the Coringuaros. Tariacuris actions as a true Chichimec when he shoots the hummingbird and wins the daughters of Zurumban reveal the proper relation of the Uacusecha Chichimecs and the truly transformative capacity of their innate powers. This transformative capacity is further displayed in the conquests that established the Tarascan kingdom and their aftermath. The Uacusecha conquered towns and transformed autonomous peoples into subjects (rather than transforming them into sacrificial victims). Finally, the Uacusecha embarked on a program that transformed certain subjects into lords in an effort to pacify the kingdom and consolidate power. Conclusion: The Ideological Construction of the Relationship between Subordinate Lords and the King As represented in the RM, the relationship between the king/royal dynasty and subordinate lords (indeed, all of society) is one of encompassment. Through the establishment of elementary categories, the narrative of the legendary history of the Uacusecha royal dynasty divides the social world into two opposed categories, the Chichimecs and the Islanders. Only the Uacusecha Chichimecs achieve the encompassment of both categories, and therefore only they have legitimacy required to rule. By definition, all other lords and social categories, subsumed under these elementary categories (whether Islanders or non-Uacusecha Chichimecs) are encompassed by the Uacusecha Chichimecs (and by extension the king). Thus the hierarchical relationship of encompassment is established rhetorically as a direct relationship between the Uacusecha royal dynasty and all other (lesser) lords. As represented in the RM, through the constitution of the 217

PAGE 218

lower nobles as Chichimecs, these lower nobles are confirmed as inferior tokens of the prototypical embodiment of legitimate authority, the king. They are inferior because they have not achieved the legitimate authority, through the actions of the encompassment of the contrary, that the king and his ancestors achieved, but have merely been transformed into possessors of that legitimate authority by the actions of the king and his ancestors. Finally, the RM claims that the same narrative was told in the capital of Tzintzuntzan at a yearly festival by the chief priest, and that traveling priests (presumably the curitiecha) told the same narrative in all of the towns of the realm. While it is an open question whether or not the narrative contained within the RM is the same as a fixed prehispanic version, or if any fixed version ever existed, the narrative contained in the RM does contain the basic themes of the encompassment of elementary categories and the origins of offices of local leadership that I believe would have featured prominently in any prehispanic version, as they were key elements in the rhetoric of the encompassment of the lesser nobility and society in general. The proclamation of these relations would have gone a long way toward rhetorically framing the relations of encompassment between the king and subordinate lords. Moreover, this practice of framing the historically constituted relations between the royal dynasty and subordinate lords was realized in other exchanges as well, as I discuss in the next chapter. 218

PAGE 219

CHAPTER 7 REPRESENTATIONS OF EXCHANGE AND THE PRACTICE OF ENCOMPASSMENT IN THE RELACIN DE MICHOACN Introduction The representations of hierarchy and the encompassment of the local level lords described in Chapter 6 largely frame the relations of the present in terms of the past. While the relation of encompassment was cosmologically authenticated at yearly festivals if not more often, the king likely would have sought additional ways to perpetuate the relationships of subordination upon which the state depended. Furthermore, cosmological authentication is, by its nature, about the past, and could have been subject to empirical disjunctions and slippages of meaning (Keane, 1997; see Chapter 4) between the world of the present and the remembered past. In this chapter I examine representations in the RM of those exchanges that would have established a direct relationship of subordination and ultimately of encompassment. The exchange relationships represented in the RM are, in a sense, doubly framed. They are representations in the early colonial era of how such exchanges were realized in the precolonial era. Therefore they are framed representations of how the exchange relationships were themselves framed by state agents in the precolonial erathey purport to be accurate recordings of what those participants involved in exchange acts in the past said or would have said or should have said. This double filter presents some advantages to investigating the framing of exchange relations in the prehispanic era. As colonial representations, they were attempts to explain the nature and extent of the power of the king to the Spaniards, and perhaps even to reiterate that power to indigenous members of society. As such, the representations within the representationsthe speeches made by the participantsshould have been constructed in order to satisfy the same end goal (the explication of a relation of encompassment to communities of witnesses in the prehispanic past) that was the very same goal the colonial era authors of the 219

PAGE 220

document (i.e. the indigenous members of the upper nobility whose testimony formed the bulk of the document) were trying to achieve. In these representations, the king constituted the capacity of the subordinate lords to fulfill their obligations as lords through certain exchanges of the objects and persons integral to that capacity. The framing of exchange relationships as incomplete, asymmetrical, and/or unreciprocated would have been one way to actively construct relations of encompassment by proclaiming the preservation of an indexical relationship between giver and gift. This preservation would have led others to abduct the role of the giver in the actions of the recipient, thereby encompassing the recipients agency. By focusing on exchange practices, the practical realization of encompassment is brought into view, against the backdrop of the cosmological authentication of the relationship between the king and subordinate lords discussed in Chapter 6. Framing Noble Marriage Exchanges in the RM The RM contains descriptions of multiple marriage ceremonies, involving members of different classes of indigenous society. These descriptions were provided to the Franciscan friar that produced the document and claim to represent typical marriage ceremonies, including the rhetorical speeches made by the priests that officiated at the ceremonies as well as members of the wedding party (including the groom and members of the grooms as well as the brides families). Within the speeches claimed to have been made by the participants in a marriage involving a subordinate lord and a woman from the kings palace, there are passages that indicate that the relation between the subordinate lord, the woman, and the king was being constructed as a relation in which the king encompassed the subordinate lord. In the representations of such speeches, the king emerged as the principle by which the subordinate lord was able to fulfill some of his most important duties as a lord. As stated above, I assume that while the priests that actually officiated at such marriages in the past might not have spoken the exact words recorded 220

PAGE 221

in the RM in the early Colonial period, their speeches should have sought to construct the kings encompassing nature by framing the exchange according to the same logic of encompassment through exchange that is explicated in the representations of the RM. Before I present the data concerning how such marriages are rhetorically framed in the RM, and therefore what the effects of framing in the prehispanic era should have been, I discuss the marriage relation between the king and subordinate lords in general, in order to provide the necessary background. On an additional note, the analysis of the relationships between the king and subordinate lords that were forged through marriage presented here is admittedly androcentric. In many ways this androcentrism is the result of the evidence in the RM itself. For example, the names of only five women are given in the entire text of the document. All other women are simply described as the mother, wife, or daughter of a named male, and so women appear important only in their relationships to men. With regard to the description of marriage relations between the king and subordinate lords, the males in these relations are the actors, while the women are the passive objects of the mens actions. I acknowledge that royal and noble women likely played important roles in the prehispanic Tarascan kingdom not only as brokers between the king and subordinate lords but also as agents in their own right who, by exercising their will, likely shaped the course of royal/noble relations. The example of Doa Beatriz de Castilleja, a woman descended from members of the royal dynasty who married three Spaniards largely to pursue her own agenda in the early colonial period was a paragon of the pursuit of self-interest (see Kuthy Saenger [1996] on the life of Doa Beatriz), at once throws the notion that women were, or were always, passive objects of masculine agency into doubt. What follows is not intended as a corrective of the male-dominated viewpoint presented in the RM, but an attempt to understand the underlying logic of the manner in which a class of events is represented and should have 221

PAGE 222

occurred in the minds of the male upper nobility. In order to analyze how the representation in the RM of marriages between the king and subordinate lords reveals a conceptualization of the encompassment of the subordinate lords, therefore, I examine marriage relations as they are represented, i.e., as relations between men that involve women. Brides of Nobles and Wives of Curicaueri: Royal/Noble Intermarriage The RM (1956, pp. 208-210) states that the king married off some of his daughters or sisters to subordinate lords (see figure 7-1). In the context of the description of the kings wives it is said that as a result of having so many wives, he had many children and female relatives, and the RM states that he married some of these women to noblemen. In the chapters concerning marriage practices among the lords (RM, 1956, pp. 208-210), the king would occasionally decide to marry a particular daughter or sister of his to a particular nobleman, 1 and it is stated in one passage that the proposed marriage (that between a relative of his and a less important lord, a principal) should take place because the king required the lesser lords help and reinforcement. The RM also describes the marriage ceremony of a more important lord, a seor. While the RM states that the marriage being described is that between the a female relative of the king and Don Pedro Cuiniarangari, who would become the kings right hand man in the context of the encounter with the Spaniards, it also states that this marriage is being described within the document because it is typical of marriage between the female relatives of the king and lesser lords. I assume that this proclamation of typicality extends to the specific use of kin terms and rhetoric used as framing devices. In the representation of the marriage of a seor and a female 1 One passage calls the receiving lord as a seor, while another calls a less important lord who receives a female relative of the king for a bride as a principal. The second passage, that involving the principal, states that the king, when drunk, proposed that a female relative be married to that particular lord, and this passage does not describe the wealth that accompanies the woman, as it is in the context of the marriage of the seor. I prefer not to take the passage concerning the kings inebriation too literally, and simply take it to mean that the women sent as brides to less important lords did not require the same amount of strategizing and deliberation that went into selecting a match for a seor. 222

PAGE 223

relative of the king, the RM describes how the woman to be married was dressed in new garments and wore a turquoise necklace and many earrings. Furthermore, other women would accompany her, carrying her jewelry, reed mats, and little baskets. The involvement of such wealth items is not indicated for marriages involving the less important principales. According to the RM, subordinate lords also gave one of their female relatives to the king in marriage (RM, 1956, pp. 183-192). In the context of the chapters concerning the state bureaucracy, the RM also contains chapters that describe the many women who lived in the kings house as well as descriptions of their duties (RM, 1956, pp. 183-192). Only women lived in the kings house, save for a single man who oversaw them. These women are described as daughters of the principales of the kingdom, and the king is said to have had many children by them. In the description of the coronation of a new king, the successor marries all of the women who had been married to his father, and with the passage of time other daughters of the lords and caciques come to be his wives (RM, 1956, p. 230). Each woman had a responsibility regarding some aspect of the kings daily life: one kept certain articles of jewelry (including one who kept track of his bezotes of gold and of turquoise), another looked after certain of his garments, another served him his drinking cup, etc. These women were, furthermore, cloistered; they appeared in public only during the religious festivals, and then only to dance with the king. They were responsible for making cloth and food for the god Curicaueri, and it was said that they were wives of that god (RM, 1956, pp. 182-183). Lastly, the penalty for a lord taking one of these women (presumably as a wife or perhaps merely to carry on an affair) was, to say the least, draconian. Not only would the offending lord be killed, but also his children, his wife, his relatives, and all who were in that lords house would be killed; because none had reported it to the king, all were sentenced to death as traitors (RM, 1956, p. 201). 223

PAGE 224

Figure 7-1. Idealized model of marriage exchange between the Tarascan king and noblemen in which the daughters (women are represented by circles) of the noblemen are given to the king in marriage, and vice versa. The size differences reflect differences in status between the women thus exchanged and the men between whom they are exchanged. Therefore women flowed from the king to subordinate lords and vice versa (figure 7-1). While the exchange of women for women appears symmetrical, in reality the women exchanged were likely not equivalent to one another; the female relatives of the king would have been more highly valued (i.e., of higher status) than their counterparts, the daughters of subordinate lords. This is indicated by the fact that the female relatives of the king were accompanied by large amounts of jewelry and wealth objects, and other women. The status of these brides was made manifest, and made visible, by such material and human objects. Even if a specific woman was not related to the king by blood, she should have been accorded some status by being so tightly bound to the person of the king as a result of her being cloistered. This leads to another consideration: as a wife of Curicaueri she enjoyed some contact with the sacred and should therefore have garnered some sacrality and therefore also higher status. On the other hand, I discuss evidence in Chapter 8 that subordinate lords would not have had the economic wherewithal to construct the high value of their own female relatives that were sent to marry the king through dowering them with valuable wealth objects. Male nobles apparently did not have access to foreign wealth objects of any kind (outside of their relationship 224

PAGE 225

to the king) with which they could reciprocate the wealth that accompanied the kings relative that he received. The only wealth objects that these male nobles were buried with presumably were attained through other connections with the king and the state system of prestige reward, such as bezotes and ornaments made of bronze. Furthermore, this evidence supports the claims of Gorenstein and Pollard (1983), based on their interpretation of the traders in the employ of the king, that the king maintained a monopoly on foreign trade goods through a monopoly on long-distance traders. However, the RM (1956, p. 184) does state that the male offspring of a woman given to the king in marriage is provided with a house and a parcel of land, but the womans kin group, i.e., the family of the lower nobleman, is obligated to come and work those lands (or more generally provide labor for the lands to be worked) as well as make cloth for the son. It is not apparent whether this offspring is affiliated with or a member of the royal dynasty, as the Tarascan kinship system as revealed by kin terms seems to have been largely bilateral, although succession to office appears to have been dominated by patrilineality in prehispanic times (Pollard, 1993, p. 59). It is possible that the land and labor that the womans family gave to her son would have balanced the exchange relationship between subordinate lords and the king. Rhetoric and its Indexical Consequences The rhetoric of certain parts of the speeches made during the wedding ceremony involving the seor and the royal bride casts the marriage thus constituted as something other than a normal marriage. This rhetoric implies that the woman, along with her productive capacities and the wealth objects she is dowered with, are not really or fully transferred. In the representation of his acceptance of the woman, the nobleman to be married states: I have here this lady that is our daughter and our lady, who is not given as a woman. She is not given as a woman, but more so that we will raise her and be tutors of hers (RM, 1956, p. 209, my translation). The speeches included in the representation in the RM of the marriage of a nobleman and noblewoman (i.e., 225

PAGE 226

between two more or less equal members of the lesser nobility) do not contain any passages that similarly negate the relation being established (RM, 1956, pp. 211-213). Similarly, the representation of a commoner marriage does not contain any rhetorical negation of the marriage relationship being realized (RM, 1956, pp. 213-214). The reference to being a tutor to the bride is noteworthy, and together with the rest of the passage indicates that the woman is given not as a wife. That is, in some sense the woman is not exactly exchanged between the king and the nobleman. Neither, therefore, would her productive capacities and wealth be exchanged and given to the lord with no strings attached. The fact that the brides productive capacity is not transferred to the lord is explicitly recognized in the speech of the priest officiating at the marriage between the royal bride and the subordinate lord. The priest tells the couple to behave well and then advises the bride to be productive and make food and cloth for the lord. The priest then admonishes the couple (and then the lord singly) as to why such efforts are important. And if someone enters your house, give him cloth. The king says that what you both will give to them [visitors], it is he who gives it. For he cannot remember all of the caciques and lords to give them cloth and do them favors and to do these for the other people. For this [reason] are you here, sir, that he [the king] has as a brother. (RM, 1956, p. 208, my translation) Within this passage, the priest is represented to remind the lord, the bride, and anyone else within earshot that it is really the king who gives the gifts of food and cloth to the people. To Western sensibilities this may seem to be only a rhetorical reminder to represent the king appropriately. However, if this rhetoric is proclaimed publicly and there is no equivalent reciprocation of the bride on the part of the cacique, as represented in the RM, both the rhetoric and the material contexts of the exchange relationship would have established the womans productive capacities as indexes of the kings agency and not the agency of the local lord. The womans role in the local context of lordly beneficence would have been a direct result of the 226

PAGE 227

kings intentions and wherewithal, and not the local lords. By extension, then, the gift objects that were ostensibly given by the lord would have been attributed to the king by anyone aware of the exchange relation that pertained between the king and the subordinate lord. In short, the agency objectified in the practice of giving gifts to the populace was abducted to be that of the king and the more proximate agency of the lord was encompassed in the process. The priests rhetoric that it is really the king who gives the people gifts through the action of giving the royal bride to the subordinate lord is the logical extension and consequence of the lords speech (as groom) in which he denies taking the royal bride as his wife. In other words, it confirms the interpretation that in denying the marriage relationship, the lord fails to establish an indexical link between himself and the products of the royal brides labor (figure 7-2). Figure 7-2. Idealized model of the exchanges of the noble economy in which noblewomen played an essential role. Dotted lines indicate that the association, or indexical relationship, between the lord and the objects that are given (and which the noblewoman) has produced is eclipsed by the relation these objects have with the noblewoman and by extension the king. In the priests speech, he also reiterates that giving gifts to the people is precisely the reason why the lord has been placed in his post. This is confirmed in other passages of the RM, 227

PAGE 228

outside of the sections describing marriage ceremonies. In the prologue of the RM the friar who compiled and wrote the document claims that the only virtue of the people of Michoacn was their generosity, for they held it to be a great offense to be stingy (RM, 1956, p. 4). This emphasis on generosity is repeated numerous times throughout the rest of the document, both in the legendary history of the Uacusecha royal dynasty and in the ethnographic section. Primary among those items given by the lords were mantas, or reams of cloth. Lords were expected to give cloth items to any persons that came calling, be they some representative of the king or a commoner (RM, 1956, p. 185). The labor of women was vital in the production of both cloth and food (RM, 1956, pp. 15-16, 65, 208). Furthermore, female labor was essential for the care of the god Curicaueri (RM, 1956, pp. 15-16, 65), another responsibility of the nobility in the towns throughout the kingdom (RM, 1956, pp. 153-155, 206). In the chapter concerning the selection of new caciques, the priest instructs those present to place women in the new lords house, so that they may make mantas and food for Curicaueri (RM, 1956, pp. 206). Therefore throughout the text of the RM and reiterated by the priest officiating at the marriage of the royal bride and the subordinate lord, a defining responsibility of a lord was to be generous and give gifts to everyone. Furthermore, the labor of the royal bride was essential to the ability of the lord to perform these duties. Not only, therefore, does the king encompass the lords agency through the preservation of the indexical link between himself and the woman he has given to the subordinate lord. The exchange of the bride also constructs the king as the means by which the lords actions are abducted to be the results of the kings will, intentionality, and agency. Through the very act of marrying the woman to the lord and by preserving the brides association with himself through the material and discursive contexts of the exchange, the king is able to induce the actions of the 228

PAGE 229

lord (i.e., giving gifts to the populace) to be abducted to be the replicas of the action that the king has willed and set in motion. This is precisely the definition of encompassment as explicated in Chapter 4. Investiture of Caciques: the Distributed Person of the King The second context for which there is important evidence for the encompassment of the lower nobility comes in the description of the death of a local leader and the selection of his successor by the king. When the king selected a new cacique, he sent a set of insignia with a priest, and the priest would inform the successor and give that successor the insignia of office that the king had given to the priest for that purpose. The insignia stood, in other words, as material signs (indexes, see Chapters 4 and 5) of the relationship between the king and the cacique, the desire of the king that the lord should be the new cacique, and the authority and legitimacy with which the king was endowing the new cacique. Therefore the RM claims that the encompassment of caciques was effected by, and objectified in, the insignia of office that were given to the officeholders. As I discuss in this section, the representation of the transferal of the insignia indicates that the insignia preserved their indexical association with the king and did not index the lords agency. Furthermore, the representation of the investiture ceremony indicates that the priest whose job it was to give the new cacique the insignia of office and thus raise him to the office of local lord framed the office to which the lord was being raised, reiterating the agency of the kings ancestors in the establishment of local offices of lordship. Death and Succession in Offices of Local Leadership The RM describes the investiture of a new lord following the death of an officeholder (RM, 1956, p. 203). Upon the death of a cacique (a cacique of the province) the king would select his successor after a few days of mourning. The relatives of the dead leader would take the insignia of office (the gold bezote, earspool, and gold bracelets) to the king so that the king 229

PAGE 230

would know of the death. After a few days of mourning had passed, the king would choose a successor from among the male relatives of the dead lord. He would tell a priest to go to the town and install the new lord, ordering that the successor be given the objects that constituted the insignia of office enumerated above. The text explicitly states that the successor is to be given new insignia, indicating that the objects given to the successor were not those that the king had reclaimed from the deceased (a point made by Beltrn, 1986; Carrasco, 1986; Pollard, 2003). The return of the insignia to the king combined with the gift of a completely new set of insignia, in spite of the fact that the new insignia were of exactly the same categories as the old insignia, renders the central role of the king unambiguous in the ceremonies of investiture. Furthermore, it underscores the fact that offices of lordship derived from the king and not through genealogical reckoning within the kin group of the deceased, even though the king chose the successor from among the male relatives of the deceased (Pollard, 2003, p. 49). The insignia of office signified multiple, but interrelated, intentions at once, and in order to understand how the insignia effected the encompassment of the subordinate lord it is necessary to explicitly delineate these intentions. Firstly, the insignia as given to the priest and subsequently given from the priest to the new officeholder indicate that the king has chosen that person to be the new cacique. Secondly, the insignia index the kings desire that the new cacique be able to represent him as an officeholder. As discussed in the legendary history (and as analyzed in Chapter 6), offices of local lordship were created by the ancestors of the king in order so that they might represent the king in the towns of the kingdom. In the representation of the investiture ceremony in the RM, the priest raising the lord to the office of cacique reminds the lord and the people of this historical fact. Third, and in direct relation to the second point, is that the insignia directly index the kings agency (and the agency of his ancestors). As I discuss 230

PAGE 231

in Chapter 4, in order for the subordinate lord to be constructed as a representative of the king, the lords actions must be abducted to be replicas of actions that the king had already desired, willed in motion, and taken some steps to insure the preservation of the indexical link between the objects so given and himself, thereby ensuring the abduction by witnesses of his own primary role in the realization of the action. It is these steps that the RM claims to present in the description of the events surrounding the investiture ceremony. I propose that this representation of the actions surrounding the investiture ceremony can best be interpreted as a claim that the insignia of office should have been perceived as the inalienable property of the king (following Weiner, 1992). As discussed in Chapter 5, by enforcing certain parameters on the exchanges that made up an exchange relationship, the giver of an object could succeed in keeping-while-giving the object, or in other words preserving the perception that the object was inalienable from the giver (Weiner, 1992, pp. 6, 26, 60-63). One prominent way in which objects can preserve their indexical associations with the giver is through cosmological authentication (Weiner, 1992; see Chapter 5) or discursively framing them more generally (Keane, 1997; see Chapter 5). This indeed occurs in the representation of the investiture ceremony. In his summation speech during the investiture of the new lord, the priest reiterates the roles of Hiripan and Tangaxoan (but not Hiquingaje, who also led the wars and instituted offices of local lordship along with his cousins) 2 as the Uacusecha leaders who instituted the offices of local lordship (RM, 1956, pp. 204-205). The priest makes it clear to the new officeholder as well as the people that without the actions of the ancestors of the king, the 2 As discussed in the previous chapter the Uacusecha Chichimecs are transformative beings par excellence. The historical institution of local offices and the act of raising a person to that office in the present are, in their fundamental aspects, transformative acts. In the legendary history of the Uacusecha, Hiquingaje is represented as an Islander in relation to his cousins Hiripan and Tangaxoan, the Chichimecs. Therefore in the context of a transformational act (raising a person to the lordship) that reproduces a prior transformational act (instituting offices of lordship), it is logical that the Chichimecs of the Uacusecha royal dynasty, Hiripan and Tangaxoan, deserve mention in this speech but not the Islander Hiquingaje 231

PAGE 232

people would not be settled, there would be no lords or offices of lordship, and the god Curicaueri would not be cared for; in short, chaos would have prevailed. The newly selected lord therefore owes his status to the actions of the ancestors of the king as well as to the present king who has just had him installed. The priest takes this opportunity to frame the relation, rendering it unambiguous in the minds of the witnesses. The relation between the office and the insignia is left implicit, but the insignia clearly had an important, and even constitutive, role in the investiture ceremony. Without the insignia, there would be no investiture ceremony. The insignia themselves are indexes of the existence of the office and, as noted above, due to their movement as they are passed from king to priest to initiate, are indexes of the kings desire that the initiate should hold that office. Another way in which objects can be constructed as inalienable possessions even when they have been given is the establishment of an expectation that they will be returned to their proper and rightful owners. This is clearly the case in the representation of the sequence of events that surround the investiture ceremony in the RM. The insignia of office derived from the king and were destined to be returned to the king. The representation of the death of a lord presents the return of the insignia of the dead lord as a precondition of the selection of a new lord. As presented in the RM, then, the return of the insignia was already guaranteed at the moment of their distribution. Viewing the insignia of office within the lifetime of the officeholder, conversely, it is apparent that they were merely on loan to the lordas the inalienable property of the king they continued to index the king, even though they might temporarily be in the hands (or on the body) of a subordinate lord. Furthermore, the nobleman thus raised to the lordship does not proffer a gift to the king, not even a nominal gift paid as rent for the insignia. The only gifts that the lord gives are the 232

PAGE 233

mantas, gourd plates, and wreaths that the lord presents to the priest who had just installed him in the lordship (there are no indications of what the priest might have done with these objects). The indexical relationship between the insignia and the king therefore should have remained unchallenged due to the lack of a return gift on the part of the newly selected lord. The lord, in other words, played no role in the exchange relationship between himself and the king. As a consequence, the lord played no role in his own constitution as an officeholder, which was the very purpose of the exchange relationship, for both the king and the lord, in the first place. The ceremony in which the priest installs the new lord is represented in the RM as public and open for all the local people to see. Therefore it should have been clear to everyone that the source of the new lords legitimate authority was embodied in these insignia that remained, and were ostensibly destined to remain, inalienable objects of the king. Insignia and the Agency of Caciques to Represent the King As enumerated above, the insignia of office indexed the intent of the king that the cacique that received them should have the ability, authority, and legitimacy to represent him. Keane (1997, p. 231, see Chapter 4) points out that people do not send representations out into the world unless they have a reasonable chance of achieving the desired effect. Therefore it is necessary to investigate how the insignia of office endowed the lord with the ability to be perceived to possess legitimacy and act authoritatively. The insignia, as I have just discussed, indexed the king due to the nature of the exchange relationship between the king and lord, at least as it was represented in the RM. As such, they would have indexed the kings legitimate authority, itself produced through the kings claims to represent a sociocosmic totality and an altogether unprecedented social being (see Chapter 6). However, the specific objects that composed the insignia of office also had historical associations with authority and leadership as well as indexical relationships to the bodies of the bearers that constituted their agencies in 233

PAGE 234

specific ways. Here I discuss the significatory associations of the insignia of office that they possessed through historical association and their use as adornment. Objects such as bezotes, earspools, and other forms of bodily adornment had for centuries been associated with high status within Michoacn (Hosler, 1994; Pollard and Cahue, 1999) and in Mesoamerica in general for millennia (e.g., Joyce, 2000). In the prehistory of Michoacn, beginning in at least the Classic period if not before, forms of bodily adornment were important markers of status. At Erongarcuaro itself, one of the Loma Alta burials from field ER-22 contained a clay earspool (Pollard, n.d.b). By the Late Classic to Epi-Classic period, emergent elites at the site of Urichu were adorning themselves with beads, plaques, and bracelets made of shell and precious stone (likely turquoise) (Pollard and Cahue, 1999, p. 270). In the Late Postclassic period during which the Tarascan kingdom rose and dominated Michoacn, elites throughout the kingdom are associated with bezotes, earspools, bracelets, and bells (Macas Goytia, 1990, 1996; Michelet, Ichon, and Migeon, 1988; Pereira, n.d.; Pollard and Cahue, 1999). In the RM, the possession or investiture of such objects of bodily adornment was explicitly connected with high status and the possession of authority, both before and following the establishment of a unified kingdom governed by the Uacusecha (RM, 1956, pp. 203-205). Due to their historical connections to, and objectification of, legitimate authority within Michoacn, and as confirmed by the RM, it is clear that the possession of such forms of bodily adornment were essential to the ability of the lords to construct their authority and legitimacy. Perhaps most emblematic of the ability of lords to exert and legitimize their authority was the bezote. Bezotes are listed first among the insignia of office in the passage from the RM cited above in which lords were transformed into officeholders. Pollard (2003, p. 51) pointed out how important bezotes were in the recognition of the authority of lords. She notes that according to 234

PAGE 235

the definitions given in the Gilberti dictionary (Gilberti, 1962, p. 15), the Tarascan word for bezote was angamecua, and the plural form of a specific title or office was angamencha, which is defined as those who wore bezotes between the lip and the beard (i.e. in a hole in their lower lip) (Lagunas, 1983[1574], p. 221). The verb angameni means to put a bezote on a noble (Lagunas, 1983[1574], pp. 221-222), and Pollard (2003, p. 51) argues convincingly that the verb angamucuni should be interpreted as standing at a social boundary, i.e. that between the king and the commoners. The primacy given the bezote in the RM (1956, p. 33, 203) among the insignia of office and the definitions of these related words as analyzed by Pollard suggest that bezotes can be viewed as the sine qua non of legitimate authority and noble status. Because the bezote was placed so that it would protrude from the bottom lip for all to see (and bezotes, as a dot of blue representing the turquoise of the bezote, are indeed easy to see in many of the illustrations of the RM), it stood in an indexical relationship to the mouth and to that which issues from the mouth, namely speech. As Seeger (1974, p. 212) advises, [t]he ornamentation of an organ may be related to the symbolic meaning of the organ in a society. Therefore in order to understand the significance of bodily adornments, the importance accorded to those practices with which the adorned body part is associated should be analyzed. Houston and Cummings (2001, p. 369-370) have noted that the bezotes in Mesoamerica possessed indexical relationships to speech due to their proximity to the mouth. Bezotes affected the way in which the speech of the bearer was perceived and received; bezotes constructed the speech of bearers as different from, and more efficacious than, that of persons who did not possess bezotes (Houston and Cummings, 2001, p369-370). The relation between bezotes, lordship, and speech, is well established in the ethnohistoric record. In addition to giving gifts of food and cloth to the people, an important responsibility of 235

PAGE 236

the caciques was to speak to the local commoners. In two speeches in the RM (1956, pp. 153, 204, 206), it is clear that the cacique is conceptualized to give (in the sense of presenting a gift) order to the populace by commanding them to be settled, pacified, and not to quarrel amongst themselves (and likely resolving disputes between commoners). With his commands in place, the people could prosper and tend to the fields whose produce would sustain them (to say nothing of the state). In return, by implication, the people should tend to the caciques fields so that he may fulfill his duties. Speech, therefore, was essential to the ability of caciques to act as caciques. By endowing caciques with bezotes and the other insignia of office, the king was endowing them with the ability to be perceived as authority figures and therefore as people who should be followed. In other words, the king desired and set in motion events and objects that would give caciques the ability to effectively govern the commoners in their charge. Insigniaand Thus Lordsas the Distributed Personhood of the King Because the insignia of office given to the caciques simultaneously indexed the ability of the caciques to effectively govern (i.e., to have their commands heard and followed), the desire of the king that the caciques should possess this ability, and the kings primary (exclusive) agency in the constitution of the relation between himself and the caciques, the insignia of office effected the encompassment of the caciques by tying all of these relations together. They constituted the ability of the caciques to effectively govern as an index of the kings agency and not as an index of the agency of the caciques themselves. By giving the caciques the objects necessary to govern, which is to say to represent himself, and by preserving the indexical relationship between himself and the insignia, the king did not so much transform the agency of the lords into his own agency but simultaneously endowed the caciques with agency and constituted that agency as primarily his. The encompassment of the actions of the caciques was a condition of the very realization of those actions by the caciques. 236

PAGE 237

It is important to emphasize that the representation in the RM of the exchange relationship between the king and subordinate lords was an exercise in framing that relationship and the insignia that it involved and which indexed it. The representation of the investiture ceremony was intended to explicate the encompassment of the subordinate nobility by purporting to explain how the interactions between the king and the subordinate nobility should have happened. The representation itself indicates that the informants that provided this representation believed that such ceremonies and the transferal of objects that they included could have effected the encompassment of the caciques if they had been realized as they are described. Whether or not these ceremonies and exchanges actually did happen as they are described is another question altogether. Conclusion As represented in the RM, the relationship that was actively produced between the king and subordinate lords through social interaction was a relation whereby the king managed to ensure the intersubjective abduction of the perception of the lords agentic capacities by proclaiming his prior and sole (i.e., not including the lords themselves) constitutive role in the transactions in which he engaged the lords. The representation of a marriage involving a cacique and a bride from the kings palace in the RM indicates that the royal bride preserved her indexical relationship to the king because the marriage was somehow incomplete and did not fully establish an indexical relationship between the bride and the cacique. The framing of the marriage as incomplete would have had serious consequences for the construction of the legitimacy of the lower nobility. The major roles of the nobles were to present all who called on them with gifts of food and cloth. They were, ostensibly, the face of the state, the location at which the tribute and service to the state by the commoners was reciprocated through the demonstrated generosity and beneficence of the state toward those same commoners. However, 237

PAGE 238

the things given to the commoners were things produced by precisely those women to which the lower noblemen denied their claim in the representation of marriage speeches. Therefore the indexical link between the noblemen and the objects given was tenuous and superceded by the relationships between the objects and the women, and ultimately between the objects given and the king. The actions of the lords therefore would have indexed the king; they would have been abducted to be actions that the king had willed, set in motion by giving the woman in marriage, and laid claim to by having a priest frame the marriage relation, to say nothing of the likelihood that the royal bride could not be reciprocated by an equivalent bride given to the king by the lord. The kings ability to encompass the actions of the caciques is also reflected in the investiture ceremony as it is represented in the RM. In this representation, the insignia were guaranteed to be returned to the king, were indexes of the office of local lordship that was cosmologically authenticated to be the result of the agencies of the kings ancestors, and were unreciprocated by the cacique himself. Not even a nominal gift was given to the king as an effort to obviate the indexical link between the insignia and the king while those insignia were on loan to the lord. The insignia therefore would have simultaneously constituted the caciques ability to be perceived as possessing legitimate authority and therefore to have their orders followed and encompassed that ability as something that the king had willed and unilaterally set in motion so that the caciques would represent him, speaking on his behalf. The actions of the lesser noblemen were thus represented as encompassed by the king. Their actions, and more specifically those actions that were central to their identity as lords and the execution of the offices they held, are intersubjectively constructed as the kings actions through the representation of kings central and often sole role in the dispersal of aspects (namely, women and bezotes) of his person. The noblemen themselves became the distributed 238

PAGE 239

personhood of the king and not actors in their own right. Or at least this is what is being represented in the RM as a typified, generalized, and hence universal aspect of the hierarchical relations in the Tarascan kingdom. Such representations can be considered merely the revelation and instantiation of a Tarascan general theory of hierarchy and encompassment through exchanges. But for such unquestioning abductions of agency to have been regularly made by commoners and witnesses, there should not have been any slippage between the rhetoric of exchange and their material constitution. In other words, the real world of objects must have been harnessed and put to use in an encompassing process in its own right. To evaluate the material parameters of such exchanges and the reality of the practice of framing these exchanges in the same manner as the RM does in a delimited and defined context, in the next chapter I examine two case studies through ethnohistoric, and more prominently archaeological means. 239

PAGE 240

CHAPTER 8 ENCOMPASSMENT IN THE TARASCAN CORE: THE CASE STUDIES OF ERONGARCUARO AND URICHU Introduction In this chapter I reconstruct how relationships between the Tarascan king and subordinate lords at two sites within the Tarascan core were materially constituted. The relationships to be examined are those between the king (and the Uacusecha royal dynasty) and the lords of the sites of Erongarcuaro and Urichu. More specifically I analyze archaeological as well as ethnohistoric evidence in order to evaluate the extent to which material practices effected the encompassment of the lords of these two sites by the king. Through such an examination, I demonstrate that the significatory relationships objectified in the objects that should have been perceived as the kings distributed personhood (see Chapter 4) were not always as simple and straightforward as the representations in the Relacin de Michoacn claimed (see Chapters 6 and 7). The data from Urichu indicate that the significatory relationships of these objects were fairly simple and should have led to the abduction of the kings primary role in the constitution of the agency and personhood of the local lord. However, the data concerning the lords of Erongarcuaro present a much greater departure from the representations contained in the RM. Rather than a simple, binary relationship between the king as producer and giver and the objects so given and the resultant abduction of these objects as the objectified and distributed personhood of the king, at Erongarcuaro there was an expansion of the significatory relationships objectified in objects that could have potentially been exploited by caciques at that site. This expansion was established by the production of bezotes and other lapidary objects at the site and associated with what was likely an elite residence at the site. However, other material practices, such as marriage exchange and discursive framing, would have effected the encompassment of the lords of Erongarcuaro and mitigated against the 240

PAGE 241

risk that a relation of encompassment would not have been abducted by witnesses. Therefore the evidence indicates that the king did indeed encompass the nobles at Erongarcuaro, but the relation of encompassment was likely more tenuous and relied on those other material practices to carry the significatory load in the construction of that relation of encompassment. Marriage Exchange at Erongarcuaro and Urichu There is evidence in the legendary history of the Uacusecha, contained in the RM (see Chapter 6), that indicates that there were apparently long-standing relationships of intermarriage and alliance between the Uacusecha and the lords of Erongarcuaro and Urichu. Moreover, these towns were said to be allied with the Uacusecha in the conquest of Zirahuen, an event that jump-started Uacusecha expansion through military conquest and ultimately led to the formation of the Tarascan kingdom. In addition, the relationship between the king and the towns of Erongarcuaro and Urichu can be investigated through more locally oriented documents. These documents contain information concerning specific marriages or specific consequences of an affinal relation (in contrast to the typified representation in the RM; see Chapter 7). Here I examine these representations in combination with the more specific ethnohistoric evidence relating to these two sites to suggest that relations of encompassment were most likely practiced and reproduced between the king and the lords of Erongarcuaro and Urichu. Representations of Relations Between the King and the Lords of Erongarcuaro in the RM In the complete text of the RM, the towns of Erongarcuaro and Urichu are mentioned only sparsely. This meager attention does present a view of how members of the Uacusecha royal dynasty viewed or represented their relationship with the lords of these two sites, however. This relationship was apparently one of alliance and intermarriage, which, although filtered through the lens of the legendary history of the Uacusecha, perhaps had some correspondence to the 241

PAGE 242

actual historical constitution of relations between the lords of these two towns and the Uacusecha royal dynasty. Affinal relations in the RM There is evidence in the RM for affinal relations between the Uacusecha and the lords of both Erongarcuaro and Urichu. In a passage in the RM (1956, pp. 94-96) the mother of Hiripan and Tangaxoan leads the youths to Erongarcuaro and then to Urichu in search of a place to stay (see figure 6-1 for the place of these characters within the Uacusecha). The mother takes them to these places because the lords at each place are brothers of hers, and therefore uncles of the youths. At both places they are cast out of the lords houses due to the fact that they neglect the duties they had promised to perform, because they are constantly gathering wood for the temples instead. 1 Upon being cast out, the youths mother remarks that the lord is niggardly, which as I discuss in Chapter 7, is a great offense among the lords. While these events took place before the advent of the wars of conquest initiated by the Hiripan and Tangaxoan (with their cousin Hiquingaje; see Chapter 6), and I have elsewhere advocated interpreting the legendary history as a whole in structuralist, rather than literalist terms (Haskell, 2003, 2008), I believe this passage likely indicates the presence of an actual historical relationship between the Uacusecha and the noble families of these towns (insofar as this relationship, once historically constituted, became structural, i.e., reproduced), even if it need not have been established in the manner portrayed in the RM. The mother of the boys would have been married to either Zetaco or Aramen, 2 cousins of Tariacuri, and this marriage would have 1 This act of piety is ultimately rewarded, as the lord of Pareo recognizes the youths and their lordly status, allowing them to gather word for the temples instead of performing household duties. It is while they are gathering wood that they are discovered by Tariacuris priests and ultimately brought to him. Under his tutelage the youths go on to conquer many towns and institute the bureaucratic structure of the Tarascan state. 2 The exact kin relationships between Zetaco and Aramen, and their sons Hiripan and Tangaxoan, are ambiguous in the RM. Zetaco and Aramen are themselves the sons of Uapeani, brother of Pauacume, the Uacusecha leader who 242

PAGE 243

taken place well before Uacusecha expansion. Following Fogelsons (1989) discussion of encapsulating events, in which long-established relations might be represented by singular events, I propose that a general and ongoing relationship of bride exchange was represented in the RM as a singular, encapsulating event. That is, the relationship between the Uacusecha and the nobles of these two towns was located at some point in the past and subsumed within a single woman where it was rhetorically convenient and logical in the context of the specific movements in time and space of the protagonists of the passage, Hiripan and Tangaxoan. 3 Military alliance in the RM Both Erongarcuaro and Urichu were named as allies in the context of the plans of the war that the Uacusecha leaders Hiripan, Tangaxoan, and Hiquingaje desired to launch against Hivacha, lord of Zirahuen. The war against Hivacha, as represented in the RM, initiated the wars of conquest that are the historical foundation of the Tarascan kingdom (see Chapter 6). The triumvirate of Uacusecha youths benefited from the help of their sage elder Tariacuri, who arranged for the help of allies in their war against Hivacha. As Tariacuri explained to the three youths: marries the fishermans daughter from Xaraquaro. The mother of Zetaco and Aramen is never explicitly named, in contrast to Tariacuris mother. Hiripan and Tangaxoan are simply named as the sons of Zetaco and Aramen, and no indication is made as to whether they are the sons of one of these brothers or Hiripan is the son of one and Tangaxoan is the son of the other. Similar to the mother of Zetaco and Aramen, the mother of Hiripan and Tangaxoan is never explicitly named, nor is it explicitly stated where she is from. As I have discussed elsewhere (Haskell, 2003), the dubious parentage of Zetaco and Aramen, and their sons Hiripan and Tangaxoan, is a rhetorical device to emphasize their Chichimecness. In this way, especially with regard to Hiripan and Tangaxoan, whose specific parentage on both sides is unclear, these two are simply reproductions of their Chichimec fathers, i.e. they are the same persons but in the next generation. As the mother of Hiripan and Tangaxoan, the woman would have been the wife of either Zetaco or Aramen. As a sister of the lords of both Erongarcuaro and Urichu (as well as Pareo), the woman would have married into the Uacusecha and was representative of the polities of this part of the Lake Ptzcuaro Basin (that is, the southwest part of corner) as a whole. 3 In this context it is the movement of the boys and their mother from Asajo, located northwest of Lake Ptzcuaro, down the west side of Lake Ptzcuaro, represented by the stops in Erongarcuaro and Urichu, that has them ending up in Ptzcuaro when the boys are brought there by their uncle Tariacuri. (RM, 1956, pp. 91-100; see also Seler, 1993). 243

PAGE 244

Look because I want to show you the town; this line here is the road where you have to go; this here is a mountain. You must go by here and the men of Cumachen must go by here and those of Erongarcuaro and Urichu and Pichataro will go by this road that you come by, that I signal to you that you will take tomorrow. (RM, 1956, p. 148, my translation) According to this representation, Erongarcuaro and Urichu were included in the war party by alliance and through the diplomatic efforts of Tariacuri, who had arranged for the help of people from other towns as well. Due to the fact that the defeat of Hivacha and the conquest of Zirahuen marked the beginning of the conquests that establishes the Tarascan kingdom, this passage indicates that these towns were incorporated into the state by peaceful means, and that this arrangement potentially included some agreement of power sharing (at least initially). As further evidence of such an arrangement of sharing power, the lengthy list of towns that the Uacusecha-led war party conquered that is given towards the end of the RM (1956, pp. 151-155) does not include Erongarcuaro nor Urichu. Additionally, there are passages in the RM (1956, pp. 151-155) that indicate that lords involved in the conquests were able to establish ties of lordship (and perhaps tributary ties or the taking of patrimonial lands) in the towns they had conquered as part of the allied forces led by the Uacusecha. Perhaps the promotion of such perquisites for allied lords who participated in the conquests were necessary in order to insure their involvement in those conquests in the first place. In any event, the lords of Erongarcuaro and Urichu appear to have been incorporated into the Tarascan state by peaceful, rather than forceful, means. Not only is this important as a potential indicator of alliance cemented by marriage exchange, but it also indicates that the Uacusecha should not have been able to claim patrimonial lands within Erongarcuaro nor Urichu as spoils of war. This is an essential point in relation to a land petition brought in the colonial era by descendants of the last king, which I discuss below. 244

PAGE 245

Intermarriage According to other Colonial Documents In addition to the representation of a lengthy relation of intermarriage and alliance between the Uacusecha and the lords of Erongarcuaro and Urichu, there is more specific ethnohistoric evidence for intermarriage between the Uacusecha royal dynasty and the noble family of Erongarcuaro. In 1579, Doa Ana of Erongarcuaro, a daughter of the cacique of Erongarcuaro, married Don Juan, a nobleman of Ihuatzio (Lpez Sarrelangue, 1965, p. 244; Roskamp and Csar Villa, 2004, p. 218n4). While this does not represent a marriage of the daughter of a principal to one of the descendants of the king (or the kings equivalent in the colonial era, the indigenous governor of Michoacn in Ptzcuaro), the marriage did involve a lord of Ihuatzio, one of the three capitals of the prehispanic Tarascan kingdom and very probably second in status to Tzintzuntzan in the prehispanic era. 4 Roskamp and Csar Villa (2004, p. 218n4) cite the marriage between the nobleman of Ihuatzio and the noblewoman from Erongarcuaro as evidence of a close relationship between the noble families of Erongarcuaro and Ihuatzio, relevant because they both had a presence in the towns of the Sierra Tarasca west of Lake Ptzcuaro. 5 It should be noted that the practice of polygyny had been outlawed by the Spanish colonial authorities (Kuthy, 2003; Warren, 1985, pp. 17-18) and so the king and descendants of past kings (i.e., the upper nobility) could not have married numerous women as 4 According to the legendary history of the Uacusecha contained in the RM, the capital was moved from Ihuatzio to Tzintzuntzan during the reign of Tzitzispandaquare, who ruled immediately prior to Zuangua, who reigned at the time of the arrival of the Spaniards. The relatively high status of the lords of Ihuatzio is also indicated by the fact that after the death of Zuangua, when the highest lords of the kingdom had selected Tzintzicha Tangaxoan to be the next king, he demurred and suggested that the lord of Ihuaztio be king instead (RM, 1956, p. 246). Gorenstein and Pollard (1983, p. 118) suggest that Ihuatzio was second in status to Tzintzuntzan, and that by the time of the Spaniards arrival in Michoacn, Ptzcuaro held a largely ceremonial or ritual significance compared to the other two capitals. 5 A document dating to 1556 states that lands in many of the towns (Peribn, Chern, Sevina, Aranza, and Paracho) listed as subjects of Erongarcuaro in the Caravajal Visita had been possessed by lords of Ihuatzio dating back into the prehispanic era (beginning with Hiripans son Tucuruan). (Martnez Baracs and Espinosa Morales, 1999, pp. 45-46, document in the Archivo Historico de la Ciudad de Ptzcuaro) 245

PAGE 246

had been the custom in the prehispanic era. Perhaps the marriages that bound subordinate lords to the king in the prehispanic era were diluted among the upper nobility as a whole in the early colonial era, as members of the royal families of the three capitals exchanged wives with lords of the towns throughout the kingdom in order to attempt to hold together the traditional alliances and relationships that the kingship had assumed in the prehispanic era. However, Kuthy-Saengers (1996) investigation of succession to important offices in the early colonial period in Michoacn demonstrates that while the highest ranking members of the indigenous elite did not marry more than one woman, they continued to father children with women other than their wives. Moreover, access to important offices was not restricted for such illegitimate children. In addition to the marriage of Doa Ana of Erongarcuaro to Don Juan of Ihuatzio, a land claim pertaining to Erongarcuaro is further evidence of a relationship between past kings and lords of that site. A 1563 document (a viceregal mandate discussed in Martnez Baracs, 2005, pp. 362-363; see Chapter 2) states that the family members of Don Antonio Huitzimengari, the indigenous governor of Michoacn from 1545 to 1562 (Martnez Baracs, 2005, pp. 306, 360) and the last surviving legitimate heir of the last Tarascan king, claimed a patrimonial estate in Erongarcuaro. According to this document, laborers from Erongarcuaro testified, one year after the death of Don Antonio Huitzimengari, that they worked a parcel of land within the boundaries of Erongarcuaro named Urecho, and had been taxed by the then-deceased governor during his lifetime (Martnez Baracs, 2005, pp. 361-362). The similarity of the name of this parcel of land to the neighboring town of Urichu is likely a coincidence and should not be taken to indicate that separate town (Pollard, personal communication, June, 2006); if the implicated parcel was the town of Urichu, then Urichu and not Erongarcuaro would likely have been named due to the fact that Urichu had its own lord and 246

PAGE 247

governing body. Furthermore, the entire town of Urichu would have constituted quite an estate, as the Late Postclassic archaeological site itself covered 90 hectares (Pollard n.d.a; Pollard and Cahue 1999), not including the adjacent agricultural lands that would have been worked by its inhabitants. On these grounds, I believe that the parcel in question was actually located within the limits of Erongarcuaro. Due to the fact that Erongarcuaro was, according to the legendary history of the royal dynasty, never militarily conquered by the Uacusecha (see above), I suggest that this land and the labor to work it did not come into Uacusecha possession through military conquest and the taking of patrimonial estates by force of arms. Rather, it probably became a Uacusecha possession through the practice of a lord granting land and labor to the son of a daughter given to the king in marriage (RM, 1956, p. 184; see also Chapter 7). Therefore the land claim of Don Antonio Huitzimengari is further evidence for a marriage relationship between the noble family of Erongarcuaro and the Uacusecha royal dynasty. The ethnohistoric evidence for marriage exchange between the Uacusecha royal dynasty and the noble families of Erongarcuaro and Urichu is, however, consistently one-way. Women were given from the noble families of Erongarcuaro and Urichu to male members of the Uacusecha royal dynasty: the mother of Hiripan and Tangaxoan married into the Uacusecha, Doa Ana of Erongarcuaro married Don Juan of Ihuatzio, and Don Antonio Huitzimengari claimed lands that would have entered into Uacusecha hands through the in-marriage of a female from Erongarcuaro. 6 There is archaeological evidence from Urichu, however, that women did 6 As noted in Chapter 7, however, this exchange of a woman, dowered with land, would likely have been to complete an exchange relationship with the king. The dowry of land would have been a last resort to contest the encompassment constituted by the presumed gift of the royal woman to the noble family of Erongarcuaro. 247

PAGE 248

flow in the opposite direction and served to ennoble, but also and as suggested in Chapter 7, to encompass, the local nobility there. Archaeological Evidence for Royal Marriage into Urichu While there is a lacuna of information in the ethnohistoric record concerning the exchange of women going from the king to the noble families of either Erongarcuaro or Urichu, there is strong archaeological evidence, in the form of mortuary data (Pollard and Cahue, 1999), for women from the royal dynasty marrying in to the noble family at Urichu. In general, the females buried in the noble residence located in Area 1 of the site of Urichu were accompanied by more valuable grave goods and more importantly grave goods that would have derived from the king. Males were buried with no such objects, save for one adult male burial that contained a bezote that likely derived from the king. I explain this disparity as the material proclamation of the affiliation between the females at Urichu and the king. Furthermore, I argue that the poverty of the male burials indicates that they were unable to procure such wealth objects and appropriate them for their own uses, including their use in their own burials, the burials of male relatives, or the dowering of females married off to the king or royal dynasty. In order to demonstrate that the disparities were not merely the result gender codes that regulated the kinds of objects that were considered proper for burial with males or females irrespective of status or social connections in society, I also discuss representations of mortuary ceremonies in the RM (1956) and archaeological data from the capital of Tzintzuntzan (Cabrera Castro, 1977-1978, 1987; Rubn de la Borbolla, 1944). Mortuary Evidence From Urichu During the Late Postclassic period (Tariacuri phase, 1350-1520 AD) females at Urichu were buried with wealth objects that either in production or in raw material originated outside of the limits of the Tarascan kingdom or were acquired from the king (Pollard and Cahue, 1999). 248

PAGE 249

For example, shell objects either definitively or possibly from the Pacific Ocean was used in pendants from burials 7, 8, 30 and 32, which were all female with the possible addition of burial 29, the sex of which was unable to be determined. Of the seven burials with metal (bronze alloys) objects whose place of manufactured is given as probably Tzintzuntzan by Pollard and Cahue (1999), five are female, one is the burial of two unsexed adolescents, and one is male. The only burial with a jadeite artifact is burial 7, a female burial. If indeed the Tarascan king controlled external trade as Gorenstein and Pollard (1983, pp. 108-111) and Pollard (1993, p. 119) suggest, these objects would have had to come from the king. Their association with females at Urichu is therefore significant. While I am not necessarily arguing that such goods were generally or universally associated with femininity or gendered female, I am proposing is that these grave goods constructed and proclaimed (i.e., represented) a relationship in which these women not only originated from the royal dynasty but also continued, even in death, to be linked to the royal dynasty via their grave goods. As discussed in Chapter 5, grave goods and mortuary treatment are not simply a reflection of status, but represent a performance in which relationships between the living and the dead and among the living through the deceased could be proclaimed. The fact that the rich female burials of Urichu included highly valued objects was not simply a reflection of their possessions in life, but the act of depositing these objects, derived from the king, with the deceased woman reiterated her ties to the king and the fact that that relationship was never severed by her marriage relationship. In essence, this representation of a demonstrated and preserved link between the noblewomen of the site and the king was analogous to the rhetoric of the priest officiating the marriage between a royal bride and a local level lord as represented in the RM (see Chapter 7). 249

PAGE 250

Insofar as the kind of wealth objects consisting mostly of foreign derived objects and/or precious metals were historically essential to the construction of authority in Michoacn, these women would have been essential to the material construction of authority of the caciques of Urichu. The male noblemen apparently could not directly appropriate these wealth objects from their wives for themselves, nor could they acquire such objects on their own accord and deposit them with their own selves or male relatives (at least as evidenced in the poverty of adult male burials at the site, including a lord buried with a bezote; see below). Therefore noblemen had to gain access to the legitimacy that these objects symbolized through their relationships with their wives. Furthermore, I assume that if noblemen at Urichu had access to the kinds of wealth objects that could have been used to dower their own women that were sent to marry the king, this wealth would have been represented in the mortuary data from Urichu. Therefore the noblemen of Urichu would not have had the material means necessary to avoid encompassment by the king through dowering women with valuables and constructing an exchange relationship with the king that was more symmetrical. By constructing, preserving, and publicly proclaiming a material link with the bride of the lord at the moment of her death, the king constructed his own presence in the ruling husband and wife pair of Urichu. It was his wealth, placed on the bodies of the females that married into the noble family of Urichu, that gave the caciques rule the legitimacy it needed but which would have been abducted to be the distributed personhood (Gell, 1998; see Chapter 4) of the king. In other words, the caciques wealth and therefore legitimacy was actually that of the woman, and in turn the womans very placement at Urichu within that noble household were abducted to be indexes of the kings agency, as the local lord could not have demonstrated any agentic capacity through reciprocation in the constitution of his relationship with the noblewoman or with the 250

PAGE 251

king. As discussed in Chapter 7, moreover, the wife of the local lord was central to the lords ability to carry out his lordly duties, by producing the objects that the lord was expected to give to the commoners in his charge and to the god Curicaueri. Therefore his agency as a lord was encompassed, as it was abducted by onlookers as originating with the king, who desired and was the sole agent responsible for setting in motion the marriage that would in the end result in the lords ability to carry out the duties of his office. While no comparable mortuary evidence exists for Erongarcuaro, I propose that much the same relationship likely existed between the male nobles of Erongarcuaro and the king as existed between the male nobles of Urichu and the king. Furthermore, the ethnohistoric record does indicate that marriage relationships were strong between the noble family of Erongarcuaro and the Uacusecha royal dynasty, even into the colonial era, as discussed above. Such a strong relationship probably involved the exchange of women in both directions, even if there is only evidence of women marrying out of Erongarcuaro and into the Uacusecha royal dynasty. This does not necessarily mean that the relationship between Erongarcuaro and the Uacusecha royal dynasty was as asymmetrical (more on the potential asymmetry of exchange below in the context of lapidary production at Erongarcuaro) as it apparently was at Urichu (and other sites in the Tarascan state, as documented by Pereira n.d.), but I believe the king likely did ensnare the local nobility of Erongarcuaro by giving them high status women, accompanied with wealth objects which the lords of Erongarcuaro would not have been able to reciprocate equitably. This interpretation relies on the assumption that there should not have been any other factor determining what kinds of objects would have been suitable for interment with males or females. I defend the validity of this assumption with reference to mortuary data from the capital of Tzintzuntzan. 251

PAGE 252

Mortuary Ceremonies at Tzintzuntzan In order to demonstrate that the grave goods interred with males and females were not simply due to the categorization of certain objects as male or female, mortuary evidence from the capital of Tzintzuntzan is necessary. The only representation in the RM of a mortuary ceremony at Tzintzuntzan pertains to the death of a king (1956, pp. 219-223). The representation had a recent instance of such an event to draw from, as king Zuangua had died roughly two decades before the production of the document, when the Spaniards had arrived in Central Mexico (RM, 1956, p. 245). Obviously, the king was the highest status individual in indigenous society, and perhaps those objects suitable for his burial might not have been suitable for burial with individuals of lower status. However, the raw materials of the objects that are described as being interred with the dead king are similar to the materials recovered in the female burials of Urichu, supporting the conclusion that the objects interred with females but not males at Urichu were not simply feminine objects that would not have been interred with males regardless of status. The evidence from the RM is reinforced by archaeological evidence from Tzintzuntzan, as I discuss below. The RM (1956, p. 219) records that the dead king was buried adorned with some or all of the jewelry he had worn in life. This included gold bells, gold bracelets, gold ear spools, a turquoise necklace, and a bezote with inlaid turquoise. The king was cremated, and his remains were bundled up and buried at the foot of the temple of Curicaueri (RM, 1956, p. 222). Those lords and ladies who had accompanied the king in life were similarly adorned with all of their insignia and were killed to accompany him in death. They were buried behind the temple of Curicaueri (RM, 1956, pp. 220-222). The fifth field season at Tzintzuntzan, as reported by Rubn de la Borbolla (1944), excavated two multiple burials adjacent to ycata 5 (ycatas are uniquely shaped temple 252

PAGE 253

platforms; see Chapter 5). Nine adult males were recovered in one of the burials, and approximately five females were recovered in the other burial. Rubn de la Borbolla only summarized the grave goods associated with the burials in the 1944 publication, but his descriptions of the contents of the burials contrast markedly with the pattern at Urichu. Numerous complete ceramic pipes, obsidian earspools (some of which had inlaid turquoise), obsidian bezotes (some of which had inlaid turquoise and gold rims), large rings made of gold/copper with inlaid turquoise, copper bells, and bracelets made of cotton cloth and copper were associated with the individuals in the multiple burial composed entirely of males. One burial contained a silver tweezer of the kind worn by Tarascan priests (tweezers were thin metal objects with crescent shapes at both ends, doubled over on themselves so that the crescent ends lined up with one another). Rubn de la Borbolla (1944, p. 130) states that the multiple burial composed entirely of females was characterized by a great quantity of ceramic vessels, in contrast to the male burials, which possessed few such objects (other than the pipes). The females were interred with numerous copper needles, brooches with copper bell tops, and small copper bells that Rubn de la Borbolla (1944, p. 130) believed were likely woven into a cotton fabric or series of knotted and interwoven strings to form bands of bells that were worn about the wrists and legs. The tenth field season carried out at Tzintzuntzan, led by Rubn Cabrera Castro in 1977-1978, recovered 17 individuals adjacent to ycata 3 (Cabrera Castro, 1987, p. 562). The field season and some of the findings are discussed in Cabrera Castro (1987). In that publication, Cabrera Castro notes that there was significant evidence of looting, likely the activities of Spaniards in the first few years of their presence in Michoacn (Cabrera Castro, 1987, pp. 535, 562). The skeletal remains were never sexed according to their morphological characteristics. 253

PAGE 254

With these tremendous caveats in mind, the field notes of the mortuary remains from the 1977-1978 field season help to augment the sample size of burials revealed by Rubn de la Borbollas excavations during the fifth field season just discussed. 7 I assume that because these individuals were buried at Tzintzuntzan they must have been members of, or closely associated with, the royal dynasty ( a conclusion reached by Cabrera Castro himself; Cabrera Castro, 1987, p. 562). Without the proper sexing of those individuals recovered in the 1977-1978 field season, I must rely on the presence of gendered objects, similar to Pereiras (n.d.) use of such objects in his analysis of mortuary contexts from various sites in the Tarascan kingdom. Pereiras examination of mortuary contexts includes an analysis of the excavated materials from Huandacareo; the mortuary remains from this site were similarly not sexed. Therefore Pereira uses spindle whorls as a reliable indicator of a feminine individual, based on his analysis of the mortuary evidence from other Tarascan sites (including Urichu; Pollard and Cahue, 1999) in which individuals were sexed, and therefore I follow his example. I also rely on the presence of projectile points, obsidian blades, and ceramic pipes as indicators of male individuals. According to Pereiras analysis, in the era prior to the formation of the Tarascan kingdom, projectile points and bifacial knives are markedly associated with adult males and youths that due to their young age could not be sexed. The continued association of projectile points with males in the Tarascan capital, in contrast to provincial sites in the Tarascan kingdom at which no such objects were interred with males (Pereira, n.d.) is logical given the importance of a male Chichimec identity (i.e., a hunter/warrior identity that incorporated projectile points as symbols 7 The few mortuary remains excavated earlier by Rubn de la Borbolla at Tzintzuntzan and Ihuatzio (1941) are published with a great amount of detail with respect to the skeletal remains, but unfortunately the grave goods related to these burials (if any) are not reported. The excavations led by Pia Chan in the 1960s, including the excavation of mortuary contexts, have not been published. Castro Leal (1986) summarizes the mortuary evidence recovered at Tzintzuntzan, mostly with reference to the grave goods recovered at the site but not their contextual associations. 254

PAGE 255

of hunting and warfare) 8 and the claims for its monopolization by the royal dynasty, as discussed in Chapter 6. Pipes, a masculine object in Pereiras analysis, were also associated with the males buried adjacent to ycata 3 (see above). With these assumptions in place it is possible to state that in the small sample of burials excavated during the 1977-1978 field season at Tzintzuntzan, males buried at Tzintzuntzan were in fact buried with objects of the same materials that females were interred with at Urichu (Pollard and Cahue, 1999) and other provincial sites in the Tarascan kingdom (Pereira, n.d.). Burial 2 at Tzintzuntzan contained 2 projectile points, 3 obsidian blades, and 2 copper artifacts, including a ring. Burial 7 contained obsidian blades, obsidian lapidary objects, and copper objects. Burial 12 contained 3 pipes, pieces of a copper bell, a copper ring, and a little copper axe. Finally, the richest burial excavated during the 1977-1978 field season (burial 13) is somewhat of a contradiction of the assumed gender markers discussed above. Spindle whorls were included in the burial, but projectile points were present as well. Additionally, the burial contained a gold bezote and 3 obsidian bezotes. It is possible that bezotes were also masculine objects, as only males were proper officeholders (RM, 1956, p. 114). This burial contained many copper artifacts, including rings, 3 axes (two of which are described as large), and tweezers (see above for a description). Finally, the burial also contained an obsidian ear spool, greenstone beads, and pieces of shell. While the presence of spindle whorls renders this 8 As I note in Chapter 5, in the legendary history in the RM, Uacusecha men are frequently encountered by others while they are busy making arrows. The god Curicaueri, patron deity of the Uacusecha and a solar and hunting deity, is also represented in the legendary history by a piece of obsidian (RM, 1956, p. 125). Finally, in the Early Postclassic elite males throughout Michoacn proclaimed their status with obsidian blades and projectile points made of various kinds of stone, part of the construction of an ethos of political leadership grounded in or legitimized through the proclamation of a warrior identity by those elites (Pollard, 2008; Pollard and Cahue, 1999; Pereira, n.d.). As demonstrated by Pollard and Cahue (1999) and Pereira (n.d.), in the Late Postclassic period the association of such objects with males no longer existed at provincial administrative centers. 255

PAGE 256

interpretation somewhat problematic, I propose that this rich burial is that of a male given the inclusion of projectile points and bezotes. The important point that can be drawn from these data is that at the capital males as well as females were buried with objects composed of materials there were, according to research into state control over precious metals and long-distance trade goods (discussed in Chapter 3), controlled by the highest levels of the nobility. There were, of course, differences in the specific objects buried with males and females. For example, bezotes composed of obsidian, turquoise, and precious metals were associated with males, but pendants containing turquoise or needles made of copper were associated with females. Most importantly, however, the documentary and archaeological evidence of mortuary contexts at Tzintzuntzan indicates that items of adornment made from materials that the king or upper levels of the state administration controlled were not simply feminine goods. Rather, they were goods controlled by the king and used by the king to proclaim a relationship between himself and others, including the women he married off to subordinate lords and through these women the subordinate lords themselves. If objects composed of state-controlled resources were not limited to feminine goods, then, the disparity between the males and females buried at Urichu in relation to the presence and absence of such objects is not explained by such resources as possessing masculine or feminine qualities. The absence of such qualities being intrinsic to the raw materials indicates that there must be an alternate explanation for the disparity between the male and female burials at Urichu. I suggest that this disparity is the result of kingly agency in efforts to materially proclaim a connection with the deceased females, a connection that was not being proclaimed in relation to the males of the Urichu. In my view this is the material equivalent of the lords denial of establishing a true marriage with a bride from the kings palace, at least as his speech is 256

PAGE 257

represented by agents of the king in the RM (see Chapter 7), and which would have by default signaled that the kings relation with the woman had not been severed by marriage. By giving the objects composed of materials controlled by his government to some of the women of Urichu, and by allowing these objects to be buried with the women if not providing objects specifically for their burial, the king was demonstrating a tie between himself and the women. This tie would have accompanied or likely superceded the material link with the women that the male lords of the site were able to muster. As indicated by the poverty of the male burials at Urichu, even the male officeholder (as indicated by his bezote) at the site, these males were not able to ensure that they would be accompanied by objects composed of state-controlled resources in death. The richness of the female burials at Urichu, then, can be explained as the continuing proclamation of the kings relationship to these women of Urichu, in essence materially framing that relation, making it public, and likely producing memories among the witnesses of the acts of interment that would affect how the deceased would be remembered, and potentially how any living women, especially those that had married in to the local noble family, would be perceived. In sum, the evidence is strong that the king successfully encompassed the local nobility of Urichu through marriage exchange, and it is probable that he was able to do the same with respect to the nobility of Erongarcuaro. The king, through the gift of royal women, their labor, and the valuables that accompanied them, constructed the local nobility as nobles and caciques, but as he did so he constructed them as subordinate and dependent upon him for their legitimacy. The wealth objects that constituted their legitimacy were in fact the distributed personhood of the king; their legitimacy directly (through the women who married into these sites) indexed the king and was in the end his legitimacy. In this way they were reduced to aspects of his social person, encompassed as parts in relation to the whole that the king embodied. 257

PAGE 258

Production of Lapidary Objects in the Tarascan Core Archaeological data concerning the production and exchange of insignia of office between the king and subordinate lords suggests that this relationship was not materially constituted as one in which the king would have been the only referent of those insignia, as implied in the RM and discussed in Chapter 7. There is evidence that lapidary production was taking place at Erongarcuaro under the supervision of the local lords there. This production indicates that other signficatory relationships, in addition to the indexical link between the insignia and kings, could have impacted the intersubjective perception of the place of various lords in the organization of the kingdom. The lords of Erongarcuaro could have used the lapidary objects so produced to attempt to engage subordinate lords of their own in practices that would have constituted a claim to an encompassing status of their own. The lack of any evidence for lapidary production at Urichu, on the other hand, suggests that the exchange of insignia derived from the king was more consistent with the representation in the RM, although even here there is some evidence that suggests that this practice did not transpire exactly as the typified representation contained in the RM indicates. Lapidary Production at Erongarcuaro and its Implications At Erongarcuaro, numerous artifacts that indicate the presence of obsidian lapidary production during the Late Postclassic were recovered in the 2005 field season. Here I describe the evidence for this production, examine the other contextual associations of this production as they potentially affected the perception of the items produced at the site, and discuss the wider implications of lapidary production at Erongarcuaro. The primary evidence for lapidary production is in the form of ground pieces of colored obsidian, predominantly green but also including red-black (Rebnegger, n.d., further study of the obsidian artifacts and comparison to other Tarascan sites is currently ongoing by Karin 258

PAGE 259

Rebnegger, a doctoral candidate at Michigan State University under Dr. Helen Pollard). The green ground pieces are all cylindrical, and ground around the outside (see figure 8-1). Some have one end pecked and ground, while others are fractured at both ends. Of particular interest is a piece that can be described as resembling two offset, ground cylindrical pieces joined together and with a ridge at their convergence (figure 8-1, center). In general form these pieces resemble the end of artifacts used for drilling and pecking (reamers) as described by Otis-Charlton (1993) in her study of lapidary production at the Aztec city-state of Otumba. The shorter pieces that are fractured on both ends could also represent the additional production of discoid beads from drill bits, also as revealed in Otis-Charltons (1993) study. The form of the double cylinder, however, seems to indicate that it was produced by drilling out the middle of an ear spool from two directions; it would be what was extracted from that drilled hole as the doughnut hole. Discoid beads could have been produced from this object as well. 259

PAGE 260

Figure 8-1. Fragments of obsidian drills (reamers) and a hollow drill core that indicate lapidary production at Erongarcuaro. Photograph by Karin Rebnegger and Chris Valvano. Secondary evidence for lapidary production comes from the prominence of multiple colors of obsidian artifacts within the total obsidian sample in this elite part of Erongarcuaro. As defined by Pollard (1993, p. 43) in her survey of the Tarascan capital of Tzintzuntzan, lapidary production sites can be identified through a high proportion of green, red/black, and clear (or clear with only light gray/black striations) obsidian in addition to the gray and gray/black obsidian that is ubiquitous at Late Postclassic Tarascan sites. The presence of a wide range of colors of obsidian is likely a result of the desire to make jewelry items of a wide range of colors. Especially prominent in lapidary workshops are green obsidian artifacts, likely due to a Mesoamerican tradition of valuing green or blue/green objects highly (e.g., Kirchhoff, 1952). The percentages of various colors of obsidian in levels assigned to the Tariacuri phase (1350260

PAGE 261

1520 A.D.) from the excavations at Erongarcuaro in the northern, elite area of the site (see Chapter 2), are given in table 8-1. Table 8-1. Counts and percentage of total sample of obsidian artifacts from Erongarcuaro by according to color. Percentages do not add up to 100% because 5 artifacts were unclassified, and are not included in the chart. Color of Obsidian Count Percent Black 12 0.31% Very Dark Gray 124 3.16% Dark Gray 1922 49.04% Gray 1348 34.40% Clear with Dark Gray 289 7.37% Almost Clear 31 0.79% Red-black 33 0.84% Brownish-clear 38 0.97% Dark Greenish Gray 103 2.63% Dark Olive Green 14 0.36% Total 3919 99.87% In addition to the artifacts recovered in excavation that are evidence of drilling lapidary objects, a local resident was kind enough to show the team of archaeologists for the 2005 field season objects that he had collected at Erongarcuaro that are clearly specific evidence for bezote production at Erongarcuaro. The locations in which he claimed to have found these objects are close to where the archaeologically provenienced indicators of lapidary production were subsequently recovered in the course of the 2005 field season (see figure 8-2). This residents collection demonstrates a production sequence for bezotes from rough blank to near-finished product. Interestingly, all of his bezote pieces were gray/black, in contrast to the green drill bits and doughnut hole recovered in excavations. 261

PAGE 262

Figure 8-2. Map of the survey units of Erongarcuaro, indicating those units in which a local informant claimed to have found bezotes in various stages of production (shaded in light gray). The units in which objects indicative of lapidary production were recovered in excavations are shaded in darker gray. The evidence of bezote production at Erongarcuaro contradicts the implied ideal biographies (Gaitn Amman, 2005; Kopytoff, 1986) of bezotes and other insignia as represented in the RM (see Chapter 7). The ideal biography of bezotes and other lapidary products contained in the RM (1956, pp. 203-205) states that the insignia of office were given to subordinate lords upon their accession to the lordship, and upon their death the insignia were returned to the king. This would have implied, in turn, that the king exercised tight control over at least the distribution of the insignia of office, of which bezotes were a particularly important 262

PAGE 263

category (see Chapter 5). The production of lapidary objects at Erongarcuaro at the very least expanded the significatory relationships of the objects so produced, and the indexical link between the producers at Erongarcuaro and the objects that they made could have potentially been emphasized over the other relations that involved the king. Following from the discussion of technologies of enchantment (Gell 1992) introduced in Chapter 5, particularly difficult productive technologies can at least potentially index the producer over other social actors who have had or will have played a role in the acquisition, transport, and exchange of objects. Through the difficulty of the production process and the high degree of skill necessary for the production of any item (see Otis-Charlton [1993] on the production process and the difficulties therein for obsidian lapidary production), the objects more immediately confront observers with the painstaking process required for their production and therefore the great skill needed. Due to the ability of these difficult objects to confront observers with the process of their making, they more directly index their producers or the patrons of their producers who have allowed them the time and support necessary to hone their craft. In relation to this issue of producers and patronage, the evidence for lapidary production at Erongarcuaro comes from an area of the site that was an elite residential area (see Chapter 2 for a full explanation of why this area is believed to have been an elite residential area). Many of the ground pieces recovered in excavations, moreover, were recovered in close proximity (within a few meters at most) to a hearth that was excavated in ER-02 and which dated to the end of the Late Postclassic (see Chapter 2). The remaining ground pieces were recovered in ER-03, which is directly down slope from ER-02, the area in which the evidence of an elite residence was excavated. Therefore it is reasonable to conclude that these artifacts eroded down the hill slope from ER-02. To that extent, the close proximity between the evidence for lapidary production 263

PAGE 264

and the elite residential area at the site fits well with expectations of attached specializations (Brumfiel and Earle, 1987; also Costin, 2001; see Chapter 5). In fact, the provenience of most of the objects in ER-02, in such proximity to a hearth, suggests that this specialization might even have been embedded within the noble family itself (Ames, 1995; also Inomata, 2001). Thus as patrons of lapidary production, the nobles not only likely controlled the strictly economic capacity to control production and distribution, but (perhaps just as importantly) the nobles were potentially, and in certain contexts, able to appropriate and emphasize the significatory relationships in which those products were embedded. By appropriating the signficatory relationships of the objects produced at Erongarcuaro, the nobles of that site could have potentially used these objects to further their own interests, resist overt and totalizing encompassment by the king, and/or give the objects to lesser nobles as acts manifesting their own encompassing nature. I return to the likelihood of these multiple possibilities below and in Chapter 9, after discussing all of the significatory relations of the lapidary objects and the contexts of their production. Wider Contextual Associations of the Lapidary Production at Erongarcuaro The wider signficatory associations of the producers and production context at Erongarcuaro suggest, however, that this production was not completely severed from its significatory relation with the king. Other relations that were established between the king, the raw material, and the producers themselves negated the ability of the nobles of Erongarcuaro to unambiguously signify themselves and establish their own autonomous legitimacy through sponsoring or engaging in lapidary production. The obsidian utilized in the lapidary production at Erongarcuaro most likely came from the king, as it comes from either sources outside the boundaries of the Tarascan kingdom (thus likely controlled by the king through his long-distance merchants) or the mines at the Ucareo/ 264

PAGE 265

Zinapcuaro source complex, the obsidian from which was apparently appropriated by the state/royal dynasty (Gorenstein and Pollard, 1983; Pollard, 2003c; also Healan, 1997). A sample of green obsidian shatter flakes associated with the obsidian drills/doughnut hole recovered at Erongarcuaro was analyzed by x-ray fluorescence and neutron activation under the supervision of Dr. Michael Glascock of the Missouri University Research Reactor. The majority of the shatter flakes was found to be from the La Primavera/Huaxtla obsidian source in Jalisco (Glascock, 2006; see table 8-2). Also represented in this sample of green obsidian associated with the lapidary production were the Pnjamo obsidian source in Guanajuato and the Cerro de la Bola source in Quertaro. All of these sources were either outside or at the very margins of Tarascan controlled territory (Rebnegger, n.d.). Therefore obsidian imported from these sources was likely controlled by the kings long distance merchants. The small sample of red/black obsidian from the elite area of Erongarcuaro that was sourced using the same techniques was dominated by obsidian from the Ucareo/Zinapcuaro source complex, and obsidian from this source complex flowed predominantly through the state administrative system (Pollard, 2003c). Therefore, the obsidian used in lapidary production at Erongarcuaro was most likely procured by the lords of Erongarcuaro via their relationship with the king, and this productive capacity was in an important way constituted by the kings act of giving or supplying the lords of Erongarcuaro with the materials necessary for its realization. 9 It is likely, then, that the king and royal dynasty must have known to what ends it would be used and at least tacitly approved of such production and allowed for its existence. 9 Although this was not necessarily the case for clear obsidian, which is known to have been acquired at the Zinparo source in north central Michoacn (see Appendix A), which was closer to Erongarcuaro than the Ucareo/ Zinapcuaro source and well within the boundaries of the Tarascan kingdom. Furthermore, evidence from non-elite areas of sites within the Ptzcuaro Basin suggests that gray/black obsidian from this source flowed through non-state distribution networks, probably including market-based exchange. See Pollard and Vogel 1994; Pollard 2003a on the distribution of Zinparo obsidian at sites in the Ptzcuaro Basin. Research in the Zinparo source area of obsidian production has been conducted by CEMCA; see especially Darras (1999). 265

PAGE 266

Table 8-2. Count and percentages of obsidian sorted by color and source, as determined by INAA and XRF. Color La Primavera/ Huaxtla Cerro de la Bola Pnjamo Ucareo/ Zinapcuaro Cerro Varal/Zinparo Sum Green 10 (66.67%) 3 (20.00%) 2 (13.33%) 0 (0%) 0 (0%) 15 Red/Black 0 0 0 4 (80.00%) 1 (20.00%) 5 To the extent that the production of lapidary objects at Erongarcuaro took place under the supervision of the local noble family or even embedded within it, the relation of that family with the Uacusecha royal dynasty would have been of crucial importance to the community of abducting witnesses. If the members of the local noble family of Erongarcuaro had been encompassed through the means of marriage exchange (see above), this would have established the producers themselves as encompassed by the royal dynasty. In other words, the relation of encompassment established through marriage relations could have been intersubjectively extended to the identity of the nobles of Erongarcuaro as lapidary producers or sponsors. Furthermore, the possible role of the discursive framing of the locally produced lapidary objects should not be overlooked. Other sate agents, potentially including the king but most likely including the traveling priests (the curitiecha; see Chapter 5), could have come to Erongarcuaro and appropriated some of these objects and deployed them in the name of the king. Even outside of outright appropriation, such state agents could have reiterated the kings role in the decision to endow a particular lord with lapidary objects included among the insignia of office as well as the role of the ancestors of the king in instituting local offices in the first place (see Chapters 6 and 7). These discursive practices, if realized at Erongarcuaro, should have mitigated against the significatory ambiguity established with the advent of local production of such important and powerful objects. 266

PAGE 267

To this extent, it should be noted that among the collection of ceramic objects used in rituals and ceremonies, including pipes and perhaps mini-bowls, recovered at Erongarcuaro there is a high frequency of sherds and fragments composed of a non-local paste possibly linked to state sponsored production at Ihuatzio. The results of efforts to chemically source ceramic objects and tie the resultant data to the paste descriptions developed by Pollard at Tzintzuntzan and Urichu are somewhat preliminary and have at times produced contradictory results (discussed in greater detail below), but at present it appears that one paste, Tecolote Orange, probably came from a southerly or southeastern location within the Basin (Hirshman et al., 2005; Pollard 2003c). Pollard (2003c) has suggested that pastes shown to be from the southeastern portion of the Lake Ptzcuaro Basin might have been utilized by craft specialists sponsored by state institutions at either Ptzcuaro or more likely Ihuatzio. The sourcing data and their relation to the paste classifications developed by Pollard is subject to revision and refinement, however, and should be regarded with caution. With that caveat in mind, the contrasts between the percentages of Tecolote Orange fragments of mini-bowls and especially pipes and Tecolote Orange sherds in the total ceramic corpus is striking. Among pipe fragments recovered in fields ER-02, ER-03, and ER-05, 18.5% (n=10/54) fragments were Tecolote Orange. Among the 7 sherds that could be classified as fragments of mini-bowls, 2 were Tecolote Orange (28.57%). Among sherds that were either feasting forms (see Chapter 5), showed resist decoration, or both, only 1.41% (n=1/82) were Tecolote Orange. If mini-bowls were indeed associated with ritual, as Pollard (1993, pp. 203-204) suggests, and they are grouped with the pipes and removed from the sample of feasting associated ceramic vessels, the difference between the two proportions is statistically significant at a 99.9% confidence level (see Appendix B for p-values of tests of difference). If, on the other hand, the mini-bowls were used more for feasting purposes and not 267

PAGE 268

necessarily for the types of communal rituals at which objects were discursively framed, the difference between the percentage of Tecolote Orange paste sherds in the pipe sample and the feasting wares sample remains statistically significant at the 99% confidence level. The contrast with the overall ceramic corpus of the elite area of Erongarcuaro is striking, in which only 0.07% (n4/5833) were Tecolote Orange. This difference between the proportion of Tecolote Orange pipes and the proportion of Tecolote Orange sherds statistically significant at the 99.99% level. This evidence indicates that pipes in particular, but perhaps also mini-bowls, found their way to Erongarcuaros elite area through different processes than feasting vessels and even more so than ceramic objects in general. Given the prominence of Tecolote Orange pipes and other ritual objects in the elite area of Erongarcuaro, I suggest that this prominence is due to the presence of representatives of the royal dynasty at communal rituals and observances at Erongarcuaro. The representatives of the royal dynasty should have framed the objects in question by emphasizing the kings role in producing and exchanging the objects involved in any exchange between the king and the lords of Erongarcuaro. The inclusion of representatives of the king in local ritual and ceremonial occasions would have also had the effect of demonstrating the relative statuses of, and relationship between, the king and the local nobles. It should be noted, however, that no evidence of a ycata (see above and Chapter 5) was found at Erongarcuaro during a complete survey of the site. The absence of evidence does not definitively indicate that ycatas were completely absent at Erongarcuaro in the prehispanic period, however; a Franciscan church and convent was established in Erongarcuaro in the 1560s, and prehispanic structures were perhaps demolished in order to provide stone for the 268

PAGE 269

churchs construction. 10 If Erongarcuaro did indeed lack a ycata in the prehispanic era, the implications of this absence are intriguing. Because ycatas were connected with the state religion, ycatas are typically present at Tarascan administrative sites, and Area 5 at Urichu contains a ycata. The absence of a ycata at Erongarcuaro would make the site a rarity among Tarascan administrative centers, and perhaps indicate that it was not an important ritual center or alternatively that Erongarcuaro was home to a religious tradition not entirely defined by the canons and doctrine of the state religion. To some extent, the absence of a ycata at Erongarcuaro would have meant that there was no permanent local reminder of the Uacusechas claim to be the encompassing religious as well as secular authority. Without a ycata to claim the contrary and implicate Curicaueri and the king by its very existence, perhaps that absence gave the lords of Erongarcuaro some intersubjective leeway as they attempted to realize self-referential acts of authority and leadership. Bezotes, and Their Absence, at Urichu At Urichu (Pollard and Cahue, 1999), there is mortuary evidence that the exchange of bezotes between the king and subordinate lords as represented in the RM masks a certain degree of variability in the actual realization of those practices. This variability does not necessarily suggest that the lords of Urichu were not encompassed via the exchange of bezotes, but merely that the practices that led to such encompassment were not, or not always, realized exactly as they were portrayed in the RM. At Urichu, no evidence for any kind for lapidary production was recovered in a total surface survey and extensive excavations in two elite areas of the site, including the area in which the noble family of Urichu apparently resided. I assume, therefore, 10 A document in the Archivo Histrico de la Ciudad de Ptzcuaro, summarized in Martnez Baracs and Espinosa Morales (1999, p. 158-159), describes a suit brought against the civil authorities of Erongarcuaro in 1588 for forcing inhabitants of nearby towns to work at the church and convent. In a footnote Martnez Baracs and Espinosa Morales (1999, p. 158n40) cite documentary evidence for the existence and rapid construction of the church and convent 269

PAGE 270

that the lords of Urichu received their insignia of office from either the royal dynasty or the lords of Erongarcuaro, to whom they were subordinate in at least the tributary hierarchy (see Chapter 2). Given the apparently tight control over insignia of office as presented in the RM and the fact that their production at Erongarcuaro was likely either attached to, or embedded within, the nobility there (see above) I believe it is reasonable to assume that insignia of office were not available in the marketplaces. As I discuss below, however, the available ceramic evidence concerning interaction between Erongarcuaro and Urichu suggests that such interactions were sparse, and so the more likely source of insignia of office acquired by the lords of Urichu was the royal dynasty and the known lapidary workshops at Tzintzuntzan (Pollard, 1993, p. 43). Burial 10 at Urichu, that of a male, included a bezote (in addition to a single polychrome spouted vessel). The interment of the bezote with burial 10 suggests numerous possibilities. This act of interment was possibly in direct defiance of a well-established and enforced practice (at least as it is presented in the RM), or it may indicate that this particular lord was permitted to keep and be buried with his insignia of noble status, or it is possible that noblemen possessed multiple jewelry objects and only some were in fact returned to the king. With respect to these possibilities, bezotes are relatively well-represented in mortuary contexts from various Tarascan administrative sites (e.g. Macas Goytia, 1996, p. 104), and perhaps the return of bezotes as described in the RM was more of an ideal representation, i.e., the extension of a principle (royal encompassment) to representations of a field of practice in which the actual nature of the practice would not have actually reflected the operation of that principle. It is apparent, in the end, that a widely enforced practice of returning all such objects to the king, as it is represented in the RM, did not in fact exist. 270

PAGE 271

In contrast to burial 10, burial 2 exhibited abrasions on the front of the mandible, indicating that for some time during his life this individual wore a bezote (Cahue, Sauer, and Pollard, 1998). Because there was no bezote interred with the body, it is possible that this bezote was returned to the king in the manner described in the RM. It is also possible that the bezote was removed during the individuals lifetime, however, as punishment for disobedience in a manner akin to the removal of bezotes in the RM as an act of the general divestment of status and legitimacy (e.g., RM, 1956, pp. 32-33). The complete lack of grave goods contained in this burial is an interesting side note with respect to this latter possibility, although there were other adult male burials without any grave goods and so this absence in burial 2 is well within the range of variation. It is also possible that this individual survived into the early colonial period, when lords throughout Mesoamerica were forced by Spanish colonial authorities to relinquish these insignia. 11 The data from Urichu (with a sample of two) are difficult to interpret in relation to the supposed practice of returning bezotes and other jewelry to the king, as described in the RM. Viewed in light of the more widespread evidence for the presence of bezotes in burials, perhaps the evidence that the individual of burial 2 wore a bezote in his lifetime suggests that the scenario in which this individual was stripped of his bezote (and status) is more plausible than the return of the bezote to the king. Of course, with better temporal control and a larger sample a pattern might emerge, but at this point in time the apparent variability of practices regarding the interment of bezotes with local lords must simply be noted, along with the fact that the representation of the return of bezotes to the king in the RM was likely an attempt to claim 11 For example, Sahagn (1950-1969) mentions drooling lords who went about drooling because they had a hole in their lip where the bezote had been before the Spanish authorities banned wearing bezotes. 271

PAGE 272

and/or explain the constitution of the relation of encompassment that should have otherwise pertained between the king and subordinate lords. Lordly Feasting and Communal Rituals: Patterns of Ceramic Acquisition at Erongarcuaro and Urichu As discussed in Chapter 7, lords in the Tarascan kingdom were required to provide food and cloth to the commoners during religious festivals. They also must have interacted with key commoners such as elders or lineage heads and possibly also the ocambecha who were in charge of counting and gathering the people for public works (see Chapter 3). Feasting, or the providing of elaborate, labor intensive, and non-quotidian foods and beverages, was likely important to both religious festivals and the micropolitics (Dietler, 2001) of interacting with, and ensuring the loyalty and support of, important community leaders and commoners in general. The king therefore could have attempted to encompass this ability of the local lords through the constitution of, and signification of himself in, the ability of the lords to carry out these feasts. One manner in which this capability would have been constituted by the king was through the gift of royal women as brides who were integral to food production, discussed in Chapter 6 and borne out in the mortuary data at Urichu as discussed above. Additionally, through the regulation of the significatory relationships of the vehicles of food service (i.e., the wares and vessels on and in which food was served), the king could have signified his own persona over and above the role of the local lord in the feasting activity. One way (though certainly not the only way) to consolidate the indexical relations of the ceramic vessels integral to feasting activities would be the control of production and distribution of those ceramic vessels. As the ceramic assemblages of both Urichu and Erongarcuaro demonstrate, however, this was not the case. While there is some evidence that elites at these sites participated in different acquisition networks from the commoners, the evidence does not support the hypothesized role 272

PAGE 273

of the king and state bureaucracy he oversaw in provisioning and controlling the distribution of feasting vessels. Instead, elites likely relied on myriad networks, perhaps including but by no means limited to, access to state-sponsored production, in order to acquire the vessels necessary to execute the duties associated with their status. Feasting Ceramic Vessels: Specialized Forms and Conspicuous Decoration Certain forms of ceramic vessels in the Tarascan corpus were specialized to allow for the dispensation of food and beverages, as I discuss momentarily. Additionally, such vessels, as key and highly visible objects for the realization of lordly feasts, would have been prime vehicles for the communication of the state ideology and the kings central role in the state and cosmos. This communication would have necessitated decoration of those vessels, both in order to convey the capacities of the state for conspicuous consumption as well as more specific symbolic meanings. In what follows I discuss the identification of vessels appropriate for feasting according to their form and decoration. Specialized vessel forms and their role in feasting Specialized vessel forms appropriate to feasting are depicted in the RM. This information identifying the kinds of forms used in feasting is buttressed by the studies of Smith and colleagues (Smith, Wharton, and Olson, 2003), who conclude that certain forms were used in feasting in Aztec-dominated towns in the what is now the Mexican state of Morelos during the Late Postclassic; many of these same forms are common in the corpus of ceramic artifacts in the Lake Ptzcuaro Basin. In the RM, vessels with spouts and handles are depicted as being important vessels for serving beverages, likely including pulque and chocolate. 12 Smith, 12 The RM contains passages in which the drinking of pulque (maguey wine) was practiced at feasts and upon the visits of lords and chiefs. The RM makes no specific reference to chocolate, but it was a drink that was common among the nobility throughout Mesoamerica. Furthermore, the RM (1956, pp. 64-68) contains a passage in which Tariacuris first wife claims to be ill, and a female relative of Tariacuri is depicted with a vessel with a spout and handle. Presumably, then, the container depicted held some sort of medicinal beverage. 273

PAGE 274

Wharton, and Olson (2003) and Smith (2008) include bowls, plates, cups, pitchers, and miniature vessels among those used in feasting. Plates and cups are absent from the sites of Erongarcuaro and Urichu, but spouted vessels are easily viewed as a kind of pitcher (as both are designed for pouring drinks), and miniature bowls and jars are present in the ceramic corpus from these sites. Bowl is perhaps an overly broad category, and I narrow this category to tripod bowls, which would have been used to establish diacritica (Dietler 2001, 2003), or small differences, between certain feasters as possessors of higher/taller bowls which sat above and off of the ground. Collectively, I refer to tripod bowls, miniature bowls and jars, and vessels with spouts and basket handles (basket handles are thin and oval in cross-section; therefore they contrast with lugs, or robust handles that are circular in cross-section and which are common on the sides of large storage vessels; see Pollard 1993, n.d.a) as feasting forms (see figure 5-1) Highly decorated vessels In addition to specialized vessel forms, I assume that highly decorated ceramics would have been preferred for feasting activities. Such decoration would have been essential to the communication of a state ideology (Bray, 2003; Versluis, 1994), enabled practices of conspicuous consumption and the display of wealth (Clark and Blake, 1994), and established another field of diacritica that separated or distinguished feasters (Dietler, 2001, 2003). Feinman, Upham and Lightfoots (1981) elaboration of the production step measure of vessel cost is useful in these regards. The ceramic production process is additive, and with each new stylistic or decorative element (slip, paint, resist decoration, etc.) the cost to produce the vessel in terms of time and labor goes up. For the sake of simplicity, I use the presence or absence of resist decoration as a convenient dividing line in order to discuss vessels likely used in feasting or acts of conspicuous consumption. Proportions of sherds classified according to paste are 274

PAGE 275

consistent in relation to other classifications of high degrees of decorative elaboration, such as the presence of 1 color of paint or more (see Appendix B). Indicators of the Differential Acquisition of Ceramics at Erongarcuaro and Urichu In order to evaluate differences in acquisition patterns between both elite and non-elite areas and between feasting vessels and non-feasting vessels, I make use of the paste classifications developed by Pollard at Tzintzuntzan (Pollard, 1993) and refined at Urichu (Pollard, n.d.a). I assume that these paste differences reflect different producers or communities of producers and that the differences between observed pastes indicate differences in the source material used for the paste, in the paste recipes (including the refinement of the clay and the inclusion of various materials as temper) developed by discrete producers of potting communities, or potentially both. While research attempting to link paste categories with specific clay sources have produced some important results but also some mixed results and should therefore be regarded with caution, I assume here that the pastes do indeed indicate some differences between communities of potters dispersed across the landscape and making use of different clay sources. Therefore the presence and proportions of sherds of different pastes in given samples are assumed to be indicative of acquisition practices of ceramic objects. Past and ongoing research by Pollard, Hirshman and colleagues (Hirshman, 2003; Hirshman et al. 2005; Pollard, 1993: Appendix 2, n.d.c; Pollard et al., 2001) into the organization of ceramic production in the Lake Ptzcuaro Basin reveals a complex interaction between raw materials used, paste recipes (including the addition of various materials, but most commonly ash, as temper; Pollard, n.d.c), and firing conditions, all of which affect the ability to tie observed paste differences to different sources and production locales through means of chemical characterization. Currently, the evidence from two separate chemical characterization studies reveals that sherds from Urichu and Tzintzuntzan can be placed into three compositional groups, 275

PAGE 276

and clay samples from the southwestern portion of the Lake Ptzcuaro Basin are divided between these two groups (Hirshman et al., 2005). While some pastes (Tariacuri Brown, Tariacuri Coarse, Yaguarato Cream, and Yaguarato Coarse) can be tied definitively to clay source areas and therefore are assumed to be produced near those sources, sherds composed of other pastes (Querenda White, Sipiho Gray, Tarerio Cream) are split between the two groups. Finally, Tecolote Orange wares have been found to compose a third group that likely originated from the south or possibly southeastern portion of the basin. Additionally, the relative frequencies of pastes should help to determine whether or not ceramic vessels were acquired through the market system or other practices involving interaction between local lords and state officials (either participation in a prestige-reward system or feasting with state officials). Market distribution is believed to have been the predominant mode of distribution and acquisition for ceramic vessels in general based on studies of production (Hirshman, 2008; Pollard, 2003c) as well as ethnohistorically based studies of marketing activity (Gorenstein and Pollard, 1983). Acquisition practices should also be reflected in profiles of paste frequencies, based on the assumption that marketplaces operate by pooling goods from a variety of sources and producers, with consumers acquiring goods from this pool (Feinman, Blanton, and Kowalewski, 1983; Flannery and Pires-Fereira, 1976; Hirth, 1999; Pires-Fereira, 1976; Smith, 2001, 2003; Winter and Pires-Fereira, 1976). Therefore, I expect a wide variety of pastes to be present in comparable frequencies in given samples, likely with some skewing in favor of locally produced goods due to lower transportation costs (assuming equality in production costs as a result of the lack of any specialized production, as demonstrated by Hirshman [2003]). Locally produced ceramic vessels need not have been acquired through the market system (the market closest to Erongarcuaro and Urichu was in Pareo, on the southern 276

PAGE 277

shore of Lake Ptzcuaro) but could have been acquired through local marketing characterized by direct producer/consumer interaction, with the possibility of acquiring other ceramics through the marketplace should availability or quality dip. Skewing could also be the result of higher outputs of particular pastes due to higher numbers of producers (rather than intensification among a steady or smaller number of producers, which apparently did not occur in the Ptzcuaro Basin; see Hirshman [2003]). Taking these possibilities for skewing into consideration, I assume that the paste proportions of total sherd sample from the different areas of Erongarcuaro and Urichu should be the result of market place and marketing (direct consumer/producer interaction) activity in the acquisition patterns for different classes of people, with skewing toward certain wares evidence of greater availability and/or lower cost, which would make them more attractive to consumers. Therefore, where profiles of specific categories of ceramics, such as feasting vessels, depart from the profiles of all sherds and/or relatively even distributions, explanations outside of acquisition through the marketplace and local marketing should be sought. Ceramic Assemblages from Erongarcuaro and Urichu Profiles of paste frequencies from excavations in the elite area of Erongarcuaro (ER-01, ER-02, and ER-07) and Areas 1, 2, and 5 of Urichu (figure 8-3) suggest that elites at Erongarcuaro and commoners at Urichu (in Area 2) acquired their ceramics predominantly through the market system and local marketing. Percentages of Yaguarato pastes (both Coarse and Cream) and Sipiho Gray are high at Erongarcuaro and Area 5 at Urichu, but this can be explained as the likely result of higher output of wares composed of these pastes combined with their local production which made them cheaper. Area 2 largely conforms to the expectations of acquisition through marketplace and direct marketing exchange with its relatively even percentages and higher frequencies of exclusively local pastes (Tariacuri Brown, Tariacuri Coarse, Yaguarato Coarse, and Yaguarato Cream). 277

PAGE 278

All Sherds by Paste0.00%10.00%20.00%30.00%40.00%50.00%60.00%ER All SherdsU1 All SherdsU2 All SherdsU5 All Sherds Querenda White Sipiho Gray Tarerio Cream Tecolote Orange Tariacuri Brown Tariacuri Coarse Yaguarato Coarse Yaguarato Cream Figure 8-3. Percentages of sherds of dominant pastes from excavations at Erongarcuaro and Urichu. In contrast, the high percentage of Sipiho Gray wares in Area 1 indicates acquisition of this ware outside of the normal marketing observed in the other excavated areas. The differences between the percentage of Sipiho Gray wares in Area 1 and the percentages of the same ware in Urichu Area 5 and Erongarcuaro are both statistically significant at the 99.9% percent confidence levels. This suggests that the elites in Area 1 of Urichu acquired ceramic goods through a range of practices that differed from the means whereby the elites of Erongarcuaro and Area 5 at Urichu acquired ceramic vessels. If it is assumed that the elites of Area 5 and Erongarcuaro acquired ceramics through predominantly marketing activity, the dominance of Sipiho Gray wares at Area 1 at Urichu indicates that elites in this area were acquiring ceramic vessels through means outside of the market system or direct marketing. Given that Sipiho Gray wares appear to have been produced using raw materials from the southwestern part of the Lake Ptzcuaro Basin, the nature of this association could have been the result of various practices and relations with producers. High levels of acquisition from local producers of Sipiho Gray wares need not entail a high degree of specialization and increased output among those potters, but 278

PAGE 279

merely first pick from among those potters. Perhaps they had unique access to a community of producers that could have ranged from attached production in which the elites were patrons of producers (Brumfiel and Earle, 1987) or embedded specialization in which elites were potters themselves (Ames, 1995) to a looser system of preferential access to the products of a group of potters (on the variation of arrangements between producers and consumers in general see Costin [2001]). The ceramic data from Urichu, at which both elite/noble and commoner residential areas were excavated, demonstrate that sherds with resist decoration were not limited to elite areas (see table 8-3). In fact, such sherds were more frequent in the commoner area of Urichu, Area 2. The same can be said of feasting vessel forms, as these are present in Area 2 at Urichu as well (see table 8-3). This suggests that commoners had some access to vessels for feasting and they were not the sole prerogative of the elites. Table 8-3. Number and proportion of resist decorated and feasting form sherds from excavated areas at Urichu and Erongarcuaro. Urichu Area 1 Urichu Area 2 Urichu Area 5 Erongarcuaro Resist Decoration Sherds n=39/2839 (1.37%) N=59/1892 (3.12%) n=23/7562 (0.30%) N=30/5833 (0.51%) Sherds Indicative of Feasting Forms n=38/2839 (1.34%) N=14/1892 (0.74%) n=18/7562 (0.24%) N=59/5833 (1.01%) However, paste classifications of those sherds with resist decoration as well as feasting sherd forms reveal some interesting differences in pastes. It is obvious that elite and commoner acquisition patterns, as indicated by paste, were quite different at Urichu (see figure 8-4). At that site, Sipiho Gray wares dominate the sample of feasting or resist decorated sherds in elite areas, while in the commoner area Tariacuri Brown is the most common paste among a relatively even distribution. This evidence indicates that the elites of Urichu had a close association with the 279

PAGE 280

producers of Sipiho Gray wares, similar to the scenario offered as an explanation for the dominance of Sipiho Gray sherds in the total sample from Area 1. The percentages of feasting or resist decorated sherds separated by paste are shown juxtaposed with the percentages of all sherds by paste in figure 8-4. In this figure, the Tariacuri Brown and Tariacuri Coarse pastes have been combined into a single category, as have Yaguarato Cream and Yaguarato Coarse. This has been done for the sake of comparability, because feasting and resist decorated vessels are more likely to have been produced using fine wares. Furthermore, the differences between the Tariacuri and Yaguarato fine and coarse pastes are minimal, reflecting only a lesser quantity of inclusions and smaller inclusions among the fine-wares, while the clay matrices between the fine and coarse wares are consistent. 280

PAGE 281

Feasting Forms or Negative Decoration By Paste0.00%10.00%20.00%30.00%40.00%50.00%60.00%70.00%80.00%ER Ecc/N.D.U1 Ecc/N.D.U2 Ecc/N.D.U5 Ecc/N.D. QW SG TareCr TecO TariComb YagComb All Sherds by Paste0.00%10.00%20.00%30.00%40.00%50.00%60.00%70.00%80.00%ER All SherdsU1 All SherdsU2 All SherdsU5 All Sherds QW SG TareCr TecO TariComb YagComb Figure 8-4. Percentages of wares within the feasting forms and resist decoration sample and the total sherd sample. I assume that the percentages of pastes within the total sherd samples reflect an absence of significant acquisition of state-produced goods in these areas of Erongarcuaro and Urichu. Even if such provisioning did occur in small amounts, it is likely masked by the large sample sizes (ER n=5833, U1 n=2839, U2 n=1829, U5 n=7562) of sherds that would not have been within the purview of state interests, i.e., sherds from utilitarian and quotidian vessels. This comparison reveals that in the elite areas (1 and 5) of Urichu, Sipiho Gray wares were more common among feasting or resist decorated sherd samples than in the total sherd samples. The difference between the proportion of Sipiho Gray sherds in the feasting and negative decoration sample and 281

PAGE 282

the proportion of Sipiho Gray sherds in the total sample in Area 1 is statistically significant at only the 90% confidence level. The evidence for Area 1 suggests, then, that elites at Urichu possibly altered their acquisition practices when it came to acquiring vessels used for feasting. This could have included acquisition from state-sponsored producers but, given the prominence of Sipiho Gray wares in the total sherd sample, more likely involved a preference for acquiring feasting vessels from those same producers with whom they already enjoyed a close relationship. In Area 5, the percentage of Sipiho Gray feasting or resist decorated sherds is markedly different from the percentage of Sipiho Gray sherds in the total sample. The difference is statistically significant at the 99% confidence level. The significance of this difference is unclear, given the nature of the area as a state religious installation (as indicated by the presence of ycatas in this area) but also as part of the local site and thus presumably overseen the majority of the time by local lords. Therefore the deposition of ceramics at the site could have been due to religious practices in which either the local lords or state religious officials participated most often. The evidence from Area 2, in which a wide range of pastes are represented relatively evenly, suggests that commoners acquired feasting vessels and vessels with resist decoration through the market system (with its pooling of goods from various producers), and through interaction with local producers. Interestingly, there is a relatively high frequency of Tecolote Orange wares in Area 2 in the total sherd sample, but Tecolote Orange sherds are completely absent in the feasting or resist decorated sample. In other words, the residents of Area 2 did not acquire highly decorated sherds or feasting vessels composed of the Tecolote Orange paste. None of the Tecolote Orange sherds from Area 2 at Urichu are decorated with anything more than a slip (i.e., no paint, no resist; see table in Appendix B). Therefore while these commoners 282

PAGE 283

apparently did acquire non-locally produced ceramics, they probably could only afford ceramic vessels that were either locally made and highly decorated or non-locally made and not highly decorated. Moreover, it should be noted that these non-locally made Tecolote Oranges wares, fairly plain as they were, likely would not have been suitable for practices of conspicuous consumption associated with feasting activities. The evidence from the elite area of Erongarcuaro demonstrates more of an even distribution of wares indicative of acquisition through markets and local marketing in the sample of feasting and negative decorated sherds. Sipiho Gray wares are prominent, but occur at a lower frequency than do Yaguarato Cream wares among feasting or resist decorated sherds. This is consistent with the total sherd sample. This suggests that elite acquisition patterns of feasting and resist decorated vessels were not different from the manner in which all kinds of vessels were acquired. The dominance of Yaguarato (both cream and coarse) ware sherds should also be noted, especially when compared to the different areas of Urichu. The prominence of Yaguarato ware sherds among the feasting and resist decorated sherd samples is particularly striking. The difference between the high proportion of Yaguarato ware sherds at Erongarcuaro and Sipiho Gray ware sherds at Urichu could simply be due to the dominance of Sipiho Gray wares at Urichu, but even in the commoner area of Urichu, where Sipiho Gray wares are less frequent, it is Tariacuri wares (both Tariacuri Brown and Tariacuri Coarse) and not Yaguarato wares that make up for this difference. Given that recent investigations by Hirshman (n.d.) have largely confirmed the difference of production regimes between the Tariacuri and Yaguarato wares, as evidenced by the different color and composition of the pastes, perhaps the dominance of Yaguarato wares at Erongarcuaro suggests that Yaguarato wares were produced at 283

PAGE 284

Erongarcuaro or closer to Erongarcuaro than Urichu and therefore were more accessible at the former and than the latter. Finally, while the percentages of Sipiho Gray and Yaguarato ware sherds are dominant among the sample of feasting and resist decorated sherds in the elite area of Erongarcuaro, all pastes are represented and occur in low but consistent frequencies. This is in contrast to all areas of Urichu. In Area 1 of Urichu, percentages of Querenda White and Tarerio Cream ware sherds are low and Tecolote Orange is completely absent among the feasting and resist decorated sherd sample. In Area 2, Tecolote Orange ware sherds are completely absent in the sample of feasting and resist decorated sherds, but percentages of Querenda White and Tarerio Cream ware sherds are higher than in Area 1. In Area 5, Tarerio Cream ware sherds are completely absent from the sample of feasting and resist decorated sherds, while Querenda White and Tecolote Orange ware sherds occur in low frequencies. Therefore while Sipiho Gray and Yaguarato Cream wares are clearly dominant at Erongarcuaro among the feasting and resist decorated sherd samples, the presence and frequencies of sherds of the other pastes indicate that acquisition of such vessels also occurred through the market system. By interpreting the acquisition patterns of the inhabitants of Urichu and Erongarcuaro according to the proportions of different wares in relation to one another at these sites in combination with the extant data relating pastes to clay sources through chemical characterization, it is possible to reach a number of significant conclusions regarding the hypothesis that the material objects necessary for lordly feasting activities would have been an obvious media through which the king could have sought to encompass the agency of the local lords and thereby signify the benevolence of the state through the actions of those local level lords. 284

PAGE 285

The first conclusion is that this implication of the king in acts of localized feasting was definitively not established through the provisioning (i.e., gift-giving) of feasting vessels to the lords of Erongarcuaro and Urichu. The evidence at Urichu, in which Sipiho Gray wares are proportionally dominant within the total sherd sample and even more dominant among feasting and resist decorated sherds in Area 1, suggests that in their efforts to acquire vessels suitable for feasting the elites at Urichu merely made use of their preferential access to local producers of these wares to acquire what they desired or needed. At Erongarcuaro, the evidence similarly indicates that state provisioning did not occur. Sherds from the elite area of Erongarcuaro display a wide range of pastes, with two dominant pastes and the rest relatively even in their frequencies. It is possible that Tecolote Orange wares were provisioned by the state in low quantities and this is overshadowed by the large sample and relatively even profile of wares at the site, but the data do not indicate that state provisioning was at all prominent. Moreover, the most frequent paste (Yaguarato Cream and Coarse) was one that, according to chemical characterization studies and its marked preeminence at Erongarcuaro compared to Urichu, was almost exclusively produced locally with respect to the southwest portion of the Lake Ptzcuaro Basin and possibly even at Erongarcuaro itself. Overall, then, the evidence from Erongarcuaro suggests at best the existence of an acquisition practices that included preferential access to local producers, supplemented by acquisition through the market system, and least of all possible provisioning by state-level officials (although this was not necessarily very likely). At any rate, current evidence does not support anything approaching near-exclusivity of state provisioning of feasting vessels that would have been widely known and which would have led witnesses to abduct the role of the king in the realization of the feast. 285

PAGE 286

Secondly, there does not seem to have been a very strong relationship between the elites of Erongarcuaro and Urichu, as indicated by the ceramic assemblages of the two sites. The reliance by the nobility of Area 1 of Urichu on their contacts with producers of Sipiho Gray wares suggests that they had no need for the wares produced at Erongarcuaro or for the involvement of the nobles of Erongarcuaro. The prominence of Yaguarato wares at Area 5 in Urichu is somewhat intriguing, however, and could indicate the participation of nobles from Erongarcuaro in rituals and religious festivals at that site. In this regard it is potentially significant that the only ycatas located at either site are located in this area (but see discussion above concerning the existence of ycatas at Erongarcuaro), and perhaps state rituals of the kind that called for the attendance of nobles from both sites (e.g., rituals in which Curicaueri, the patron deity of the royal dynasty and the state, was worshipped) were held at Area 5 in Urichu. Lastly, the possibility remains that stylistic components of feasting vessels were the primary significatory media, rather than the exclusivity of exchange or acquisition networks, through which the king implicated himself in lordly local feasting. Versluis (1994), on the issue of the stylistic composition of the decoration of vessels from Urichu, concluded that the Tarascan state ideology was in the process of consolidation and was not yet uniform in its materialization. Moreover, if the state was to be signified by the new stylistic cannons of the Tarascan polychromes, I would expect that at least some specialized production must have occurred at one or more of the three Uacusecha centers in which a recognized school could develop and be copied. At this point, there is no archaeological evidence suggestive of such state-sponsored production, although the RM (1956, p. 178) states that a number of craft specialists, including potters, were under the supervision of the king. The wide range of pastes and recipes for ceramic production, as well as the evidence that production was, at least in the southwest portion 286

PAGE 287

of the basin, relatively unstandardized (Hirshman 2003), is at least suggestive that the stylistic elements and iconography of the Tarascan polychromes was more of a general and shared phenomenon, not developing in any single locale. This, however, could be subject to revision, particularly if evidence of state-sponsored craft specialists in the eastern portion of the basin is brought to light. Conclusion The evidence discussed in this chapter demonstrates that the practices through which the king claimed (in the RM; see especially Chapter 7, also Chapter 5) to have encompassed the lesser nobility were not consistently realized at Erongarcuaro or Urichu. This is not to say that relations of encompassment were not materially realized at Erongarcuaro and Urichu, however. I suggest that while the relations of encompassment were likely more tenuous and open to slippage than the representations in the RM indicate, the relations with material objects would have led to the abduction of the kings primary, and therefore encompassing, role in the constitution of the lords of these two sites. At Urichu, there is evidence that the marriage relationships between the local lords and the king as described in the ethnohistoric record were indeed realized. Therefore the lords of Urichu could not have consummated a mutual exchange alliance and could not have appropriated the labor of their royal brides. The evidence concerning the indexical relations of the bezotes from Urichu is somewhat equivocal; bezotes almost assuredly came from the king, but at least one was not returned to the king (as the RM claims it should have been; see Chapter 7). It is entirely likely, however, that the status of the bezote as a gift from the king would have been known by the populace at large and that the local lords would have been appropriately perceived as encompassed by the king. The ceramic evidence for feasting reveals that the king did not provision the lords of Urichu with the objects necessary to hold feasts. Therefore the 287

PAGE 288

demonstration of the generosity and benevolence toward the populace that was the central function of those feasts most likely did not directly implicate the kings role; the lords benevolence was not directly transformed into the kings benevolence through the construction of feasting vessels as the distributed personhood of the king. The king and state likely relied on the signficatory extension of the encompassment of the local lord of Urichu through other means, such as royal intermarriage and the prestation of insignia of office, to effect such a transformation. At Erongarcuaro, there is no direct evidence that women who married into the noble family from the royal dynasty would have effected a relation of encompassment as appears to have pertained between the king and the noblemen of Urichu. However, there is evidence that there was a close relationship of marriage exchange between the local noble family and the royal dynasty, and it would have been improbable that these relations had been symmetrically constituted. Therefore I suggest that the king was indeed able to encompass the noblemen of Erongarcuaro through marriage exchange. The king likely relied on this relation of marriage exchange to carry the lions share of the significatory weight necessary for a relation of encompassment to have been abducted by witnesses. This is borne out by the evidence for lapidary production at Erongarcuaro. Ostensibly, the noblemen there could have laid claim to the lapidary objects produced under their auspices, but there is also evidence that the king was involved in this production and the appropriation of its products through a range of practices. In such contexts of at least potential ambiguity the abduction of agency carried out by witnesses likely tipped in favor of the king, particularly if he was able to implicate himself in the public rituals whereby the lord came into official possession of his bezote and the other insignia of office. The implication of the king in the 288

PAGE 289

perception of such events is indicated by the prominence of Tecolote Orange pipes and mini-bowls in comparison to the total sherd sample and even the sherds related to lordly feasting at that site. Such objects indicate the presence of state agents who, present at such events or others like them, must have seized the opportunity to discursively frame the objects at the center of these events and the relations they were integral in consummating. Furthermore, I address the issue of tipping when abducting agency, in effect giving the king the benefit of the doubt, in the concluding chapter. This practice, if the kings grandeur could ensure it, also likely helped eliminate some of the significatory ambiguity that resulted from many hands being involved in the production of the bezotes and other lapidary objects at Erongarcuaro. This should not be taken to mean that the lapidary production at Erongarcuaro was meaningless, useless to the lords of Erongarcuaro if they could not use the objects so produced to actively realize self-referential acts of speaking and ruling and in the process to resist encompassment by the king. Rather, the most plausible impact of the lapidary production at Erongarcuaro is that the nobles of that site could have used the objects produced to claim an encompassing status of their own. I discuss this possibility, and its implications, in the concluding chapter. Lastly, the relationship between Erongarcuaro and Urichu, linked in a superior/ subordinate relationship in the tributary hierarchy, was not also manifested as a relationship of encompassment. There is no evidence of marital exchange between the two sites. The lords of Urichu likely received their insignia of office from the king and not the lords of Erongarcuaro. The lords of Urichu appear to have been autonomous to the lords of Erongarcuaro with regards to their ability to acquire ceramic vessels used for feasting. Therefore while Erongarcuaro might have held a higher place in local economic (productive and distributive) networks and the 289

PAGE 290

lords of that site might have enjoyed a higher status or more prestige, particularly owing to their sponsorship of lapidary production, they do not appear to have influenced the lords of Urichu politically nor directly engaged them in practices that would result in the encompassment of the lords of Urichu by the lords of Erongarcuaro. 290

PAGE 291

CHAPTER 9 CONCLUSION: ENCOMPASSMENT, THE TARASCAN CORE, AND IMPLICATIONS FOR THE STUDY OF THE TARASCAN KINGDOM AND ITS DEVELOPMENT Introduction In this final chapter I summarize the interpretations of the representations of encompassment explicated in the Relacin de Michoacn (RM) and review the evidence for the realization of material practices that would have effected the encompassment of the local lords of Erongarcuaro and Urichu. Specifically with regard to those material practices I discuss how the objects that were produced at Erongarcuaro would have been suitable for use in certain interactions with certain actors. I suggest that the lords of Erongarcuaro could not have used such objects to resist being encompassed by the king due to the kings investment of those objects with his agency via certain practices. However, I also suggest that the lords of Erongarcuaro could have used such objects in order to engage subordinate lords in relations of encompassment. Interpreting exactly how such relations would have been perceived by the different actors that they involved requires a discussion of the potential for different abductions to be made regarding the roles of the other actors in these relations. Additionally, I discuss the implications of the specific archaeological case study for the study of the development of the Tarascan kingdom. I believe that the evidence for lapidary production at Erongarcuaro indicates a royal concession, and that such a concession would not have been optimal for the king. Therefore I suggest that this arrangement signals some degree of power sharing in the earliest stages of the consolidation of Uacusecha power. This conclusion runs contrary to Gorenstein and Pollards (1983) model of the consolidation of the Tarascan core prior to state expansion but is in agreement with Pollards (1993, p. 88) discussion of state formation and extra-basin raids by the Uacusecha. More generally, I discuss the wide-ranging implications of the formulation of a model of intersubjective encompassment as appropriate to 291

PAGE 292

the nature of the kings power within Tarascan society. The fact that kingly encompassment involved the abduction of agency in relation to total webs of relations and practices is particularly relevant when considering that relations between the king and lords in the core and between the king and lords in more provincial areas were likely interconnected and impacted one another. Finally, I discuss how the encompassing nature of the Tarascan king and the historical development of that nature might be investigated through further research, and why the manner in which that encompassing nature was constituted requires investigation in multiple contexts throughout the kingdom. Logic and Claims of Encompassment in the Tarascan Kingdom As discussed in Chapters 6 and 7, the king and royal dynasty claimed to have encompassed all of indigenous society in general, and subordinate lords more specifically. These claims were made through the rhetoric of a historical narrative that related the origins of legitimate authority. By using constructions of the past as cosmological authentication (Weiner, 1992; see Chapter 5), the royal dynasty claimed to have been the means by which legitimate authority was produced and then distributed to the subordinate lords, who were transformed into replicas of the king in the process. Additionally, claims of the kings encompassing nature were made in the context of the production of the RM, as the priests and members of the upper nobility used the opportunity they were given to describe indigenous society and politics to frame (Keane, 1997; see Chapter 5) how exactly encompassment worked, or should have worked, in practice. In the representations of interactions between the king and subordinate lords, these informants presented visions of how encompassment was continually achieved in (what was at the time) the very recent past by claiming that certain relations involving the objects at the heart of those interactions were the norm in the prehispanic era. 292

PAGE 293

As the legendary history of the RM reveals, the king and the royal dynasty claimed to have fundamentally reordered the sociocosmic universe, and were themselves embodiments of that newly ordered universe. The two processes were one and the same. In order to reorder the universe the royal dynasty had to subsume within itself its disparate but essential aspects (its elementary categories; see Chapter 6), and as the royal dynasty was transformed so, too, was their ability to unite the wider universe. As part of the reordering of the social universe, the royal dynasty instituted local offices of leadership. These offices, because they supposedly originated with the actions of the royal dynasty, were not continuations of preexisting local power relations, at least not in Tarascan political theory. They were indexes of the kings (and his ancestors) agency across space and down through time. According to this origin narrative, local level lords owed their position and powers to the ancestors of the king. They were but the result of the desire and will of the legendary leaders Hiripan, Tangaxoan, and Hiquingaje that there should have been lords in the towns to settle and pacify the people, and just as importantly of these leaders newfound power and their ability to institute such offices. Through the actions of the Uacusecha, local lords were made, and they were made as replicas of the king. That is, by claiming to have instituted these local offices of leadership, the royal dynasty claimed to have been the historical means by which the subordinate lords were transformed into replicas of the king, in perfect concordance with Turners model of processual encompassment discussed in Chapter 4. The legendary history also managed to maintain a separation between the royal dynasty and the subordinate lords, even as the latter were replicas of the former. This difference was maintained as a permutation and elaboration of the simple difference between prototype and replica discussed in Chapter 4. In any relation of replication, the prototype precedes the replica, 293

PAGE 294

and this primacy is the basis of its higher valuation. The royal dynasty claimed in the RM that not only were they the first kings but they were the first kings precisely because they had invented what it was to be a king. They had encompassed the elementary categories of indigenous society, they were hierarchically superior (in Dumontian terms of their superior valuation; see Chapter 4), and so it was natural that they should hold power. This is in marked contrast to the nature of the subordinate lords, as represented in the RM. The lords were Chichimecs just as the king was; they were individuals who possessed the legitimate authority necessary to rule. However, the members of the royal dynasty were the original Chichimecs who had transformed themselves and in the process produced legitimate authority, grounded in the encompassment of elementary categories, in the first place. The subordinate lords were merely the recipients of the legitimate authority that the royal dynasty had produced; they only received legitimate authority through their relationship with the king and did not produce it themselves. What is as important as the logic of sociocosmic encompassment manifested in the structure of the narrative is the claim in the RM that the narrative was frequently and publicly proclaimed to townspeople throughout the kingdom. Frequent public proclamation of the historical basis of the relationship between the king and the subordinate lords as a whole would have inculcated the people to the states way of seeing the world (following Bourdieu [1999]) and more specifically the place of their own town and lord within it. It was an effort to regulate the semiotic ideology (Keane, 2005) of the people, that is, the way that they would abduct, and abduct hierarchically, the agencies that went into the production of any particular object, exchange act, ritual, or ultimately the cultural order as a whole. The states concern with discursively framing the past is precisely what led it to, again according to the RM, actively produce scenes at which it could tell its own history and therefore explain the historically 294

PAGE 295

constituted relations among its elements. It is no accident of history that a Tarascan priest just happened to be able to recite the history of the royal dynasty to a Franciscan friar in the first decades of the colonial encounter, whether or not that priests narrative was a faithful reproduction of some prehispanic version that may or may not have existed in the first place. The upper nobility of the Tarascan kingdom took their history seriously, and knew that in order for it to be politically useful it needed to be told often and passed on from one generation to the next. The RM, among other things, represented a claim of how the past determined how things were, or should have been, perceived at the moment of its production and in years pervious. In addition to a claim on how the past should have been perceived, the RM was also a claim that practices in the the present, loosely defined, should have operated according to the principle that the king encompassed society in general and subordinate lords in particular. Given the opportunity to express to Spanish civil and religious authorities how things worked in the not-so-distant past before the Spaniards arrived, members of the highest levels of the indigenous political structure claimed kingly encompassment to have been produced through typified exchange relations and the public proclamations that framed those exchanges. In other words these representations in the RM explained how the king, in theoretical practice or practical theory, encompassed the subordinate lords of the realm through actions in the present. The king claimed to be everywhere at once, his personhood extended in the bodies of his subordinates because he had placed his own objectified personhood on, or in the case of royal marriage exchange, next to, those bodies. In order to accomplish this, the exchanges of those objects and women needed to be represented as something less like exchange and more simply akin to transferal or endowment. This is why bezotes and other insignia of office were represented as, firstly, unreciprocated gifts and, secondly, as objects that automatically devolved back to the 295

PAGE 296

king upon the death of the lord. Because Tarascan political theory held that the offices that local lords held were the extensions of the kings agency (or extensions of his ancestors agency, which amounted to the same thing), this principle was reflected in the representation of the objects that objectified those offices as inalienable property (Weiner, 1992; see Chapter 5) of the king. State history and rites of investiture were transformations of the same basic principle of the encompassing, and therefore distributed, nature of the king. The same principle in which the insignia were not exactly given to subordinate lords was also operative in the representation of marriage practices involving local lords and royal women of the kings house (RM, 1956, pp. 208-210; see Chapter 7). These exchanges were described as asymmetric. The royal woman was dowered with exquisite wealth objects. In contrast, the ability of the local lords to reciprocate with anything other than land and labor would have been severely limited. The exchange of the royal women was discursively negated in the representation of the marriage ceremony as well. The lord accepted the woman not as a wife, but rather so that he may tutor or raise her. The total effect would have been to cast the marriage as a non-exchange. The implications of this severely impacted the endeavors of the lord to act in a manner befitting a lord, as the wife was responsible for making food and cloth for the gods and for the people, the distribution of which was one of the most essential duties of the lords. Without a true marriage to bind the wife and her productive capacities to him in the abductions of commoners, however, the lord could not have unambiguously appropriated these fruits of her labor for his own ends. They remained indexes of the womans agency, and through her were indexes of the kings agency. In other words, the will and agency responsible for the realization of the lords ability to rule, the lords ability to distribute goods to the people in acts of lordly beneficence, the fact that those goods would have indexed the bride and therefore also the king, 296

PAGE 297

and finally that the lord should be a representative of the king would have been abducted to be the kings will and not the will of the local lord. Material Practices of Encompassment in the Tarascan Core The archaeological and ethnohistoric evidence pertaining to kingly interactions with the lords of Erongarcuaro and Urichu indicate that relations of encompassment were indeed materially realized through real world practices. The king was able to directly engage the lords of both Erongarcuaro and Urichu in exchange relations that would have led witnesses to abduct the primacy of the kings agency in the constitution of the agencies of the subordinate lords at these two sites. Here I summarize the evidence for materially constituted relations of encompassment with respect to marriage exchange, the indexical relationships of the insignia of office, and the indexical relationships of the ceramic vessels used in feasting. I also discuss how the lapidary objects produced at Erongarcuaro in particular must be interpreted contextually according to how they should have been interpreted contextually by witnesses in the past, and therefore how they might have been used by the nobles of Erongarcuaro. While there is definitive evidence for the encompassment of the lords of Erongarcuaro and Urichu in the form of marriage exchange and investiture ceremonies, the material evidence is perhaps not as totalizing as might be expected in light of the claims of encompassment made in the RM. For example, the ceramic data at Erongarcuaro and Urichu indicates that state provisioning of feasting vessels that would have implicated the role of the king and state in local contexts of lordly beneficence was nonexistent. Moreover, the location of lapidary production at Erongarcuaro under the auspices of the lords there would have introduced new indexical relationships in addition to the significatory relations between insignia of office and the king that, according to the representations of the RM, should have been exclusive. The presence of lapidary production at Erongarcuaro should have been seen by the king and royal dynasty as less 297

PAGE 298

than optimal, if only due to potential problems that might arise in the future as semiotic ideologies might be reworked in order to challenge the kings claims to have monopolized the origins and origination of legitimacy and authority. There was, after all, a reason why members of the upper elite proclaimed the kings exclusive role in the distribution of insignia of office and not a more diffuse arrangement of production and distribution in the RMthey wanted to tie these objects to the king as unambiguously as possible. When attempting to analyze how the lapidary objects produced at Erongarcuaro would have been perceived and, just as importantly, how they could have been used in social action (and the two are interrelated, as I discuss momentarily), it is necessary to consider the complex and contingent ways that different agencies were invested in the objects and could be abducted from them. The lapidary objects were produced by the lords of Erongarcuaro or under their direct auspices, but the king at least tacitly approved of this production and demonstrably provided the lords of Erongarcuaro with the raw materials necessary for its existence (see Chapter 8). The lords of Erongarcuaro themselves were very probably encompassed by the king through relations of marriage exchange. Finally, and as evidenced by the sample of pipes and mini-bowls at Erongarcuaro, the king or other state agents came to Erongarcuaro to discursively frame the objects produced at Erongarcuaro, reminding the populace of the relationship between the office that the objects signified and the role of the king and his ancestors in instituting that office. State agents likely also reminded the lords of Erongarcuaro and the populace at large that the king had provided the raw materials necessary for production, and laid claim to the objects so produced. With the establishment of indexical relationships between the king and the lapidary objects produced at Erongarcuaro through provisioning the raw material necessary, and the likelihood that this production took place with some kind of royal concession or sanction, it 298

PAGE 299

is probable that the king saw the products as his own property, i.e., as objectifications of his own personhood. Therefore the king or royal emissaries could, moreover, have simply appropriated the locally produced insignia, enforcing the claim the king held over them by dint of having sponsored and allowed their production, and used them in the investiture of the local lords of Erongarcuaro and perhaps elsewhere. This scenario appears likely given the ceramic evidence (non-local pipes and mini-bowls) for visitations by state agents. That such scenes were produced at Erongarcuaro suggests that the lords of Erongarcuaro could not have used the lapidary products to endow themselves with the ability to govern self-referentially, without also implying the kings role in constituting and enabling such acts. In short, these objects could not have been used to resist kingly encompassment through investiture ceremonies and the conferral of insignia of office. Lapidary objects could also have potentially been used to dower women from Erongarcuaro that were married off to the king. Such jewelry objects would have more evenly matched the jewelry and wealth object of the kings women who married into Erongarcuaro. Therefore the lords of Erongarcuaro could have achieved a more balanced exchange relationship more akin to alliance than encompassment if they had been able to use lapidary objects to dower the women they married off to the king. I believe that the potential negation of a relation of encompassment through an asymmetric marriage exchange relation was unlikely, however. Here again the kings role in enabling the lapidary production at the site and the likely perception on his part that they belonged to him created a problem for any potential efforts by the local nobility to establish more symmetrical marriage relations. If the king insisted that the locally produced lapidary objects were his, they therefore they would have been unsuitable as return gifts 299

PAGE 300

proffered by the local lords to the king in an effort to circumvent an impending relation of encompassment. In short, a gift cannot be given if it already belongs to the recipient. In the face of the improbability that the lords of Erongarcuaro were able to resist kingly encompassment, the sponsorship of lapidary production on the part of the lords of Erongarcuaro may seem pointless. This was not likely the case, however. The local nobles of Erongarcuaro could have used their sponsorship of lapidary production to their advantage in interactions with lesser nobles. Ultimately, the nobles of Erongarcuaro could have attempted to foist materially constituted relations of encompassment upon those lesser nobles. The bezotes and other jewelry items produced at Erongarcuaro could have enabled the lords of Erongarcuaro to enter into the same kinds of encompassing relationships with their own subordinate lords that the king claims to have monopolized in the RM (see Chapters 6 and 7). This possibility is speculative and cannot be addressed with the available archaeological evidence, except with regard to the relationship between Erongarcuaro and Urichu. Even the relation between Erongarcuaro and Urichu can only be addressed in a fragmentary fashion. There is not enough evidence regarding lapidary objects or their by-products at Erongarcuaro and Urichu to resolve this issue with regard to the flow of lapidary objects from the lords of Erongarcuaro to the lords of Urichu. Utilizing the other data available, however, I suggest that the relationship between Erongarcuaro and Urichu was not a relationship of encompassment. There is no evidence of marital exchange between the two sites. The lords of Urichu appear to have been autonomous with regards to their ability to acquire ceramic vessels used for feasting. Therefore while Erongarcuaro might have held a higher place in local economic (productive and distributive) networks and the lords of that site might have enjoyed a higher status or prestige due to the fact that they had received a royal concession to produce lapidary objects, they do not 300

PAGE 301

appear to have politically influenced the lords of Urichu or directly engaged them in practices that would result in encompassment. The encompassment of subordinates in the core was, at least as indicated by the evidence at present, the sole reserve of the king. It is nonetheless possible that the lords of Erongarcuaro could have engaged lords or officials other than the lords of Urichu in would-be relations of encompassment. Urichu was, after all, a fairly important administrative site within the Ptzcuaro Basin, and it might have been important enough that the king directly encompassed the lords of Urichu through material and discursive practices. As the Caravajal Visita documented, numerous sites, most of which were located to the west in the Sierra Tarasca, were subject to Erongarcuaro in the tributary hierarchy (see Chapter 3), and perhaps some of these towns and smaller settlements were targets of encompassment by the lords of Erongarcuaro. In such a scenario, the indexical relationships between the king and the lapidary objects produced at Erongarcuaro could have been muted by the nobles of Erongarcuaro as they sought an encompassing status of their own. The potential role of state agents in the investiture ceremonies of lesser lords must be confronted here again, as such a presence would have likely overshadowed the pretensions of the lords of Erongarcuaro. This potential dissonance between the archaeological possibilities and the universalistic claims of the royal dynasty as presented in the RM is difficult to evaluate given the available evidence. However, it is reasonable to at least posit that if the lords of Erongarcuaro had been able to receive a royal concession for lapidary production within their purview in the first place then they might also have attained a concession to use those lapidary objects in the realization of relations of encompassment involving lords subordinate to them. In the kings view of such a practice, the fact that he had encompassed the lords of Erongarcuaro through other relations meant simply that the lords of Erongarcuaro were acting on his behalf, i.e., as his 301

PAGE 302

representatives. In other words it was the kings intent that the lords of Erongarcuaro be able to produce and use insignia in investiture ceremonies of yet lesser lords, and the lords of Erongarcuaro were merely carrying out his intent. By implication, then, the king would have encompassed any lords that the lords of Erongarcuaro had in turn managed to encompass. It is also necessary to remember when considering the attempts of the nobles of Erongarcuaro to encompass subordinate lords, however, that the meanings of certain actions and the objects involved need not have been mutually agreed upon by the parties engaged in acts of investiture or other exchanges. In other words it is possible or even likely that there was some slippage (Keane, 1997; see Chapter 4) in the relation between intended and received meanings of such exchange acts. The lords of Erongarcuaro could have sought to emphasize the significatory relations that they had established between themselves and the lapidary objects by virtue of sponsoring or producing the objects themselves. They also could have sought to intervene in the machinations of the king and the agents of the upper nobility by somehow preventing those agents from being present at acts which would have effected the encompassment of tertiary subordinate lords (e.g., acts of investiture or marriage ceremonies). Without such state agents present, the lords of Erongarcuaro would have been free to de-emphasize the role of the king and emphasize their own role as a superior authority. From the perspective of the lords of Erongarcuaro, then, they would be encompassing their own subordinate lords through the realization of asymmetric exchange relations. If the lords of Erongarcuaro were indeed able to engage subordinate lords in relations of encompassment, they would have become encompassing beings, and therefore like the king. This status and the position of subordinate lords as their own distributed personhood could have been quite useful in efforts to improve their own lot in the state apparatus. 302

PAGE 303

From the point of view of the subordinate lord so engaged, however, the lords of Erongarcuaro could have simply been acting in the name of the king. Thus the kings encompassment of the lord of Erongarcuaro could have allowed subordinate lords engaged by the lords of Erongarcuaro to interpret their own encompassment as constituted by the king first and foremost, and the actions of the lords of Erongarcuaro would only have been the execution of the kings will and agency and not autonomously agentive and encompassing acts in their own right. Lords so engaged by the lords of Erongarcuaro likely would have recognized some aspect of ranking or status of the lords of Erongarcuaro that was higher than their own, as clearly the lords of Erongarcuaro were able to receive a kingly concession to produce such objects whereas they possessed no such ability. In their eyes, however, this prestige that the lords of Erongarcuaro had managed to acquire in relation to their own could have been de-emphasized in relation to their own perception that both themselves and the lords of Erongarcuaro were beings encompassed by the king, with more in common at a fundamental level than the mere matter of prestige that differentiated them. The subordinate lords, therefore, would have perceived their own situation in much the same way that the king didthat the king was the only encompassing being in the Tarascan kingdom. In both perspectives (that of the lords of Erongarcuaro and that of the subordinate lords), there is an attempt to save face and/or seek status, which are really two sides of the same coin. Based on the material realities of the exchange acts so realized, and just as importantly the multiple agencies objectified in the lapidary objects produced at Erongarcuaro that were at the heart of these exchanges, both perceptions were reasonable and potentially correct. Thus lords ostensibly encompassed by the lords of Erongarcuaro could prefer to see the situation as merely their own encompassment by the king. Hence they would have emphasized that the lords of 303

PAGE 304

Erongarcuaro were encompassed by the king, just as they were. On the other hand, the lords of Erongarcuaro could prefer to emphasize the fact that they were in some manner encompassing beings just like the king, and thus were different from the lords that they had encompassed. These perceived permutations on the structure of encompassment are diagrammed in figure 9-1. 304

PAGE 305

Figure 9-1. Different permutations of the structure of nested relations of encompassment according to the place of the perceiving person within that structure, as specific relations were differentially emphasized by those actors. 305

PAGE 306

The king, secondary lords, and tertiary lords (if they existed) therefore could have chosen to perceive their place within a generalized and potential structure of kingly encompassment. Each actor could have chosen to make the most of the potential significatory vagaries that might have arisen as subordinate lords attempted to invest objects with their own agencies. The choice of a perspective, however, also at least implicitly acknowledges the other potential viewpoints. For example, a tertiary lord engaged in an encompassing relationship with the lord of Erongarcuaro might choose to interpret that relationship as merely the realization of a relationship that the king had willed into existence, thus perceiving himself as encompassed by the king and not the lord of Erongarcuaro. By making a choice, however, such a tertiary lord must have recognized other potential viewpoints, namely the viewpoint of the lord of Erongarcuaro (who should choose to see himself as an encompassing actor in relation to the tertiary lord), even if he ultimately rejected it. Similarly, the lord of Erongarcuaro should at least have acknowledged the kings choice to perceive that he, the king, had encompassed the lord of Erongarcuaro and therefore should have been the encompassing actor in any of that lords future endeavors. The same, of course can be said of the king, who would have perceived such relationships in the most flattering light but also would have recognized the potential agendas of the subordinate lords. If the actors in engaged in such relationships chose to perceive the nature of those relationships in certain ways but also implicitly acknowledged the potential for other viewpoints, in other words if these actors are presumed to be knowledgeable actors (Giddens, 1984, p. 3-4, 90-92), these actors would have constructed (if only to consciously background or argue against) the total structure of kingly encompassment that appears as a series of nested relations of encompassment at the top of figure 9-1. This total structure is the sum of the perceptions of 306

PAGE 307

those actors engaged in these relations, the perceptions conscientiously selected by each actor and the potential perceptions attributed by each actor to the other actors. Implications for Investigating Tarascan State Formation The model of intersubjective encompassment developed in Chapter 4 and applied throughout the present study can be used in general to study how material objects and the manner in which they were deployed by a burgeoning Uacusecha elite was the defining process of Tarascan state formation. Within a model of intersubjective encompassment, legitimacy and order are built into the interactions between what was becoming the royal dynasty and the subordinate lord and the intersubjective consequences of those interactions. Therefore the application of a model of encompassment can address some of the shortcomings of other approaches to Tarascan state formation. More specifically, the findings from Erongarcuaro and Urichu of the present study have certain implications for the development of the Tarascan kingdom, particularly with regard to the interrelationship between the processes of the consolidation of Uacusecha power in the core and Uacusecha-led military expansion outside of the core. I suggest that the evidence at Erongarcuaro indicates that the process of Uacusecha consolidation of power within the core did not occur prior to state expansion but likely occurred in step with state expansion. This interpretation contrasts with the study of Gorenstein and Pollard (1983), who posited that the consolidation of the core occurred prior to Uacusecha expansion as an enabling factor of that expansion. In my view, the differential distribution of the spoils of war, with the Uacusecha appropriating the bulk of the spoils, made the material encompassment of lords in the core and throughout the kingdom possible. This interpretation is more in line with Pollards (1993, p. 88) view that state formation (the political consolidation of the core) took place after the initial Uacusecha conquests that are recorded in the legendary history of in the RM and which that history claims were not followed by any institutionalized 307

PAGE 308

control of the conquered towns. I go on to build on this discussion of the relationship between encompassment in the core and in more provincial areas in order to discuss more fully how the integral role of abduction in the constitution of relations of encompassment potentially impacted the realization of kingly encompassment throughout the kingdom. Encompassment and Tarascan State Formation The model of intersubjective encompassment developed in the present study, and the necessary role of material objects and social relations with those objects, can be the basis of more detailed understandings of the processes involved in the formation of the Tarascan kingdom. This model, because it is inherently temporal, can be readily applied to social transformations such as state formation. As I have discussed, a necessary process in the constitution of the kings hierarchical superiority was the ability to induce social actors to perceive the kings role in the material constitution of subordinates as agents of his will. This process of inducement could only have taken place in historical time. Moreover, it could only have taken place by incorporating objects and the realization of certain human/object interactions, which would have led to the investment of the kings agency in those objects. In other words, the transformation of social relations between what was becoming the Uacusecha upper nobility and other lords can be invested archaeologically through the objects and human/object relations that were integral aspects of those social relations. By positing the essential nature of exchange, while at the same time preserving indexical relationships between object and giver through the abductions of others, the flow of objects and the wider relations of production and exchange can be investigated archaeologically. The problem of how the Uacusecha royal dynasty produced an order that defined the rule of the royal dynasty as legitimate is in fact built into the model of encompassment developed here. Because, according to Dumonts definition of hierarchy as the encompassment of the 308

PAGE 309

contrary, the king was able to encompass social others and act through them, his nature as a hierarchically superior kind of being was self-evident. He was not like ordinary people, and the subordinate lords could not have achieved the same effects that he had. The subordinate lords, more specifically, were not agents in their own right but were part of the kings personhoodthat is why they could not have acted on their own accord, even in the context of the turmoil of the early colonial period (see Chapters 1 and 3). To act against the king would have contradicted everyones knowledge of the relations between the subordinate lords and the king and their expectations of present and future actions based on that knowledge. Additionally, as the king managed to encompass subordinate lords and transformed himself into a kind of super-agent (Mosko, 1992), his nature as such would have transformed those objects that came from him into valuable objects; they were indexes of his agency and power. As the king encompassed subordinate lords and reaped the material benefits of his encompassing nature (i.e., he could have appropriated resources from subjugated areas without needing to siphon off resources in order to buy the complacency of subordinate lords), he became a repository of efficacy and value. The goods that emanated from his person shared in, or signified, the kings value. I believe this is how a new cultural order of the valuation of goods proceeded, such that subordinate lords would have more or less accepted the new Tarascan style status markers such as ceramic vessels in place of objects from foreign cosmological centers that they had previously gone to great lengths to acquire. Furthermore, the resources that he was able to appropriate from the subjugated towns of the kingdom could have been invested into the production of a monumental capital. The model of intersubjective encompassment can formulate, in summary, explanations for how exactly a domestic cosmological center (Pollard and Cahue, 1999) was produced and perceived as powerful and legitimate by the subordinate 309

PAGE 310

lords of the realm. The model, in other words, can help explain how and why the world of material objects, their flows, and their values, were reworked in the process of producing and reproducing the Tarascan kingdom. The essential aspect of the model of intersubjective encompassment developed in the present study is that objects had to have been distributed in order to achieve any social effect, i.e., the construction of the kings nature and power. This model therefore moves away from the investigation of the states ability to accumulate resources as an end in and of itself, which I believe is a weakness of many of the models of the Tarascan kingdom and its formation discussed in Chapter 3. Rather, some resources were gathered, or produced, by and under the auspices of the king precisely so that they may be given away to subordinates. It was the act of giving itself that effected the subordination of the subordinate lords. Only with the establishment of a (materially signified) relation could subordinate lords have become subordinate, because to be subordinate was to be subordinate to something else (i.e., subordination is relational). The flow of gifts from king to subordinate lords can be investigated archaeologically. Furthermore, exactly what kinds of objects (women, insignia of office, ceramic vessels used for feasting) were or were not used as gifts to subordinate lords can be compared between one local context and another so that the different ways in which lordly and kingly agency were perceived in local contexts can be investigated. Giving gifts in order to subordinate other lords was, however, the easy part for the Uacusecha as they sought to consolidate power. The harder part was ensuring that those gifts continued to index their own agency. This was the problem of keeping those gifts that were given (following Weiner, 1992; see Chapters 4 and 5), and the ability of the Uacusecha to keep the gifts they had given meant that they were not really given in the first place. This is why I 310

PAGE 311

have placed gift in quotation marksthese objects had been given not so that they might elicit a return gift or establish friendly relations or even some ill-defined relation of fealty, but precisely so that they might simultaneously demonstrate and constitute the disparity in agency between king and subordinate lord. By placing an object in the hands of a subordinate lord and subsequently demonstrating the superiority of his agency in comparison with the subordinate lord through asymmetric or even unreciprocated exchanges, what the king was doing was constituting an indexical relation between the actions of the subordinate lord and his own person. The king was encompassing that subordinate, and thus constituting his own agency as less bounded by time and space because he could act through his subordinate lords. The kings ability to ensure that these objects that were transferred to the hands and bodies of his subordinates continued to index his own person would have in turn necessitated a whole range of other material relations, ranging from acquiring long-distance trade goods to undercutting the ability of would-be subordinates to reciprocate the gifts that had been proffered by the king. Monopolizing and bureaucratizing long-distance merchants as part of his government (Gorenstein and Pollard, 1983, pp. 103-104) was likely an important step in this effort, as it would have simultaneously resulted in the kings possession of raw materials and goods that could be used as gifts while ensuring that subordinate lords would not have had anything comparable that could be used to reciprocate such gifts. Expanding and monopolizing productive technologies was also part of the kings ability to give objects that could not be reciprocated to subordinates. This appears to have been the case particularly in the metallurgical industry. As discussed in Chapter 3, Pollard (1987) has suggested that a number of metalsmiths were employed by the king, or at least produced metal objects at his direction. Additionally, by militarily expanding into the areas of Michoacn rich in copper and other precious metals, the 311

PAGE 312

Uacusecha ensured a flow of metal to royal coffers that could have been put to use in the kings projects of encompassment rather than finding their way to the hands of local lords throughout the kingdom through non-state networks, in which case they could have been used to resist kingly encompassment. To the same extent, the numerous craft producers working under the supervision of the king likely produced what were the finest, and therefore incomparable, objects known in the kingdom, which, if given to subordinate lords would have been difficult to reciprocate. An important part, perhaps even the base, of these economic expansions in the effort to monopolize the significatory relations of objects given, was the Uacusechas possession of highly productive land. Yields from such lands could have been used to fund wars, pay craft producers, and pay merchants, all of which would have been essential to the kings ability to encompass subordinate lords, as I have just discussed. Furthermore, the acquisition of lands from lords who sought to reciprocate kingly gifts of women and/or valuable objects and thereby avoid kingly encompassment would have only widened the gap between the Uacusecha and other lords in the long run. By dispossessing other lords of their lands through gift relations, the royal dynasty would have increased their future capacity to out-produce their rivals and convert the yields of such lands into valuable objects that could, once again, be used in projects of encompassment. This process of acquiring land through more peaceful means of asymmetric gift relations could have been an integral part of consolidating power within the Ptzcuaro Basin, as the RM indicates that the Uacusecha did not engage all of the lords of the basin in war but brought them into the fold through more peaceful means. More generally, the issue of land and land ownership is an important one in the process of Tarascan state formation and reproduction. I have not addressed the problem of land to a great 312

PAGE 313

degree in the present study, but it was likely an essential factor not only in the ability of the Uacusecha to out-produce rival lords and convert surplus into valuable objects but also in the very constitution of relations of encompassment themselves. As I discuss in Chapter 3, local land titles and many of the Relaciones Geogrficas claim that all rights to land and other resources derived from the king. The RM (1956, p. 14-15, 173) similarly indicates that, in theory at least, the god Curicaueri (and therefore the king) owned the land. State ownership of the land was likely a consequence of the conquests of the towns of the kingdom. If the king claimed to own all land and it was given to subordinate lords and to commoners so that they might provide for themselves, this could have been the foundation of relations of encompassment. Such permission or allocation, essentially a gift, would likely have gone unreciprocated. What could the commoners have possibly given to the king, other than recognition of the fact that they were encompassed and therefore that they owed their own labor and productive capacities to the king? On the other hand, the indication in the RM (1956, p. 184; see Chapter 7) that a subordinate lord would have provided a plot of land and labor to work it to the offspring of a woman who had married into the kings palace perhaps means that in actual practice the king did not own all land. In other words, the king might have claimed ownership of all land, but this claim might have been subject to interpretation and refutation by local lords and commoners in certain contexts. Local lords and commoners might have recognized and affirmed the kings claim in some discursive arenas (such as the production of land titles and other documents intended to be official), but at other times might have gone about their business as if that claim was irrelevant. At this point I can merely suggest that the role of land and the degree to which subjects of the king actually did perceive that access to land was a gift from the king are topics that require more research within the model of encompassment developed here. 313

PAGE 314

Relations in the Core and Tarascan State Formation The results of the research at Erongarcuaro and Urichu have certain implications for the study of Tarascan state formation more specifically. The likelihood that certain concessions were made to local and preexisting power bases within the core during the rise of Uacusecha leadership within that core, exemplified by the allowance of lapidary production at Erongarcuaro, suggests that the rise and expansion of the Tarascan kingdom was not necessarily the result of the development of a monolithically dominated core. This interpretation relies on the view that lapidary production at Erongarcuaro was in some sense less than optimal from the viewpoint of the king, as discussed above. While the king could perceive that the lords of Erongarcuaro acted on his behalf if those lords did indeed engage lesser lords in relations of encompassment, this could have enabled the lords of Erongarcuaro to claim for themselves an encompassing and superior nature, at least potentially threatening the exclusivity of the kings legitimacy. Again, the upper nobility that served as informants in the production of the RM implied that all subordinate lords received their insignia of office from the kingthey did not want to admit the possibility that subordinate lords could have behaved as encompassing beings just as the king did. Therefore the presence of lapidary production at a secondary administrative site indicates that some kind of negotiation between the king and the lords of Erongarcuaro likely took place, as both sought to shape the terms under which the lords of Erongarcuaro took part in the developing state system led by the Uacusecha. As discussed in Chapter 7, the RM claims that both Erongarcuaro and Urichu were allied with the Uacusecha at the outset of the conquests that created the Tarascan state. By implication, and as represented in the RM directly, these towns were never conquered outright by the Uacusecha. Granting concessions as rewards or as part of the terms of that alliance might have been essential aspects of these negotiations that in the end 314

PAGE 315

would have brought the lords of Erongarcuaro and perhaps also Urichu into the fold of what was to become the Tarascan kingdom. This alliance and reward could also be interpreted as indications of the power of these towns and their lords in the pre-state era. I would therefore describe the processes that led to power differentials in the core and the rise of the Tarascan state as the amalgamation of the core rather than its consolidation. In this process of amalgamation, the Uacusecha managed to garner support from autonomous lords in a joint enterprise that they would be the primary beneficiaries of. Over time, as the Uacusecha lords benefited differentially in comparison to the other lords of the Ptzcuaro Basin core, the growing wealth differentials between the Uacusecha lords and the other lords of the basin enabled the Uacusecha lords to more fully engage these local lords of the core in asymmetric exchanges that would have encompassed those lords by means of the abduction of their agency as part of the kings personhood. This is a different model than that presented by Gorenstein and Pollard (1983), who proposed that the development of Uacusecha dominance within the core, as indicated by the (re)organization of the settlement pattern within the core to suit the states needs of effective administration, enabled the extra-basin conquests that expanded the dominion of the Tarascan kings. In their model, the development of a settlement pattern suitable for effective administration, but little else, was the enabling factor that allowed the Uacusecha to direct activities in the core and ultimately to transform this administrative advantage into the capacity for military conquest. On the other hand, Pollard (1993, p. 88) places the consolidation of the core, i.e., Tarascan state formation, after the earliest conquests led by the Uacusecha and before the institution of local offices and a semblance of a state bureaucracy (RM, 1956, pp. 151-152). The model of amalgamation and encompassment within the core advocated here can be rather easily 315

PAGE 316

accommodated with this suggestion by Pollard that extra-basin raids likely played a part in the process of the consolidation of the core. The spoils of those raids could have been used to initiate the kinds of asymmetric exchange relations that, reproduced over time, would have effected the encompassment of the recipients who had in the process become incipient subordinate lords. I would, however, point out that according to the legendary history of the Uacusecha royal dynasty contained in the RM, these extra-basin raids that occurred at the outset of the Uacusecha military expansion led by Hiripan, Tangaxoan, and Hiquingaje were not the first raids undertaken by members of the Uacusecha. Rather, Tariacuri himself had, according to that same legendary history, gone on raids following his marriage to the daughters of Zurumban (the spoils of which the Coringuaros failed to extort from Tariacuri; see Chapter 6). To the extent that raids for booty were a recurrent phenomena in the years immediately preceding Tarascan state formation, the spoils of such raids could have been used to bring the other lords and polities into the developing Uacusecha-led alliance that ultimately was transformed into the Tarascan kingdom. Viewing state formation and consolidation as a process of producing and then reproducing certain actions requires that the process of Tarascan state formation was as much of a process of reproducing such asymmetric exchanges (and thus keeping-while-giving; see above) as it was definable by the initial ability of the Uacusecha to give objects that could not, at that moment at least, be reciprocated. As I discuss in Chapter 4, for the abduction of agency to be ensured both in the present and in the future (and the two tend to dissolve into one another as expectations of future actions can influence perceptions in the present), there should have been a sense among witnesses that the actions that should have produced the abduction of agency and thus the perception (and conferral) of encompassment would be reproduced. All of this is to say that 316

PAGE 317

while the spoils of Uacusecha led raids could have initiated a process of consolidation and brought the lords of the Ptzcuaro Basin into the sphere of Uacusecha influence, the process of state formation that relied on such objects would have required more objects sooner rather than later that could be given to these lords in order to reinforce the perception of an increasingly asymmetric and encompassing relation. This would have necessitated increased production locally (which could have been achieved by the Uacusecha, particularly if they had managed to dispossess other lords of agricultural land) or the constant waging of raids and ultimately wars of conquest, or likely both. In other words, conquest was only the beginning of the process of state formation, as further steps must have been taken to encompass conquered lords through the establishment of certain material and social relations. On this point I agree with Pollard (1993, p. 88; 1994; 2003a), although I believe she would probably describe this process as the bureacratization of the subordinate lords as a definable and subordinated class. Encompassment in the Core in Relation to Encompassment in Provincial Areas In an expansionist state formed largely through conquest, the question of how representative the case studies of Erongarcuaro and Urichu are with respect to the Tarascan kingdom as a whole must be addressed. As represented in the RM and as I suggest above, the likelihood that Erongarcuaro and Urichu were incorporated into the developing state system through peaceful, if not uncontested, means, makes them anomalies in the Tarascan kingdom. As I discuss shortly, the relationship between king/subordinate lord interactions in the core and in more provincial areas is not a matter of idle pondering but could have had have important consequences with respect to the lords of Erongarcuaro and Urichu and their encompassment by the king. One fundamental difference between the towns of the core and the towns throughout the rest of the kingdom was the difference between peaceful, if negotiated and at times contested, 317

PAGE 318

acquiescence to Uacusecha authority and legitimacy in the core and forceful subjugation outside of it. As represented in the RM (1956, pp. 156), inherent in the act of conquest and subjugation was the exchange of the lives of the subjugated for their service to the king; the conquered were not sacrificed but instead were obligated to work for the state. Uacusecha authority therefore could have looked very different outside of the core. In the act of conquest, the agencies of the Uacusecha king (and his followers) had definitively bested the agencies of local lords and populaces. This event was preserved both in the historical memory of the populace as is evidenced by the legendary history of the Uacusecha and in the structural arrangement of debt/work obligations that existed between the populace and the Uacusecha. Additionally, if economies in more provincial areas are demonstrated (by further research) to have lacked developed economies in which the local lords could either produce or acquire the kinds of goods that were required for their efforts of constructing the legitimacy of their rule (finely crafted feasting wares, lapidary products, and highly valued foreign objects) I expect that the state would have filled this niche exclusively. This would have had the effect of ensuring the abduction of the kings agency in the actions of the local nobility by commoners and other witnesses. Through the states ability to expand the range of practical and material means through which kingly encompassment was realized and signified (which are essentially the same thing) the relation of encompassment should have been itself less tenuous and less open to the potential for slippage than perhaps arose in the core where only one or two practices was expected to carry the significatory load (e.g., as at Erongarcuaro). A potential countervailing factor to the likelihood of the more firmly established encompassment of subordinate lords outside of the core was the distance between the capital and these farther flung areas of the kingdom. The performative and discursive construction or 318

PAGE 319

representation of the encompassing nature of the king was fairly easy to achieve in the core due to the ease of transportation afforded by the lake, to say nothing of the relatively small distances necessary to traverse the landscape in the Ptzcuaro Basin overall. At greater distances from the capital, the realization of communal rituals involving the king or his emissaries might have been more problematic. This is an important point, because these ceremonies produced the cosmological authentication of the social categories, the relations between the social categories, and the objects that objectified and constituted those relations. The central question in this issue is how universally realized was the practice, represented in the RM, of kingly emissaries playing integral roles in the investiture ceremonies and other events crucial to the construction of the encompassed nature of the local lord? Did such events really occur as they are stereotypically portrayed in the RM? Or was there some variation in the practical enactment of the cosmological authentication of the kings encompassing nature, ranging all the way to nonexistent? As they relate to this issue, certain representations in ethnohistoric documents can be brought to bear in favor of the exportation of a kingly presence in more provincial areas. For example, the representation in the Cdice Plancarte of the role of members of the royal dynasty in the naming, and therefore constitution, of the community of Carapan (see Chapters 2, 5) can be interpreted as the result of precisely these kinds of rituals and performances. In the same corpus of documents pertaining to Carapan, members of that community claimed that the indigenous governor Don Antonio Huitzimengari visited the community in the middle of the sixteenth century, inspecting the community and its environs (Roskamp, 2003). If such a practice was the extension of a prehispanic practice, and assuming that the indigenous governor of Michoacn in the early colonial era was in some way the equivalent of the king of the 319

PAGE 320

prehispanic era, this too would have been the kind of practice that would have made the claim of the kings encompassing nature public. Lastly, the claim that an emissary of the king was sent to Tetlamn in order to resolve a dispute (Carrasco, 1969, p. 219) suggests that in one way or another the kings effective reach was fairly extensive. If emissaries could be sent to the corners of the kingdom to resolve a dispute, certainly the king could afford to send royal representatives to occasions as important as the investiture ceremonies of new lords, local celebrations of the festival at which the legendary history of the Uacusecha was related to the townspeople, and the mortuary ceremonies of a recently deceased lord or noblewoman). Guilt By Association and the Relations Between Lords in the Core and Provincial Areas To this point I have largely focused my attention on the vertical relations of encompassment enacted between the king and subordinate lords, as well as potential relations between subordinate lords and tertiary subordinate lords. In such interactions, the encompassing and encompassed agent are directly defined by the nature of the relation between the two. In this section, however, I discuss the implications of the probability that in the state system of classification, imposed upon the bureaucracy of subordinate lords, a principle of guilt by association was fostered by the state in the minds of the populace. The implications of such a principle might have been important factors in the perception and conception of kingly encompassment and ultimately in the historical development of the Tarascan kingdom. By naming offices and the lords that occupied those offices (as recorded in the RM; see Chapter 3), the state classificatory system established paradigmatic relations (i.e., relations that established membership of terms within categories) between lords who might otherwise not be related through direct contact or interaction. Such a classificatory system would have had effects that ran in a parallel and complementary fashion to the direct interactions and interrelationships, such as exchange, that shaped how subordinate lords would be defined in relation to the king. If 320

PAGE 321

a relational definition of social beings and their agencies is accepted (and it has been the foundation of the present study), no two persons will be exactly alike, that is, constituted of exactly the same interactions and relations. While I have argued in favor of such an approach throughout the present study, the potential effects of a different sort of relational interaction should also be appreciated. Instead of relying solely on redefining the identities and agencies of subordinate lords through direct interaction, the state could impose what was to some degree an artificial paradigmatic relation upon the subordinates lords of the realm by giving them a similar name (be it angamencha, caracha-capacha, etc.). I say artificial because such a relation, by insisting on similarity, would have attempted to erase the inevitable differences between the subordinate lords that resulted from their interactions with the king, interactions which would have been impossible to exactly reproduce from one subordinate lord to the next due simply to the contingencies of history, material and economic fluctuations, and so on. This artificial horizontal paradigmatic relation could have redefined the social beings so named in relation to the practices by which the king ought to have engaged other similarly named lords (following Bourdieu, 1977, 1999). Of course, the relation that this artificially produced class of officeholders ought to have had with the king was yet another matter that was subject to manipulation through discursive framing and the production of typified representations of the kind that are contained in the RM. Through such productions, the unambiguous encompassment of one lord could have impacted how the relation of a more ambiguously encompassed lord was perceived, so long as the community of abductors accepted the paradigmatic association between the lords being linked by occupying the same named office. The reverse was also potentially the case, and I return to the importance of these related points for the rise of the Tarascan kingdom shortly. 321

PAGE 322

The integral role of the witnesses and on-lookers who abducted the primary agency behind the manifestation of any important act in this paradigmatic horizontality of relations presupposes, however, that such persons were aware of relations between the king and subordinate lords in other local contexts, or at least that they thought they were. With respect to such issues, certain integrative mechanisms that brought people together and allowed them to gossip about their own local political contexts, such as the marketplace, likely played an essential role in the construction of social relations and social knowledge beyond the community. Kin relations might have facilitated such gossip across community boundaries, of course, but the RM indicates that marriage among commoners was by and large community endogamous (RM, 1956, pp. 211-214; also Pollard, 1993, p. 59). The nobles themselves were important witnesses of the relations between the king and members of their ranks. The RM (1956, pp. 211-213) does indicate that nobles married among themselves, i.e., that a noble from one area or town married a noble from another area or town. These marriages would have provided important routes of communication pertaining to the nobles relations with the king. All of the lords were apparently required to attend court regularly, as well (RM, 1956). Such interactions could have aided in the construction of a more uniform and conscious class of subordinate lords or served as the arena in which rivalries over the nature of the relationship that each had constituted with the king became a point of contention and competition among the nobles, or possibly even both at the same time. Additionally, the extent to which commoners and other on-lookers thought they knew what was going on in the rest of the kingdom and the relations forged between the king and other lords was an arena in which the state could have potentially had a significant impact. Through such performances as the yearly recitation of the history of the Uacusecha royal dynasty, the state could establish an imagined community (Anderson, 1981) at the level of the kingdom. The 322

PAGE 323

RM states, after all, that not only was this history recited in the capital but that priests were sent to all of the towns of the kingdom to tell this history. Within such ceremonies it might be presumed that the state attempted to impress upon the people of the realm a community of imagined nobles, all basically the same, each duly encompassed by their interactions with the king and the historical roots of those interactions. To that end, and to return to the implications of the importance of regularly abducting the kings encompassing persona, what the people of one particular locale thought about the relations between the king and lords in other locales shaped how they perceived the kings relationship with their own lord. Therefore the development of the kings persona as an encompassing one likely had something of its own inertia. To the extent that the king had encompassed some lords through materially constituted practices and was perceived by others as an encompassing being, this would have made it easier to achieve the abduction of his agency as the primary agency in other relations, even in relations in which a certain amount of ambiguity would have otherwise existed. The encompassment of some lords made the encompassment of others more assured. Encompassment was, therefore, according to a certain intersubjective calculus, the means of its own perpetuation. Following this logic, then, by establishing encompassing relations with most or even some lords and secondarily by constructing a state system of categorization and thus uniformity within the state bureaucracy in order to impose a vision of everyones place within the kingdom, the king could have forged a relation of encompassment that might have otherwise been problematic or tenuous. 1 The state endeavor to organize, classify, and make visible the paradigmatic 1 My thinking in this regard is influenced by Bourdieus (1999, p. 65) discussion of the competition in fifteenth century Spain between the hereditary nobility and a new nobility who had acquired their position through a relationship with the king, were beholden to the king for their position, and thus were loyal to the king. The 323

PAGE 324

associations between lords, thus furthering the kings cause in his efforts to encompass all subordinate lords, would have been facilitated by the proclamation of the history of the royal dynasty (discussed above and in Chapter 6), the development of specific insignia or adornments indicative of a common rank within the state bureaucracy, and even simple projects such as encouraging lordly visitations and communal festivals involving two or more towns at which the insignia of the lords in the rites would do the talking. More than simply embarking on a project of simplification (Yoffee, 2005, pp. 91-94) for its own sake, the king could have attempted to simplify the perception of subordinate lords as uniformly encompassed by the king, promulgating the concomitant view among the populace that any one lord was basically the same as any other lord holding the same named office. Moreover, this would have facilitated the perception that the king was an encompassing being by virtue of his encompassment of (some to most) subordinate lords. The promulgation of a state bureaucracy in which holders of named offices were assumed to be essentially alike was, in essence, yet another tool that the king could use to regulate the abduction of his own primacy and thereby encompass subordinate lords more by association than by any particular practices. Therefore, and to the degree that such projects might have been successful, the relationship that the king enjoyed with lords in the core would have impacted the relationship that the king enjoyed with lords in more peripheral areas, and vice versa. If the lords of one area or another were more difficult to encompass in practice, the potential existed for the king to strengthen relations with the lords of another area and thereby encompass the more reticent lords by implication and by force of his own burgeoning persona. Of course, such implied or associative encompassment was just as vulnerable, and likely more so, to the problems of slippage traditional nobility must have found themselves in a new and unfamiliar position, their own relationship to the king being altered by the presence of the new class of nobles. 324

PAGE 325

between intent and recognition discussed in Chapter 4. After all, in the end all politics is local, and the failure to register any kingly impact upon the local relationship between a lord and those commoners under his charge would have been conspicuous in its absence. The dissonance between state propaganda (i.e., efforts toward encompassment through classification) and the local context in which the king was conspicuously materially absent would have called out for reflection and evaluation, and such dissonance could have been exploited by the subordinate lord in his own effort to minimize the kings role in the constitution of his agency within that local political context. The importance of classificatory schemes to kingly projects of encompassment is a final reminder of the weaknesses of positivist investigations of political structures and their organizations as they have been traditionally employed by archaeologists and ethnohistorians in investigations of the structure, functioning, and vulnerabilities of political entities. The fact that naming is a political device should not be lost on those who would simply count up the number of similarly named officials reporting to other similarly named officials and offer pronouncements as to the relative strength, stability, and efficiency of the political structure in question. Rather, the act of imposing names was a means to a political end; names are merely reflections of the various power relations that through their interaction have settled upon what are literally more or less agreed upon reference points for talking about each other and the world at large. In relation to many of these points, Pereiras (n.d.) analysis of the mortuary remains in the Zacapu Basin in comparison to Urichu, Huandacareo, and the Sayula Basin of Jalisco is relevant. Pereiras analysis demonstrates that at Hunadacareo and in the Zacapu Basin, females were buried with valuable objects that were made from foreign or state controlled materials while 325

PAGE 326

males were not, just as at Urichu. This indicates that in these areas, and by one measure of encompassment, kingly encompassment was realized in relation to the subordinate lords of these sites, and would have been mutually reinforcing. In contrast, in the Sayula Basin at the western margins of the Tarascan kingdom, males were buried with valuable objects made from foreign or state controlled objects, while women were buried with less such objects. This would indicate that the lords of this area were able to resist kingly encompassment and constitute a more symmetric relationship, even if these lords were participating to some extent in state projects or acquiring Tarascan status markers, as indicated by the presence of Tarascan style polychromes in this area. Directions for Future Research into the Development of the Tarascan Kingdom I have briefly outlined above how the general model of kingly encompassment can be used to model Tarascan state formation in terms of processes of acquiring, producing, and ultimately deploying objects in ways that would have ensured that these objects were abducted to be the distributed personhood of the king, thus effecting the encompassment of the subordinate nobility. This process must have been one that played out in historical time, as objects were exchanged and the concomitant human/object interactions that ensured the abduction of the kings agency were realized and organized so that they might be reproduced. Each step in the process of producing such exchanges and then producing the ability to reproduce those exchanges (e.g., by launching wars of conquest, claiming the best agricultural lands, monopolizing foreign trade) would have been material, and therefore amenable to archaeological investigation. Additionally, state formation would have been a process that proceeded as a regional enterprise, as what happened in one area of the Tarascan kingdom impacted how the events in another area would have been perceived. This is not an application of world systems theory or perspective (e.g. Chase-Dunn and Hall, 1997; Peregrine and Feinman, 1996; Wallerstein, 1974), i.e., the 326

PAGE 327

recognition that economic (i.e., human/object) relations in multiple areas or regions are mutually interdependent. Rather, it is the result of analyzing the intersubjectivity of social relations as partially determinative of those relations in coordination with the fact that intersubjective relations constituted in potentially disparate areas were mutually interdependent. The nature of the relationship between the king and one subordinate lord had potential consequences for the nature of the relationship between the king and another subordinate lord as participants and witnesses alike actively constructed the nature of their own relationships with the king in relation to their knowledge of the kings other relations. Therefore one of the most important implication of the adoption of a relational ontology of social identities, and of the discussion of the myriad ways in which the kings encompassing personhood was made manifest, is that much more research pertaining to the Tarascan kingdom is needed. Because social interactions in the core impacted how social interactions were realized in more provincial areas, and vice versa, this mutual interdependence requires that more work needs to be done throughout the kingdom, and done in such a way that allows for detailed comparisons of human-material interactions and the biographies of those objects that constituted relations of encompassment across the Tarascan kingdom. With respect to the issue of the reconstruction of object biographies and the interactions that established relations of encompassment, more research can be done without the need for further field research. Rather, the materials from those provincial areas such as the Zacapu and Cuitzeo basins could be sourced in order to reconstruct the production and exchange networks through which the local lords came to possess the objects needed in order to construct their legitimate authority. The ceramic paste sourcing research within the Ptzcuaro Basin of Pollard, Hirshman, and colleagues has been integral to the present study, and other similar projects could 327

PAGE 328

be realized using clay samples and samples from extant archaeological collections from these other areas in an effort to answer important questions regarding the role of the king in endowing subordinate lords in these areas with the objects necessary for meaningful action. Were such objects derived from the king? Were they locally produced? What other contextual associations surrounded any potential local production, as gleaned from the collections and data that already exist? Following from the demonstration that the lords of Erongarcuaro were able produce lapidary objects, the opportunity exists to determine whether or not these lords did, in fact engage subordinate lords in relations of encompassment. Urichu was very likely not engaged in a relation of encompassment by the lords of Erongarcuaro, but there were many other towns subject to Erongarcuaro in the tributary hierarchy, some of which might have been prime targets of encompassment by the lords of Erongarcuaro. Research at such towns could help to determine the nature of interactions between the lords of Erongarcuaro and the lords of these towns, and therefore the interplay between the interests of the lords of Erongarcuaro and the king. For example, in contexts where there was significant interaction between the lords of Erongarcuaro and towns west of the Ptzcuaro Basin, are there also strong indications of state emissaries that would frame the kings primacy in the actions of the lords of Erongarcuaro as well? Or was the king satisfied with allowing the lords of Erongarcuaro to act on his behalf in such towns, with the attendant risks that the kings primacy might not have been abducted by such tertiary lords? The ability of the king and uppermost levels of the state administration to endow subordinate lords with the objects needed to construct their own legitimacy, and thereby encompass these lords, would have required a fairly high level of state sponsorship of craft 328

PAGE 329

production. Currently, little is known archaeologically of the presence of craft specialists under the auspices of the king, save for Pollards (1993, Appendix B) identification of lapidary workshops at the capital of Tzintzuntzan and Pollards (1987) discussion of kingly control over at least some metallurgists. Given the clay sourcing studies and ethnohistoric information that suggest that Ihuatzio was a likely candidate for the location of a number of craft specialists, more research at this site is sorely needed (Pollard, 2003c, 2003d). Such research could answer the following questions, to name just a few. What kinds of objects were being produced for the state? What scale were they being produced on? To what purposes might these objects have been put by the royal dynasty? On a related point, I note in Chapter 8 that the potential exists for ceramic vessels used in feasting ceremonies to have indexed the king through stylistic attributes rather than their origination from the king (the latter does not appear to have happened given the present evidence). The presence of ceramic production of feasting wares at one of the Uacusecha capitals could greatly elucidate the role of state sponsored production in the spread of a uniform suite of stylistic elements and forms that characterize Late Postclassic Tarascan ceramic vessels. Exactly how uniform this suite was across the kingdom also could be a fruitful line of inquiry in investigating the nature of local-state interactions. Analyses of Tiwanaku (Janusek, 2004) and Inka (Bray, 2003) feasting wares suggest that within broad stylistic canons minor variations may or may not be perpetuated that signify the amalgamation, rather than imposition, of the general characteristics that compose the stylistic canons themselves. The degree to which a state school of ceramic production and decoration (if one actually existed) was copied closely could be a clue to the degree to which the king and royal dynasty set the terms by which local lords could have signified their relation to the Tarascan king. 329

PAGE 330

Finally, I believe that a model of intersubjective encompassment, and the manner in which encompassment would have produced a hierarchical cultural order and the legitimacy of the ruler within that order, can be fruitfully applied to comparative studies of state formation and reproduction. I the present study I discuss how the Tarascan king ensured the abduction of his agency in the actions of subordinates dispersed across space. This need not have been the case in the constitution of relations of encompassment, however. Therefore comparative analysis might proceed by investigating the manner in which the king sought to implicate himself in the actions of subordinates, who might have been privy to those actions, and thus what were the spatial and/or temporal limits of the perceived encompassment of subordinates? By posing such questions, the differential construction of legitimacy and the extent to which that legitimacy was made manifest in the nature of real-world interactions can be investigated. Conclusion In this study I have constructed a model of intersubjective encompassment in the Tarascan kingdom. This has been essential to investigating the active and ongoing production of legitimate authority and order within the Tarascan kingdom. Previous investigations of the Tarascan kingdom have not investigated the production of legitimate authority or order. Rather, they are simply assumed to have been produced by the kings control over material resources or the nature of the bureaucracy itself. I have argued that this assumption is problematic, however, for two main reasons. First, the actual social processes by which legitimacy was produced are necessary if comparative models of state formation are to be formulated. Second, the models of Tarascan state formation, particularly the delineation of a prestige reward system, cannot adequately explain the loyalty of the subordinate lords to the Tarascan king in the early years of subjugation by the Spaniards. Put simply, exactly how or why the Tarascan kings power should have been perceived as legitimate and how that legitimacy and the cultural order in which it was 330

PAGE 331

embedded were historically produced are mostly unknown because they have not been the explicit subjects of modeling and theorization. In an effort to model the production of legitimacy, I have elaborated on Dumonts theory of hierarchical valuation manifested in relations of encompassment. The model of intersubjective encompassment presented here posits that an encompassing person can encompass, or subsume within his own person, the subjectivities of others. I have posited that a prominent way in which this could have been achieved was through the materially and discursively prompted abductions of yet others that the encompassed actor was merely the means by which the encompassing person achieved his own will. This is an inherently relational ontology of personhood and agency; persons are defined not only in relation to one another, but in relation to what the abductions that yet others make with respect to the initial relation. By encompassing others, namely the subordinate lords of his realm, the king constituted his own superior value and therefore the legitimacy of his rule. Furthermore, I have contributed to the burgeoning literature on the archaeology of personhood (e.g., Bruck, 2005; Fowler, 2002, 2004; Gillespie, 2001, 2009; Heckenberger, 2005; Jones, 2005) by suggesting that personhood was relational according to a triadic (at minimum) relation involving two interacting actors and the witnesses that interpreted the nature of that interaction. Additionally, I have outlined (see Chapter 5) an approach that can evaluate the nature of those interpretations made by witnesses by which the nature of the actors would have been constituted. I have also presented evidence for the realization of the relations of encompassment at two sites, Erongarcuaro and Urichu, in the Tarascan core. The king was able to invest his agency in objects that were given to the lords of the two sites, and witnesses at these two sites would have abducted the kings agency from these investments. Helping the witnesses along in the process 331

PAGE 332

of abduction were priests and other state agents who discursively framed the objects and the relations between the king and subordinate lords that they signified. The evidence at Erongarcuaro in particular demonstrates that state agents did indeed travel to that site in order to frame the objects that were so essential to the kings ability to encompass subordinate lords. In addition to a summary of the evidence for kingly encompassment, in this final chapter I have discussed the wider implications of the adoption of a relational ontology, particularly one founded on relations of encompassment, for understanding hierarchy and power relations in the Tarascan kingdom. In any relation of encompassment, there is a tension between perceiving similarity and difference, and different actors within the state system (e.g., king, secondary lord, tertiary lord) could have perceived the same interactions in different ways, differentially emphasizing the sameness or difference of actors in the total structure of encompassment. Furthermore, a relational ontology, in which agents are continually constructed through their interactions with others (and the interpretations of those interactions made by witnesses), the totality of relations in which any particular actor is involved must be investigated. This leads to the recognition that if the historical development of the Tarascan kingdom (and the consolidation of kingly superiority and power at its heart) is to be understood, the manner in which the king engaged subordinate lords throughout the kingdom and down through time must be investigated. In other words, such a theoretical framework can explain much about the Tarascan kingdom, but it also requires that much more work be done. 332

PAGE 333

APPENDIX A DATA ON THE OBSIDIAN ARTIFACTS FROM ERONGARCUARO Introduction Here I present the data related to the obsidian artifacts from Erongarcuaro. This presentation is limited to that data pertaining to my goal in the present work, which is to document the relationships between the king and subordinate lords of Erongarcuaro (see Chapter 8). A complete analysis of the obsidian from Erongarcuaro, in addition to (and as compared to) the obsidian from the site of Urichu and Pollards survey of Tzintzuntzan is in preparation by Karin Rebnegger, a graduate student under Helen Pollard at Michigan State University. Below I present data pertaining to the sample of 50 obsidian artifacts that were sourced according to their chemical composition under the supervision of Dr. Michael Glascock of the Missouri University Research Reactor (MURR) (Glascock 2006). The obsidian artifacts were classified according to color and other visual characteristics by Karin Rebnegger before they were sent to be analyzed at MURR. Table A-1. Results of the sourcing by chemical characterization of 50 obsidian samples from Erongarcuaro. In the source names, Cerro is abbreviated as C. The numbers next to particular sources indicate a specific flow. The sourcing study was unable to differentiate between the La Primavera-1 (abbreviated as LP1) and Huaxtla source flows, although they are both part of the La Primavera volcano just west of Guadalajara. The asterisk indicates that based on the XRF study, the most likely source of this sample is the Zinparo source in northern Michoacn. The Cerro de la Bola source is in modern Quertaro, the Pnjamo source is in modern Guanajuato, and the remaining the sources (except the La Primavera source noted above) are in Michoacn. Sample ID# Color Field Excavation Unit and Level Phase Source MER101 Red-black translucent ER-02 N97E8Lvl3 Tariacuri Zinapcuaro MER102 Dark greenish gray ER-02 N97E8Lv3 Tariacuri LP1/Huaxtla MER103 Dark greenish gray ER-02 N97E8Lv4 Tariacuri LP1/Huaxtla MER104 Dark greenish gray ER-02 N97E8Lv4 Tariacuri LP1/Huaxtla MER105 Dark greenish gray ER-02 N97E8Lv4 Tariacuri LP1/Huaxtla MER106 Matte gray ER-02 N97E8Lv4 Tariacuri C Varal 333

PAGE 334

MER107 Striated gray-clear ER-02 N107E8Lv2 Tariacuri C Varal MER108 Dark greenish gray ER-02 N107E8Lv2 Tariacuri LP1/Huaxtla MER109 Feathery gray ER-02 N107E8Lv2 Tariacuri Ucareo MER110 Dark greenish gray ER-02 N107E8Lv2 Tariacuri C de la Bola MER111 Smokey gray ER-02 N107E8Lv2 Tariacuri C Varal MER112 Black ER-02 N107E8Lv2 Tariacuri Ucareo MER113 Striated gray-clear ER-02 N107E8Lv2 Tariacuri C Varal MER114 Smokey gray ER-02 N107E8Lv2 Tariacuri C Varal MER115 Red-black translucent ER-02 N107E8Lv2 Tariacuri Ucareo MER116 Striated gray-clear ER-02 N109E0Lv3 Tariacuri C Varal MER117 Striated gray-clear ER-02 N109E0Lv3 Tariacuri C Varal MER118 Striated gray-clear ER-02 N109E0Lv3 Tariacuri C Varal MER119 Smokey gray ER-02 N109E0Lv3 Tariacuri Ucareo MER120 Feathery gray ER-02 N109E0Lv3 Tariacuri Ucareo MER121 Feathery gray ER-02 N109E0Lv3 Tariacuri Ucareo MER122 Black ER-02 N109E0Lv3 Tariacuri C Varal MER123 Smokey gray ER-02 N109E0Lv3 Tariacuri Ucareo MER124 Dark greenish gray ER-02 N109E0Lv4 Tariacuri LP1/Huaxtla MER125 Dark greenish gray ER-02 N109E0Lv4 Tariacuri LP1/Huaxtla MER126 Striated gray-clear ER-02 N109E0Lv4 Tariacuri C Varal MER127 Striated gray-clear ER-02 N109E0Lv4 Tariacuri C Varal MER128 Striated gray-clear ER-02 N109E0Lv4 Tariacuri C Varal MER129 Striated gray-clear ER-02 N109E0Lv4 Tariacuri Ucareo MER130 Feathery gray ER-02 N109E0Lv4 Tariacuri Ucareo MER131 Black ER-02 N109E0Lv4 Tariacuri C Varal MER132 Black ER-02 N109E0Lv4 Tariacuri Ucareo MER133 Smokey gray ER-02 N109E0Lv6 Tariacuri C Varal MER134 Smokey gray ER-02 N109E0Lv6 Tariacuri C Varal MER135 Smokey gray ER-02 N109E0Lv6 Tariacuri Ucareo MER136 Smokey gray ER-02 N109E0Lv6 Tariacuri Ucareo MER137 Smokey gray ER-02 N109E0Lv6 Tariacuri C Varal MER138 Red-black non translucent ER-02 N111E8LvPz-1 Tariacuri Unassigned* MER139 Red-black translucent ER-02 N111E8LvPz-1 Tariacuri Zinapcuaro MER140 Red-black translucent ER-02 N111E8LvPz-1 Tariacuri Zinapcuaro MER141 Black ER-02 N89E0Lv3 Tariacuri C Varal MER142 Dark greenish gray ER-02 N89E0Lv3 Tariacuri Penjamo1 MER143 Dark greenish gray ER-02 N89E0Lv2 Tariacuri LP1/Huaxtla MER144 Dark greenish gray ER-02 N89E0Lv2 Tariacuri LP1/Huaxtla MER145 Dark greenish gray ER-02 N89E0LvPz-1 Tariacuri C de la Bola MER146 Dark greenish gray ER-02 N89E0LvPz-1 Tariacuri C de la Bola MER147 Dark greenish gray ER-02 N89E0LvPz-1 Tariacuri LP1/Huaxtla 334

PAGE 335

MER148 Dark olive green ER-03 N72W225Lv4 Loma Alta Pachuca MER149 Dark olive green ER-03 N72W225Lv5 Loma Alta Pachuca MER150 Dark greenish gray ER-03 N70W240Lv4 Unknown Penjamo1 Table A-2. Counts of Tariacuri phase obsidian samples from particular sources sorted by color. Among gray/black obsidian there is a high proportion of Ucareo/Ucareo obsidian, indicating provisioning from, and close ties to, the state (following Pollard 2003c). The high percentage of striated gray/clear sourced to the Cerro Varal/Zinparo source complex perhaps indicates that similar clear obsidian was not plentiful at the Ucareo/Zinapcuaro complex. Obsidian from the La Primavera/Huaxtla source complex dominates the sample of green obsidian. Color LP1/Huaxtla C de la Bola Penjamo1 Zinapcuaro /Ucareo C Varal/ Zinparo Striated Gray/Clear N=0 N=0 N=0 N=1 N=8 Gray/Black N=0 N=0 N=0 N=10 N=9 Red N=0 N=0 N=0 N=4 N=1 Green N=10 N=3 N=1 N=0 N=0 335

PAGE 336

APPENDIX B CERAMIC DATA FROM ERONGARCUARO AND URICHU WITH STATISTICAL MEASURES OF DIFFERENCE Introduction Here I present raw data concerning the ceramic assemblages from both Erongarcuaro and Urichu. This data basically sorts, counts, and calculates the total sherd samples from different areas of these two sites based on criteria including paste (ware), surface finish and decoration, and vessel form (in the case of feasting vessels). I also present statistical tests of differences with the resultant p-values for those tests that were deemed significant and are included and discussed in Chapter 8. Ceramic Data From Erongarcuaro and Urichu Below I present raw data from the ceramic samples of Erongarcuaro and Urichu. Table B-1. Ceramic Data of the Total Sherd Sample from Urichu Area 1, categorized by paste and decoration. IC=Ichupio Coarse, QW=Querenda White, SG=Sipiho Gray, TCr=Tarerio Cream, TBr=Tariacuri Brown, TCo=Tariacuri Coarse, TecO=Tecolote Orange, UF=Urichu Fine, YCo=Yaguarato Coarse, YCr=Yaguarato Cream. Cr=Cream Slip, P/Cr=Paint on Cream Slip, N/Cr=Negative on Cream, PN/Cr=Paint and Negative on Cream, Red=Red Slip, P/R=Paint on Red Slip, N/R=Negative on Red Slip, PN/Red=Paint and Negative on Red Slip, BP=Black Polish, Un=Unslipped. Paste Cr P/Cr N/Cr PN/C Red P/R N/R PN/R BP U Sum %/S IC 2 0 0 0 1 0 0 0 0 0 3 .11 QW 45 14 0 1 7 0 0 0 0 3 70 2.47 SG 1039 196 2 30 120 2 1 0 0 32 1422 50.09 TCr 17 31 0 3 15 1 0 1 0 1 69 2.43 TBr 95 17 0 0 26 2 0 0 0 6 146 5.14 TCo 232 10 0 0 61 0 0 0 0 32 225 11.45 TecO 10 7 0 0 6 0 0 0 0 1 24 .85 UF 2 0 0 0 0 0 0 0 0 0 2 .07 YCo 407 14 0 0 39 0 0 0 0 9 469 16.52 YCr 220 53 0 1 28 1 0 0 0 6 309 10.88 Sum 2059 342 2 35 303 6 1 1 0 90 2839 100.01 %/T 72.53 12.05 .07 1.23 10.67 .21 .04 .04 0 3.17 336

PAGE 337

Table B-2. Ceramic Data for the Total Sherd Sample from Area 2 of Urichu. Paste Cr P/Cr N/Cr PN/C Red P/R N/R PN/R BP U Sum %/S IC 12 0 0 0 5 0 0 0 0 12 29 1.53 QW 53 32 0 2 110 0 1 0 0 41 239 12.63 SG 105 35 1 6 63 1 2 2 0 34 249 13.16 TCr 24 39 0 9 33 3 0 0 0 2 111 5.87 TBr 20 52 1 21 53 6 0 0 0 0 153 8.07 TCo 211 1 0 0 152 0 0 0 0 107 471 24.89 TecO 14 0 0 0 10 0 0 0 0 22 43 2.27 UF 4 0 0 0 0 0 0 0 0 7 11 .58 YCo 136 4 0 0 104 0 0 0 110 354 18.71 YCr 76 56 4 9 77 6 0 0 0 4 232 12.26 Sum 655 219 6 47 607 16 4 2 0 306 1892 99.97 %/T 34.62 11.58 .32 2.48 32.08 .85 .21 .11 0 17 .76 100.01 Table B-3. Ceramic Data for the Total Sherd Sample from Area 5 of Urichu. Paste Cr P/Cr N/Cr PN/C Red P/R N/R PN/R BP U Sum %/S IC 44 3 0 0 8 0 0 0 0 3 58 .77 QW 293 239 0 0 112 4 0 0 0 25 663 8.77 SG 1458 238 0 15 270 4 1 0 0 48 2044 27.03 TCr 111 67 0 0 98 1 0 0 0 11 288 3.81 TBr 207 11 1 1 97 1 0 0 0 12 330 4.36 TCo 547 11 0 0 310 0 0 0 0 60 928 12.27 TecO 46 14 1 0 29 0 0 0 0 2 92 1.22 UF 1 0 0 0 0 0 0 0 0 0 1 .01 YCo 1445 63 0 0 142 0 0 0 0 87 1741 23.02 YCr 860 250 0 4 282 1 0 0 0 20 1417 18.74 Sum 5006 906 2 20 1348 11 1 0 0 268 7562 100 %/T 66.20 11.98 .03 .26 17.83 .15 .01 0 0 3.54 100 337

PAGE 338

Table B-4. Aggregate Ceramic Data from fields ER-01, ER-02, ER-07, ER-08 at Erongarcuaro. Paste Cr P/Cr N/Cr PN/C Red P/R N/R PN/R BP U Sum %/S IC 19 0 0 0 3 0 0 0 0 4 26 .45 QW 61 34 0 0 16 0 0 0 0 1 112 1.92 SG 886 150 3 7 321 7 1 0 4 88 1467 25.15 TCr 213 87 2 2 94 20 1 0 1 78 498 8.54 TBr 103 3 1 0 173 1 0 0 0 66 347 5.95 TCo 11 0 0 0 13 0 0 1 0 10 35 .60 TecO 0 1 0 0 2 0 0 0 1 0 4 .07 UF 21 2 0 0 9 0 0 0 0 0 32 .55 YCo 859 22 0 0 664 0 0 0 0 76 1621 27.79 YCr 839 151 1 9 525 34 2 0 4 126 1691 28.99 Sum 3012 450 7 18 1820 62 4 1 10 449 5833 100.01 %/T 51.64 7.71 .12 .31 31.2 1. 06 .07 .02 .17 7.70 100 Table B-5. Classification of sherds by paste and presence of feasting form, negative decoration, or both from the fields ER-01, ER-02, and ER-07 at Erongarcuaro. Paste Feasting Form Negative Decoration Both Sum Percent QW 1 0 0 1 1.12 SG 12 9 2 23 25.84 TCr 3 4 1 8 8.99 TecO 3 0 0 3 3.37 TBr 0 1 0 1 1.12 TCo 0 1 0 1 1.12 YCo 6 0 0 6 6.74 YCr 34 12 0 46 51.69 Sum 59 27 3 89 99.99 Table B-6. Classification of sherds by paste and presence of feasting form, negative decoration, or both from Area 1 at Urichu. Paste Feasting Form Negative Decoration Both Sum Percent QW 2 1 0 3 3.85 SG 20 31 2 53 67.95 TCr 1 4 0 5 6.41 TecO 0 0 0 0 0 TBr 3 0 0 3 3.85 TCo 0 0 0 0 0 YCo 1 0 0 1 1 YCr 11 1 0 12 16.67 Sum 38 37 2 77 100.01 338

PAGE 339

Table B-7. Classification of sherds by paste and presence of feasting form, negative decoration, or both from Area 2 of Urichu. Paste Feasting Form Negative Decoration Both Sum Percent QW 3 3 0 6 8.64 SG 1 11 0 12 14.81 TCr 3 10 0 13 22.22 TecO 0 0 0 0 0 TBr 2 22 0 24 33.33 TCo 3 0 0 3 3.70 YCo 2 0 0 2 2.47 YCr 0 13 0 13 14.81 Sum 14 59 0 73 99.98 Table B-8. Classification of sherds by paste and presence of feasting form, negative decoration, or both from Area 5 of Urichu. Paste Feasting Form Negative Decoration Both Sum Percent QW 2 0 0 2 6.82 SG 8 12 4 24 59.09 TCr 0 0 0 0 0 TecO 2 1 0 3 6.82 TBr 0 2 0 2 4.55 TCo 3 0 0 3 6.82 YCo 3 0 0 3 7 YCr 0 4 0 4 9.09 Sum 18 19 4 41 100.01 Table B-9. Counts of pipe fragments classified by paste from ER-02, ER-03, ER-05. Paste Quantity %/Sum Sipiho Gray 27 50.00 Tarerio Cream 11 20.37 Tecolote Orange 10 18.52 Yaguarato Cream 6 11.11 Sum 54 100.00 Table B-10. Counts of discrete pipes (refit sherds counted as one) classified by paste from ER-02, ER-03, and ER-05. Paste Quantity %/Sum Sipiho Gray 26 52.00 Tarerio Cream 11 22.00 Tecolote Orange 9 18.00 Yaguarato Cream 4 8.00 Sum 50 100.00 339

PAGE 340

Statistical Tests of Difference Below I present the results of statistical tests of difference, with the percent confidence level that the difference is significant. The difference between proportions have been calculated using the formula TS=p1-p2/(p(1-p)(1/n1+1/n2)), where TS is the test statistic to be compared to a t-table with degree of difference of infinity, p1 is one proportion, p2 is the other proportion, n1 is the total sample size of the first proportion, n2 is the sample size of the second proportion, and p equals(n1 x p1 +n2 x p2)/(n1+n2) (Fletcher and Lock 2005:99). This test is appropriate so long as n1 and n2 are equal to or greater than 20 (Fletcher and Lock 2005:99). Other tests of the significance of the difference between proportions include the Fisher Exact test, which can accommodate smaller sample sizes, as well as the Pearsons Chi-Square and resultant exact p-values of the likelihood that the proportions are the result of the same processes and practices. 340

PAGE 341

Table B-11. Statistical Tests of Difference for Proportions among different sherd samples. Difference Tested Quantities of Difference Difference of Proportion (percent confidence level) Fisher Exact (two-sided p-value) Pearsons Chi-Square and p-value ER TecO pipe frags vs. ER TecO sherds 10/54 vs. 4/5833 27.67 (99.9% diff) 0 767.683 (p=0.0000) ER TecO ritual (pipes+mini-bowls) vs. ER TecO Feasting (form+negative decoration, no mini-bowls) 12/61 vs. 1/82 3.7963 (99.9% diff) 0.0007 11.799 (p=0.0005) ER TecO pipe frags vs. ER TecO Feasting (incl mini-bowls) 10/54 vs. 3/89 3.0606 (99% diff) 0.0079 7.553 (p=0.0059) U1 SG vs. ER SG 1422/2839 vs. 1467/5833 23.31 (99.9% diff) 0 253.008 (p=0.0000) U1 SG Feasting Forms vs. U1 SG Negative Decoration 22/37 vs. 33/39 2.4522 (98% diff) 0.0206 6.009 (p=0.0142) U2 SG Negative Decoration vs. U2 SG Feasting Forms 11/59 vs. 1/15 1.1232 (not significant at 90% confidence level) 0.4388 1.263 (p=0.1643) U5 SG Negative Decoration vs. U5 SG Feasting Forms 16/23 vs. 13/26 1.3910 (not significant at 90% confidence level) 0.2450 1.934 (p=0.1643) ER SG Negative Decoration vs. ER SG Feasting Forms 11/29 vs. 11/34 0.46 (not significant at 90% confidence level) 0.7920 0.214 (p=0.6434) U1 TBr Feasting Forms vs. U1 TBr Negative Dec 3/37 vs. 0/39 1.8099 (90% diff) 0.1105 4.832 (p=0.0279) 341

PAGE 342

U2 TBr Negative Decoration vs. U2 TBr Feasting Forms 22/59 vs. 5/15 0.2845 (not significant at (90% confidence level) 1 0.081 (p=0.7763) U5 TBr Feasting Forms vs. U5 TBr Negative Decoration 4/26 vs. 2/23 0.7082 (not significant at 90% confidence level) 0.6707 0.508 (p=0.4759) ER TBr Negative Decoration vs. ER TBr Feasting Forms 2/29 vs. 0/34 1.5570 (not significant at 90% confidence level) 0.2079 3.501 (p=0.0613) 342

PAGE 343

LIST OF REFERENCES Acosta, J. (1939). Exploraciones arqueolgicas realizadas en el estado de Michoacn durante los aos de 1937 y 1938. Revista Mexicana de Estudios Antropolgicos 3(2):85-99. Acosta, N. R., and Uruuela, G. (1997). Patrones de enterramiento en la Cuenca de Sayula: la fase Amacueca en Atoyac. In Malvido, E., Pereira, G., and Tiesler, V. (eds.), El Cuerpo humano y su tratamiento mortuorio, Coleccin Cientfica, Instituto Nacional de Antropologa, Centre dEtudes Mexicaines et Centramricaines, Mexico City, pp. 179-191. Alcal, Fr. J. de (2000). Relacin de Michoacn coordinated by Moiss Franco Mendoza, with paleography by Clotilde Martnez Ibez y Carmen Molina Ruiz, El Colegio de Michoacn, Gobierno del Estado de Michoacn, Zamora, Mexico Ames, K. (1995). Chiefly Power and Household Production on the Northwest Coast. In Price, T. D., and Feinman, G. M., Foundations of Social Inequality, Plenum, New York, pp. 155-187. Appadurai, A. (1986). Introduction: Commodities and the politics of value. In Appadurai, A. (ed.), The Social Life of Things: Commodities in Cultural Perspective, Cambridge University Press, Cambridge, pp. 3-63 Apthorpe, R. (1984) Hierarchy and Other Social Relations: Some Categorial Logic. In Galey, J.-F. (ed.), Diffrences, Valeurs, Hirarchie: Textes Offerts Louis Dumont, Editions de lEcole des Hautes Etudes en Sciences Sociales, Paris, pp. 283-298. Arnauld, M.-C., and Faugre-Kalfon, B. (1998). Evolucin de la ocupacin humana en el Centro-Norte de Michoacn y la emergencia del Estado Tarasco. In Darras, V. (ed.), Gnesis, culturas, y espacios en Michoacn, Centre dEtudes Mexicaines et Centramricaines, Mexico, pp. 13-34. Bailey, F. G. (1991). Religion and Religiosity: Ideas and Their Use. Contributions to Indian Sociology (n.s.) 25(2):211-231. Baines, J., and Yoffee, N. (1998). Order, Legitimacy, and Wealth in Ancient Egypt and Mesopotamia. In Feinman, G. and Marcus, J. (eds.), Archaic States, SAR Press, Santa Fe, pp. 199-260. Baines, J., and Yoffee, N. (2000). Order, legitimacy, and wealth: setting the terms. In Richards, J. and Van Buren, M. (eds.) Order, Legitimacy, and Wealth in Ancient States, Cambridge University Press, Cambridge pp. 13-17. Barnes, R. H. (1980). Marriage, Exchange and the Meaning of Corporations in Eastern Indonesia. In Comaroff, J. L. (ed.), The Meaning of Marriage Payments, Academic Press, New York, pp. 93-122. 343

PAGE 344

Barrett, J. C. (1990). The Monumentality of Death: The Character of Early Bronze Age Mortuary Mounds in Southern Britain. World Archaeology 22(2):179-189. Barrett, J. C. (2000). A thesis on agency. In Dobres, M.-A. and Robb, J. E. (eds.), Agency in Archaeology, Routledge, London, pp. 61-68. Battaglia, D. (1990) On the Bones of the Serpent: person, memory and morality in Sabarl Island society. University of Chicago Press, Chicago. Battaglia, D. (1992) The Body in the Gift: Memory and Forgetting in Sabarl Mortuary Exchange. American Ethnologist 19(1):3-18. Beltrn, U. (1986). Estado y sociedad tarascos. In Carrasco, P. et al., La Sociedad indgena en el centro y occidente de Mxico, El Colegio de Michoacn, Zamora, pp. 45-62. Berdan, Frances F., et al. (1996) Aztec Imperial Strategies. Dumbarton Oaks, Washington D.C.. Berreman, G. (1971). The Brahmanical View of Caste. Contributions to Indian Sociology (n.s.) 5:16-23. Beteille, A. (1987). Reply to Dumont 1987. Current Anthropology 28(5):672-677. Binford, L. R. (1971). Mortuary practices: their study and their potential. In Brown, J. A. (ed.), Approaches to the Social Dimensions of Mortuary Practices,. Memoirs of the Society for American Archaeology 25. Society for American Archaeology, Washington D.C., pp. 6-29 Blanton, R., and Feinman, G. (1984). The Mesoamerican World System. American Anthropologist 86:673-682. Blanton, R. E., Kowalewski, S. A., Feinman, G. M. and Finsten, L. M. (1993). Ancient Mesoamerica: a Comparison of Change in Three Regions, Second Edition, Cambridge University Press, Cambridge. Blanton, R. E., Feinman, G. M., Kowalewski, S. A., and Nicholas, L. M. (1999) Ancient Oaxaca: the Monte Alban State, Cambridge University Press, Cambridge. Blanton, R. E., Feinman, G. M., Kowalewski, S. A., and Peregrine, P. N. (1996) A Dual Processual Theory for the Evolution of Mesoamerican Civilization. Current Anthropology 37(1):1-14. Bloch, M. (1971). Placing the Dead: tombs, ancestral villages and kinship organization in Madagascar, Seminar Press, New York. 344

PAGE 345

Bourdieu, P. (1977). Outline of a Theory of Practice, Nice, R. (transl.), Cambridge University Press, Cambridge. Bourdieu, P. (1984). Distinction: A Social Critique of the Judgement of Taste, Nice, R. (transl.), Harvard University Press, Cambridge. Bourdieu, P. (1999). Rethinking the State: Genesis and Structure of the Bureaucratic Field. In Steinmetz, G. (ed.), State/Culture: state formation after the cultural turn, Cornell University Press, Ithaca, pp. 53-75. Brand, D. D. (1943). An historical sketch of geography and anthropology in the Tarascan region, pt. 1, Southwest Journal of Anthropology, 6-7: 37-108. Brand, D. D. (1971). Ethnohistoric Synthesis of Western Mexico. In Archaeology of Northern Mesoamerica, Part 2, The Handbook of Middle American Indians, Ekholm, G. F., and Bernal, I. (eds.), University of Texas Press, Austin, pp. 632-656. Bray, T. L. (2003). To Dine Splendidly: Imperial Pottery, Commensal Politics, and the Inca State. In Bray, T. (ed.), The archaeology and politics of food and feasting in early states and empires, Kluwer Academic, New York, pp.93-142. Brck, J. (2005). Material metaphors: the relational construction of identity in Early Bronze Age burials in Ireland and Britain. Journal of Social Archaeology 4(3):307-333. Brumfiel, E. M. (1983). Aztec State Making: Ecology, Structure, and the Origin of the State. American Anthropologist 85:261-284. Brumfiel, E. M. (2000). The Politics of High Culture: Issues of Worth and Rank. In Richards, J. and Van Buren, M. (eds.), Order, Legitimacy, and Wealth in Ancient States, Cambridge University Press, Cambridge, pp. 131-139. Brumfiel, E. M., and Earle, T. K. (1987). Specialization, exchange, and complex societies: an introduction. In Brumfiel, E. M., and Earle, T. K. (eds.), Specialization, exchange, and complex societies, Cambridge University Press, Cambridge, pp. 1-9. Cabrera Castro, R. (1977-1978). Exploraciones en Tzintzuntzan, Michoacn: Dcima Temporada de excavaciones (notas de campo). Cabrera Castro, R. (1987). Tzintzuntzan: dcima temporada de excavaciones. In Homenaje a Romn Pia Chan, Serie Antropolgica 79. Universidad Nacional Autnoma de Mexico, Instituto Nacional de Antropologa e Historia, Mexico, pp. 531-65. Cabrera Castro, R. (1996). Cermica suntuaria de Tzintzuntzan, Michoacn. In Crespo, A. M., and Viramontes, C. (eds.) Tiempo y Territoria en Arqueologa: el Centro-Norte de Mxico, Coleccin Cientfica, 323. Instituto Nacional de Antropologa e Historia, Mexico City. 345

PAGE 346

Cahue, L., Sauer, N. J., and Pollard, H. P. (1998). Lip plug (bezote) abrasion facets in a Tarascan burial from Urichu, Michoacn (abstract). American Journal of Physical Anthropology Supplement 26:117 Calnek, E. E. (1982). Patterns of Empire Formation in the Valley of Mexico, Late Postclassic Period. In Collier, G. A., Rosaldo, R. I., and Wirth, J. D. (eds.), The Inca and Aztec States, 1400-1800: Anthropology and History, Academic Press, New York, pp. 43-62. Carrasco, P. (1971). Social Organization of Ancient Mexico. In Ekholm, G. F., and Bernal, I. (ed.), Archaeology of Northern Mesoamerica, Part 1, Handbook of Middle American Indians Vol. 10, University of Texas Press, Austin, pp. 349-375. Carrasco, P. (1986). Economa poltica en el reino Tarasco. In Carrasco, P. et. al, La Sociedad indgena en el centro y occidente de Mxico, El Colegio de Michoacn, Zamora, pp. 63-102. Castro Leal, M. (1986). Tzintzuntzan: Capital de los Tarascos. Gobierno del Estado de Michoacn, Morelia. Chadwick, R. (1971). Archaeological synthesis of Michoacan and adjacent regions. In Ekholm, G. F., and Bernal, I. (eds.), Handbook of Middle American Indians Vol. 11, Archaeology of Northern Mesoamerica, Part 2, University of Texas Press, Austin, pp. 657-693. Chase-Dunn, C., and Hall, T. D. (1997). Rise and Demise: Comparing World-systems, Westview Press, Boulder, CO. Chapman, J. (2000). Tension at funerals: Social practices and the subversion of community structure in later Hungarian prehistory. In Dobres, M.-A., and Robb, J. E. (eds.) Agency in Archaeology, Routledge, London, pp. 169-195. Chase, A. F., and Chase, D. Z. (1996). More Than Kin and King: Centralized Political Organization among the Late Classic Maya. Current Anthropology 37(5):803-810. Clark, J., and Blake, M. (1994). The power of prestige: Competitive generosity and the emergence of rank societies in lowland Mesoamerica. In Brumfiel, E., and Fox, J. (eds.), Factional Competition and Political Development in the New World, Cambridge University Press, Cambridge, pp. 17. Clarke, D. L. (1972). Models and paradigms in contemporary archaeology. In Clarke, D. L. (ed.), Models in Archaeology, Methuen & Co., London, pp. 1-60. Comaroff, J. L. (1980). Introduction. In Comaroff, J. L. (ed.), The Meaning of Marriage Payments, Academic Press, New York, pp. 1-47. 346

PAGE 347

Coppet, D. de (1992). Introduction. In de Coppet, D. (ed.), Understanding Rituals, Routledge, New York. Corona Nez, J. (1957). Mitologa Tarasca. Fondo de Cultura y Econmica, Mexico City. Costin, C. L. (2001). Craft Production Systems. In Feinman, G. M., and Price, T. D. (eds.), Archaeology at the Millennium: A Sourcebook, Kluwer Academic, New York pp. 273. Darras, V. (1999). Tecnologas Prehispanicas de la Obsidiana: Los Centros de Produccin de la Regin de Zinparo-Prieto, Cuadernos de Estudios Michoacanos 9, Centre dEtudes Mexicaines et Centramericaines, Mexico City. Das, V., and Uberoi, J. S. (1971). The Elementary Structure of Caste. Contributions to Indian Sociology (n.s.) 5:33-43. Demarest, A. A. (1992). Ideology in Ancient Maya Cultural Evolution: the dynamics of galactic polities. In Demarest, A. A., and Conrad, G. W. (eds.), Ideology and pre-Columbian Civilizations, SAR Press, Santa Fe, pp. 135-157. Dietler, M. (2001). Theorizing the Feast: Rituals of Consumption, Commensal Politics, and Power in African Contexts. In Dietler, M., and Hayden, B. (eds.), Feasts: archaeological and ethnographic perspectives on food, politics and power, Smithsonian Institution Press, Washington D.C., pp. 65-114. Dietler, M. (2003). Clearing the Table: Some Concluding Reflections on Commensal Politics and Imperial States. In Bray, T. L. (ed.), The archaeology and politics of food and feasting in early states and empires, Kluwer Academic, New York, pp. 271-283. Dobres, M.-A., and Robb, J. E. (2000). Agency in Archaeology, Paradigm or Platitude? In Dobres, M.-A., and Robb, J. E. (eds.), Agency in Archaeology, Routledge, London, pp. 3-17. Dornan, J. (2002). Agency and Archaeology: Past, Present, and Future Directions. Journal of Archaeological Method and Theory 9(4):303-329. Dumzil, G. (1949). LHeritage Indo-Europeens Rome. Fourth Edition. Gallimard Press, Paris. Dumont, L. (1979). The Anthropological Community and Ideology. Social Science Information 18(6):785-817. Dumont, L. (1980). Homo Hierarchicus: the Caste System and its Implications, Revised English Edition, University of Chicago Press, Chicago. 347

PAGE 348

Dumont, L. (1986). Essays on Individualism: Modern Ideology in Anthropological Perspective, University of Chicago Press, Chicago. Feeley-Harnik, G. (1985). Issues in Divine Kingship. Annual Review of Anthropology 14:273-313. Feinman, G. M., Blanton, R. E., and Kowalewski, S. A. (1983). Market System Development in the Prehispanic Valley of Oaxaca, Mexico. In Hirth, K. G. (ed.), Trade and Exchange in Early Mesoamerica, University of New Mexico Press, Albuquerque, pp. 157-178. Feinman, G. M., Upham, S., and Lightfoot, K. G. (1981). The Production Step Measure: an Ordinal Index of Labor Input in Ceramic Manufacture. American Antiquity 46(4):871-884. Fisher, C. (2005). Abandoning the Garden: Demographic and Landscape Change in the Lake Ptzcuaro Basin, Mexico. American Anthropologist 107(1): 87-95 Fisher, C. T., Pollard, H. P., and Frederick, C. (1999). Intensive Agriculture and Sociopolitical Development in the Lake Ptzcuaro Basin, Michoacn, Mexico. Antiquity 73:642-649. Fisher, C., Pollard, H. P., Israde-Alcntara, I., Garduo-Munroy, V. H., and Banerjee, S. (2003). Reexamination of Human Induced Environmental Change Within the Lake Ptzcuaro Basin, Michoacn, Mexico. Proceedings of the National Academy of Sciences 100(8):4957-4962. Flannery, K. V. (1972). The Cultural Evolution of Civilizations. Annual Review of Ecology and Systematics 3(399-426. Flannery, K. V. (1999). Process and agency in early state formation. Cambridge Archaeological Journal 9: 3. Fletcher, M., and Lock, G. R. (2005). Digging Numbers: Elementary Statistics for Archaeologists, Oxford University School of Archaeology Monograph, Oxford. Fogelson, R. D. (1989). The Ethnohistory of Events and Nonevents. Ethnohistory 36:133147. Fowler, C. (2004). The Archaeology of Personhood: an Anthropological Approach, Routledge, London. Fox, J. W., Cook, G.W., Chase, A. F., and Chase, D. Z. (1996). Questions of Political and Economic Integration: Segmentary versus Centralized States among the Ancient Maya. Current Anthropology 37(5):795-800. 348

PAGE 349

Gaitn Amman, F. (2005). With a Hint of Paris in the Mouth: Fetishized Toothbrushes or the Sensuous Experience of Modernity in late 19 th Century Bogot. In Meskell, L. (ed.), Archaeologies of Materiality, Blackwell Publishers, Malden, pp. 71-95. Gal, R. (1946). Arqueologa de Tzintzuntzan. Anales del Museo Michoacano 2(4):5062. Garca Alcaraz, A. (1976). Estratificacin social entre los tarascos prehispnicos. In Carrasco, P., and Broda, J. et. al, Estratificacin social en la Mesoamrica prehispnica, Centro de Investigaciones y Estudios Superiores en Antropologa Social, Instituto Nacional de Antropologa e Historia, Mexico City, pp. 221-44. Gell, A. (1992). The enchantment of technology and the technology of enchantment. In Coote, J. and Shelton, A. (eds.), Anthropology, Art, and Aesthetics, Clarendon Press, Oxford, pp. 40-63. Gell, A. (1993). Wrapping in Images: Tattooing in Polynesia, Oxford University Press, New York. Gell, A. (1998). Art and Agency: An Anthropological Theory, Clarendon Press, Oxford. Gell, A. (1999). Strathernograms, or the Semiotics of Mixed Metaphors. In E. Hirsch (ed.) The Art of Anthropology: Essays and Diagrams, Athlone, London, pp. 29. Gibson, C. (1971). Structure of the Aztec empire. In Ekholm, G. F., and Bernal, I. (eds.), Archaeology of Northern Mesoamerica, Part 1, Handbook of Middle American Indians Vol. 10, University of Texas Press, Austin, pp. 376-394. Giddens, A. (1984). The Constitution of Society: Outline of the Theory of Structuration, University of California Press, Berkeley. Gilberti, Fr. M. (1962[1559]). Diccionario de la Lengua Tarascan o de Michoacn, Ugarte, J. B., preliminary note, Coleccin Siglo XVI (9), Mexico City. Gillespie, S. D. (1989). The Aztec Kings: the construction of rulership in Mexican history. University of Arizona Press, Tucson. Gillespie, S. D.(1999). Olmec Thrones as Ancestral Altars: The Two Sides of Power. In Robb, J. E. (ed.), Material Symbols: Culture and Economy in Prehistory, Southern Illinois University Press, Carbondale, pp. 224-253. Gillespie, S. D. (2001). Agency, Personhood, and Mortuary Ritual: A Case Study from the Ancient Maya. Journal of Anthropological Archaeology 20(1):73-112. 349

PAGE 350

Gillespie, S. D. (2009). Embodied Persons and Heroic Kings in Late Classic Maya Imagery. In Boric, D., and Robb, J. E. (eds.), Past Bodies: Body-Centred Research in Archaeology, Oxbow Press, Oxford (UK), pp. 125-134. Glascock, M. (2006). Report submitted to Dr. Helen Pollard and the author, March, 2006. File in possession of the author. Goody, J. (1962). Death, property and the ancestors: a study of the mortuary customs of the LoDagaa of West Africa, Stanford University Press, Stanford. Gorenstein, S. S. (1985). Acmbaro: Frontier Settlement on the Tarascan-Aztec Border, Vanderbilt University Publications in Anthropology 32, Nashville. Gorenstein, S. S. (1993). Introduction. In Pollard, H. P., Taracuris Legacy: the Prehispanic Tarascan State, University of Oklahoma Press, Norman, pp. xiii-xx. Gorenstein, S. S., and Pollard, H. P. (1983). The Tarascan Civilization: a Late Prehispanic Cultural System. Vanderbilt University Publications in Anthropology, 28, Nashville. Gosden, C. (1994). Social Being and Time, Blackwell, Oxford. Gosden, C. (1999). Anthropology and Archaeology: a changing relationship. Routledge, London. Gosden, C. (2001). Making Sense: Archaeology and Aesthetics. World Archaeology 33(2):163-167. Gosden, C., and Marshall, C. (1999). The Cultural Biography of Objects. World Archaeology 31(2):169-178. Gregory, C. (1982). Gifts and Commodities, Academic Press, London. Harrison, R. (2003). The Magical Virtue of These Sharp Things: Colonialism, Mimesis, and Knapped Bottle Glass Artefacts in Australia. Journal of Material Culture 8(3):311-336. Hassig, R. (1985). Trade, Tribute, and Transportation: the Sixteenth-Century Political Economy of the Valley of Mexico, University of Oklahoma Press, Norman. Hassig, R. (1988). Aztec Warfare: Imperial Expansion and Political Control, University of Oklahoma Press, Norman. Healan, D. M. (1997). Pre-Hispanic quarrying in the Ucareo-Zinapcuaro obsidian source area. Ancient Mesoamerica 8: 77-100. 350

PAGE 351

Healan, D. M., and Hernndez, C. E. (1999). Asentamiento Prehispnico y Cronologa Cermica en el Noreste de Michoacn. In Williams, E., and Weigand, P. C. (eds.), Arqueologa y Etnohistoria: La Region del Lerma, El Colegio de Michoacn, Zamora, Mexico, pp. 133-155. Heckenberger, M. J. (2005). The Ecology of Power: culture, place, and personhood in the southern Amazon, A.D. 1000-2000, Routledge, New York. Helms, M. (1988). Ulysses Sail: an ethnographic odyssey of power, knowledge, and geographical distance, Princeton University Press, Princeton. Helms, M. (1993). Craft and the Kingly Ideal: art, trade, and power, University of Texas Press, Austin. Hertz, R. (1960[1907]). Death and the Right Hand, Needham, R., and Needham, C. (transl.), Free Press, Glencoe. Heusch, L. de (1982). The Drunken King, or, the Origin of the State. Willis, R. (transl.). University of Indiana Press, Bloomington. Hirshman, A. J. (2003). A Case Study in the Relationship Between Social Complexity and the Organization of Ceramic Production from the Lake Ptzcuaro Basin, Michoacn, Mexico. Ph.D. Dissertation, Department of Anthropology, Michigan State University, East Lansing. Hirshman, A. J., in press. Tarascan Ceramic Production and Implications for Ceramic Distribution, accepted for publication in the journal Ancient Mesoamerica, fall 2008. Hirth, K. G. (1999). The Distributional Approach: A New Way to Identify Marketplace Exchange in the Archaeological Record. Current Anthropology 39:451-476. Hocart, A. M. (1969). Kingship. Oxford University Press, Oxford. Hocart, A. M. (1970). Kings and Councillors. University of Chicago Press, Chicago. Hodder, I. (1986). Reading the Past: current approaches to interpretation in archaeology. Cambride University Press, Cambridge. Hodge, M. G. (1991). Land and Lordship in the Valley of Mexico: The Politics of Aztec Provincial Administration. In Harvey, H. R. (ed.), Land and Politics in the Valley of Mexico, University of New Mexico Press, Albuquerque, pp. 113-140. Hodge, M. G. (1994). Polities Comprising the Aztec Empires Core. In Hodge, M. G., and Smith, M. E. (eds.), Economies and Polities in the Aztec Realm, Institute for Mesoamerican Studies, Albany, pp. 43-71. 351

PAGE 352

Hosler, D. (1994). The Sounds and Colors of Power: the Sacred Metallurgical Technology of Ancient West Mexico, Massachusetts Institute of Technology Press, Cambridge. Hosler, D. (1995). Sound, Color and Meaning in the Metallurgy of Ancient West Mexico. World Archaeology 27(1):100-115. Houston, S. D., and Cummins, T. (2001). Body, Presence, and Space in Andean and Mesoamerican Rulership. In Evans, S. T., and Pillsbury, J. (eds.), Palaces of the Ancient New World, Dumbarton Oaks, Washington D.C., pp. 359-398. Huntington, R., and Metcalf, P. (1979). Celebrations of Death: the Anthropology of Mortuary Ritual, Cambridge University Press, Cambridge. Ingold, T. (1993). The Temporality of the Landscape. World Archaeology 25(2):152-174. Inomata, T. (2001). The Power and Ideology of Artistic Creation: Elite Craft Specialists in Classic Maya Society. Current Anthropology 321-349. Janusek, J. W. (2004). Identity and Power in the Ancient Andes: Tiwanaku Cities Through Time, Routledge, London. Johnson, M. H. (1989). Conceptions of Agency in Archaeological Interpretation. Journal of Anthropological Archaeology 8:189-211. Jones, A. (2005). Lives in fragments? Personhood and the European Neolithic. Journal of Social Archaeology 5(2):193-224. Jones, A. (2007). Memory and Material Culture, Cambridge University Press, Cambridge. Joyce, R. A. (2000). High culture, Mesoamerican civilization, and the classic Maya tradition. In Richards, J., and Van Buren, M. (eds.), Order, legitimacy, and wealth in ancient states, Cambridge University Press, Cambridge, pp. 64-76. Joyce, R. A. (2001). Burying the Dead at Tlatilco: Social Memory and Social Identities. Archeological Papers of the American Anthropological Association 10:12. Joyce, R. A., and Lopiparo, J. (2005). Postscript: Doing Agency in Archaeology. Journal of Archaeological Method and Theory 12(4):365-374 Keane, W. (1997). Signs of Recognition: Powers and Hazards of Representation in an Indonesian Society, University of California Press, Berkeley. Keane, W. (2005). Signs are not the Garb of Meaning: On the Social Analysis of Material Things. In Miller, D. (ed.) Materiality, Duke University Press, Durham, pp. 182-205. 352

PAGE 353

Kirchhoff, P. (1952). Mesoamerica: its geographic limits, ethnic composition, and cultural character. In Tax, S. (ed.), Heritage of Conquest: the Ethnology of Middle America, Macmillan, New York, pp. 17-30. Kirchhoff, P. (1956). Estudio Preliminar: La Relacin de Michoacn como fuente para la historia de la sociedad y cultura Tarasca. In Relacin de las ceremonas y ritos y poblacin y gobierno de los indios de la provincia de Michoacn, reproduccin facsimilar del Ms IV de El Escorial, Madrid, transcription, prologue, introduction, and notes by Jos Tudela, Aguilar Publicistas, Madrid. Knsel, C. J. (2006). Of No More Use to Men Than in Ages Before?: The Investiture Contest as a Model for Funerary Interpretation. In Gowland, R., and Knsel, C. J. (eds.), Social Archaeology of Funerary Remains, Oxbow Books, Oxford, pp. 209-223. Kopytoff, I. (1986). The cultural biography of things: commoditization as process. In Appadurai, A. (ed.), The Social Life of Things: Commodities in Cultural Perspective, Cambridge University Press, Cambridge, pp. 64-94. Krippner-Martnez, J. (2001). Rereading the Conquest: Power, Politics, and the History of Early Colonial Michoacn, Mexico, 1521-1565, Pennsylvania State University Press, University Park. Kchler, S. (1991). Malangan: Art and Memory in a Melanesian Society. Man (n.s.) 22(2):238-255. Kuijt, I. (1999). Place, Death, and the Transmission of Social Memory in Early Agricultural Communities of the Near Eastern Pre-Pottery Neolithic. Archaeological Papers of the American Anthropological Association 10:80-99. Kuper, A. (1982). Wives for Cattle: bridewealth and marriage in Southern Africa, Routledge, London. Kus, S. M., and Raharijaona, V. (2001). To Dare to Wear the Cloak of Another Before Their Very Eyes: State Co-optation and Local Re-appropriation in Mortuary Rituals of Central Madagascar. Archaeological Papers of the American Anthropological Association 10:114-131. Kuthy-Saenger, M. de L. (1996). Strategies of Survival, Accommodation and Innovation: the Tarascan Indigenous Elite in Sixteenth Century Michoacn, unpublished Ph.D. dissertation, Michigan State University. Kuthy, M. de L. (2003). El Control de los Puestos Polticos: la elite Tarasca en el Siglo XVI. In Paredes Martnez, C., and Tern, M. (coords.), Autoridad y Gobierno Indgena en Michoacn, El Colegio de Michoacn, Centro de Investigaciones y Estudios Superiores en Antropologa Social, Instituto Nacional de Antropologa e Historia, Universidad Michoacana, Zamora, Mexico, pp. 153-172 353

PAGE 354

Lagunas, Fr. J. B. de (1983[1574]). Arte y Diccionario con otras obras en lengua michuacana. Warren, J. B., (ed.), Fimax Publicistas, Morelia, Mexico. Leach, E. R. (1965). Political Systems of Highland Burma: a Study of Kachin Social Structure, Beacon Press, Boston. Lele, V. (2006). Material habits, identity, semeiotic. Journal of Social Archaeology 6(1):48-70. Looper, M. (2003). From Inscribed Bodies to Distributed Persons: Contextualizing Tairona Figural Images in Performance. Cambridge Archaeological Journal 13(1):25-40. Lpez Austin, A. (1981). Tarascos y Mexicas, Fondo de Cultura y Econmica, Mexico City. Lpez Sarrelangue, D. E. (1965). La Nobleza Indgena de Ptzcuaro en la Epoca Virreinal, Instituto de Investigaciones Histricas, Universidad Nacional Autnoma de Mexico, Mexico City. Lynch, O. (1977). Method and Theory in the Sociology of Louis Dumont: A Reply. In The New Wind: Changing Identities in South Asia. Mouton, the Hague. Macas Goytia, A. (1989a). Los entierros de un centro ceremonial tarasco. In Estudios de Antropologa Biolgica, serie antropolgica, vol. 100, Universidad Nacional de Autnoma de Mxico, Mexico City, pp. 531-559. Macas Goytia, A. (1989b). Proyecto cuenca de Cuitzeo. Boletn del Consejo de Arqueologa, Instituto Nacional de Antropologa e Historia, Mexico City, pp. 71-81. Macas Goytia, A. (1990). Huandacareo: Lugar de Juicios, tribunal, Coleccin Cientfica, Number 222, Instituto Nacional de Antropologa e Historia, Mexico City. Macas Goytia, A. (1996). Una presencia tarasca en Cuitzeo. In Barba de Pia Chan, B. (coord.), Estudios del Mxico Antiguo, Coleccin Cientfica, Instituto Nacional de Antropologa e Historia, Mexico City, pp. 79-108. Marcus, J., and Flannery, K. V. (1996). Zapotec Civilization: How Urban Society Evolved in Mexicos Oaxaca Valley. New York: Thames and Hudson. Martnez Baracs, R. (1997). El Vocabulario en Lengua de Mechuacan (1559) de Fray Maturino Gilberti Como Fuente de Informacin Histrica. In Paredes Martnez, C. (coord.), Lengua y Etnohistoria Purpecha: Homenaje a Benedict Warren, Universidad Michoacana and Centro de Investigaciones y Estudios Superiores en Antropologa Social, Morelia, Mexico, pp. 67-162. 354

PAGE 355

Martnez Baracs, R. (2003). Etimologas Polticas Michoacanas. In Paredes Martnez, C., and Tern, M. (coords.), Autoridad y Gobierno Indgena en Michoacn, El Colegio de Michoacn, Centro de Investigaciones y Estudios Superiores en Antropologa Social, Instituto Nacional de Antropologa e Historia, Universidad Michoacana, Zamora, Mexico, pp. 61-90. Martnez Baracs, R. (2005). Convivencia y Utopia: el gobierno indio y espaol de la ciudad de Mechuacan, 1521-1580. Instituto Nacional de Antropologa e Historia and Fondo de Cultura y Econmica, Mexico City. Martnez Baracs, R., and Espinosa Morales, L. (1999). La Vida Michoacana en el Siglo XVI: Catlogo de los documentos del siglo XVI del Archivo Histrico de la Ciudad de Ptzcuaro, Coleccin Fuentes, Instituto Nacional de Antropologa e Historia, Mexico City. Mateos Higuera, S. (1948). La pictografa tarasca. In El Occidente de Mxico, Sociedad Mexicana de Antropologa, Mexico City, pp. 160-174. Mauss, M. (1954[1925]). The Gift. Cunnison, I. (transl.), Free Press, Glencoe. Meskell, L. (2005). Introduction: Object Orientations. In Meskell, L. (ed.), Archaeologies of Materiality, Blackwell Publishers, Malden, pp. 1-17. McHugh, F. (1997). Theoretical and Quantitative Approaches to the Study of Mortuary Practice, BAR International Series, 785, Oxford. McKinnon, S. (1991). From a Shattered Sun: Hierarchy, Gender, and Alliance in the Tanimbar Islands, University of Wisconsin Press, Madison. Michelet, D. (ed.) (1992). El Proyecto Michoacn 1983-1987. Medio Ambiente e introduccin a los trabajos arqueolgicos, Centre dEstudes Mexicaines et Centramricaines, Mexico City. Michelet, D. (1996). El origen del reino tarasco protohistrico. Arqueologa Mexicana 19:24-27. Michelet, D., Ichon, A. and Migeon, G. (1988). Residencias, barrios, y sitios posclsicos en el Malpas de Zacapu, Instituto Nacional de Antropologa e Historia, Quertaro. Michelet, D., Pereira, G., and Migeon, G. (2005). La llegada de los Uacsechas a la region de Zacapu, Michoacn: datos arqueologcos y discussion. In Manzanilla, L. (ed.), Reacomodos demogrficos del Clsico al Posclsico en el centro de Mxico, Universidad Nacional Autnoma de Mxico, Mexico City, Mexico, pp. 137-153. Migeon, G. (1998). El Poblamiento del Malpas de Zacapu y de sus alrededores, del Clsico al Posclsico. In Darras, V. (ed.), Gnesis, culturas, y espacios en Michoacn, Centre dEtudes Mexicaines et Centramericaines, Mexico City, pp. 35-45. 355

PAGE 356

Miller, D. (1987). Material Culture and Mass Consumption, Blackwell, Oxford. Miller, D. (2005). Materiality: an Introduction. In Materiality, Miller, D. (ed.), Duke University Press, Durham, pp. 1-50. Moedano, H. (1941). Estudio preliminary de la cermica de Tzintzuntzan. Temporada III. Revista Mexicana de Estudios Antropolgicos 5(1):21-42. Mosko, M. S. (1985). Quadripartite Structures: categories, relations, and homologies in Bush Mekeo Culture, Cambridge University Press, Cambridge. Mosko, M. S. (1992). Motherless Sons: 'Divine Kings' and 'Partible Persons' in Melanesia and Polynesia. Man (n.s.) 27(4):697-717. Munn, N. D. (1986). The Fame of Gawa, Cambridge University Press, Cambridge. Noguera, E. (1931). Exploraciones arqueolgicas en las regions de Zamora y Ptzcuaro, estado de Michoacn. Anales del Museo Nacional de Mxico 4(7):88-104. Oliveros, J. A. (1974). Nuevas exploraciones en El Opeo, Michoacn. In Bell, B. (ed.), The Archaeology of West Mexico, Sociedad de Estudios Avanzados del Occidente de Mxico, Ajijic, Jalisco, pp. 182-201. Orenstein, H. (1980). Asymmetrical Reciprocity: a Contribution to the Theory of Political Legitimacy. Current Anthropology 21(1):69-91. Ortner, S. B. (1984). Theory in Anthropology since the Sixties. Comparative Studies in Society and History 26(1):126-166. Otis Charlton, C. L. (1993). Obsidian as jewelry: lapidary production in Aztec Otumba, Mexico. Ancient Mesoamerica 4(2):231-243. Pader, E.-J. (1980). Material Symbolism and Social Relations in Mortuary Studies. In Rahtz, P., Dickinson, T., and Watts, L. (eds.), Anglo-Saxon Cemeteries, Oxford, BAR British Series, 82, pp. 143-159. Parker Pearson, M. (1982). Mortuary practices, society and ideology: an ethnoarchaeological study. In Hodder, I. (ed.), Symbolic and Structural Archaeology, Cambridge University Press, Cambridge, pp. 99-113. Parkin, R. (2003). Louis Dumont and Hierarchical Opposition, Berghahn Books, New York Parmentier, R. J. (1987). The Sacred Remains: Myth, History, and Polity in Belau, University of Chicago Press, Chicago. 356

PAGE 357

Parry, J. (1986). The Gift, The Indian Gift, and the Indian Gift. Man (n.s.) 21(3):453473. Peebles, C. S., and Kus, S. M. (1977). Some Archaeological Correlates of Ranked Societies. American Antiquity 42:421-448. Peregrine, P. M., and Feinman, G. M. (eds.) (1996). Pre-Columbian World Systems. Prehistory Press, Madison, WI. Pereira, G. (n.d.). El Destino Post Mortem del Gnero y la Emergencia del Estado Tarasco. Paper presented at the Annual Meeting of the Society for American Archaeology, April 29, 2006, San Juan, Puerto Rico. Pia Chan, R., and O, K. (1982). Exploraciones Arqueolgicas en Tingambato, Michoacn. Instituto Nacional de Antropologa e Historia, Mexico City. Pires-Ferreira, J. W., and Flannery, K. V. (1976). Ethnographic Models for Formative Exchange. In Flannery, K. V. (ed.), The Early Mesoamerican Village, Academic Press, New York, pp. 286-292. Pires-Ferreira, J. W. (1976). Obsidian Exchange in Formative Mesoamerica. In Flannery, K. V. (ed.), The Early Mesoamerican Village, Academic Press, New York, pp. 292-306. Pollard, H. P. (1980). Central Places and Cities: a Consideration of the Protohistoric Tarascan State. American Antiquity 45(4):677-696. Pollard, H. P. (1982a). Ecological Variation and Economic Exchange in the Tarascan State. American Ethnologist 9(2):250-268. Pollard, H. P. (1982b). Water and politics: paleoecology and the centralization of the Tarascan State. Paper presented at the 44 th International Congress of the Americanists, Manchester. Pollard, H. P. (1987). The Political Economy of Prehispanic Tarascan Metallurgy. American Antiquity 52(4):741-752. Pollard, H. P. (1991). The Construction of Ideology in the Emergence of the Prehispanic Tarascan State. Ancient Mesoamerica 2:167-179. Pollard, H. P. (1993). Tariacuris Legacy: the Prehispanic Tarascan State, University of Oklahoma Press, Norman. Pollard, H. P. (1994). Ethnicity and political control in a complex society: the Tarascan state of prehispanic Mexico. In Brumfiel, E. M., and Fox, J. W. (eds.), Factional Competition and Political Development in the New World, Cambridge University Press, Cambridge, pp. 79-88. 357

PAGE 358

Pollard, H. P. (1997). Recent Research in West Mexican Archaeology. Journal of Archaeological Research 5:345-384. Pollard, H. P. (2003a). El gobierno del estado tarasco prehispnico. In Paredes Martnez, C., and Tern, M. (coords.), Autoridad y Gobierno Indgena en Michoacn, El Colegio de Michoacn, Centro de Investigaciones y Estudios Superiores en Antropologa Social, Instituto Nacional de Antropologa e Historia, Universidad Michoacana, Zamora, pp. 49-60. Pollard, H. P. (2003b). The Tarascan Empire. In Smith, M. E., and Berdan, F. F. (eds.), The Postclassic Mesoamerican World, University of Utah Press, Salt Lake City, pp. 78-86. Pollard, H. P. (2003c). Development of a Tarascan Core: the Lake Ptzcuaro Basin. In Smith, M. E., and Berdan, F. F. (eds.), The Postclassic Mesoamerican World, University of Utah Press, Salt Lake City, pp. 227-237. Pollard, H. P. (2003d). Central Places and Cities in the Core of the Tarascan State. In Sanders, W. T., Guadalupe Mastache, A., and Cobean, R. H. (eds.), Urbanism in Mesoamerica, volume 1, Instituto Nacional de Antropologa e Historia, The Pennsylvania State University, University Park, pp. 345-390. Pollard, H. P. (2008). A Model of the Emergence of the Tarascan State. In Press, to be published in Fall, 2008, in the journal Ancient Mesoamerica. Pollard, H.P. (n.d.a). Proyecto Desarollo del Estado Tarasco: Los Seoros Urichu, Xarquaro, y Pareo (1990-1998). Informe Final al Consejo de Arqueologa, INAH. 3 volumes. Pollard, H. P. (n.d.b). Proyecto Erongarcuaro, Temporada I (campo) 2001, Temporada II (Laboratorio) 2002-2004. Informe Tcnico Parcial al Consejo de Arqueologa, INAH. Pollard, H. P., and Cahue, L. (1999). Mortuary Patterns of Regional Elites in the Lake Ptzcuaro Basin of Western Mexico. Latin American Antiquity 10(3):259-280. Pollard, H. P., and Gorenstein, S. (1980). Agrarian Potential, Population, and the Tarascan State. Science 209(4453):274-277. Pollard, H. P., and Haskell, D. L. (n.d.). Desarollo del Estado Tarasco: Proyecto Erongarcuaro. In El Boletn del Consejo de Arqueologa, Instituto Nacional de Antropologa e Historia, Mexico, http://www.inah.gob.mx/index/html. Vol. 1. Pollard, H. P., Hirshman, A., Neff, H., and Glascock, M. D. (2001). Las elites, el intercambio de bienes y el surgimiento del rea nuclear tarasca: anlisis de la cermica de la cuenca de Ptzcuaro. InWilliams, E., and Novella, R. (eds.), Estudios cermicos en el occidente y norte de Mexico, El Colegio de Michoacn, Zamora, Mexico, pp. 159-182. 358

PAGE 359

Pollard, H. P., Hirshman, A., Neff, H., and Glascock, M. D. (2005). Exchange, Elites, and the Emergence of the Tarascan Core. In Dillon, B. D., and Boxt, M. A. (eds.), Archaeology: Papers in Honor of Clement W. Meighan, Labyrinthos, Lancaster, CA, pp. 295-307. Pollard, H. P., and Vogel, T. A. (1994). Late Postclassic imperial expansion and economic exchange within the Tarascan domain. In Hodge, M. G., and Smith, M. E. (eds.), Economies and Polities in the Aztec Realm, Institute for Mesoamerican Studies, SUNY, Albany, pp. 447-470. Preucel, R. W. (2006). Archaeological Semiotics, Blackwell, Malden, MA. Preucel, R. W., and Bauer, A. A. (2001). Archaeological Pragmatics. Norwegian Archaeological Review 34(2):85-96. Rakita, G. F. M. and Buikstra, J. E. (2005). Introduction. In Rakita, G. F. M., Buikstra, J. M., Beck, L. A., and Williams, S. R. (eds.), Interacting With the Dead: Perspectives on Mortuary Archaeology for the New Millennium, University Press of Florida, Gainesville, pp. 1-11. Rebnegger, K. J. (n.d.). Obsidian Production and Changing Consumption Patterns in the Lake Ptzcuaro Basin, Michoacn, Mexico. Paper presented at the Annual Meeting of the Society of American Archaeology, April 29, 2006, San Juan, Puerto Rico. Relacin de Michoacn (RM) (1956). Relacin de las ceremonas y ritos y poblacin y gobierno de los indios de la provincia de Michoacn, Facsimile reproduction of Ms IV de El Escorial, Madrid, Tudela, J., transcription, prologue, introduction, and notes, Aguilar Publicistas, Madrid. Relacin Geogrfica de Tiripeto (2002[1580]). Cerda, I., transcription, notes and study. Universidad Michoacana, Morelia, Mexico. Renfrew, C. (1975). Trade as Action at a Distance: Questions of Integration and Communication. In Lamberg-Karlovsky, C. C., and Sabloff, J. A. (eds.), Ancient Civilization and Trade, University of New Mexico Press, Albuquerque, pp. 3-60. Richards, J. (2000). Modified order, responsive legitimacy, redistributed wealth: Egypt, 2260-1650 BC. In Richards, J., and Van Buren, M. (eds.), Order, legitimacy, and wealth in ancient states, Cambridge University Press, Cambridge, pp. 36-45. Richards, J. and Van Buren, M. (eds.) (2000). Order, legitimacy, and wealth in ancient states, Cambridge University Press, Cambridge. Richardson, M. (1982). Being-in-the-Market versus Being-in-the-Plaza: Material Culture and the Construction of Social Reality in Spanish America. American Ethnologist 9(2):421-436. 359

PAGE 360

Robb, J. E. (1999). Secret Agents: Culture, Economy, and Social Reproduction. In Robb, J. E. (ed.), Material Symbols: Culture and Economy in Prehistory, Southern Illinois University Press, Carbondale, pp. 3-15. Robb, J. E. (2004). The Extended Artefact and the Monumental Economy: a Methodology for Material Agency. In DeMarrais, E., Gosden, C., and Renfrew, C. (eds.), Rethinking Materiality: the engagement of mind with the material world, McDonald Institute for Archaeological Research, Cambridge. Roskamp, H. (1998). La historiografa indgena de Michoacn: El Lienzo de Jucutacato y los Titulos de Carapan, CNWS Publications, 72, Leiden, The Netherlands. Roskamp, H. (2001). Warriors of the sun: the eagle lords of Curicaueri and a 16 th century coat of arms from Tzintzuntzan, Michoacn. Mexicon 23(1): 14-17. Roskamp, H. (2003). Los Ttulos Primordiales de Carapan: Legitimacin e Historiografa en una Comunidad Indgena de Michoacn. In Paredes Martnez, C., and Tern, M. (coords.), Autoridad y Gobierno Indgena en Michoacn, El Colegio de Michoacn, Centro de Investigaciones y Estudios Superiores en Antropologa Social, Instituto Nacional de Antropologa e Historia, Universidad Michoacana, Zamora, pp. 305-359. Roskamp, H., and Csar Villa, G. (2003). Iconografa de un pleito: El Lienzo de Aranza y la conflictividad poltica en la sierra tarasca, siglo XVII. In Paredes Martnez, C., and Tern, M. (coords.), Autoridad y gobierno indgena en Michoacn, El Colegio de Michoacn, Centro de Investigaciones y Estudios Superiors en Antropologa Social, Instituto Nacional de Antropologa e Historia, and Universidad Michoacana, Zamora, Mexico, pp. 217-239. Rubn de la Borbolla, D. F. (1939). Antropologa Tzintzuntzan-Ihuatzio: Temporadas I y II. Revista Mexicana de Estudios Antropolgicos 3(2):99-121. Rubn de la Borbolla, D. F. (1941). Exploraciones arqueolgicas en Michoacn: Tzintzuntzan Temporada III. Revista Mexicana de Estudios Antropolgicos 5(1):5-20. Rubn de la Borbolla, D. F. (1944). Orfebrera Tarasca. Cuadernos Americanos 3:125-138. Sahagn, Fr. B. de (1950-1975). General History of the Things of New Spain; Florentine Codex, Anderson, A. J. O., and Dibble, C. E., (trans.), School of American Research, Santa Fe. Sahlins, M. (1972). Stone Age Economics, Aldine-Atherton, Chicago. Sahlins, M. (1981). Historical Metaphors and Mythical Realities: Structure in the Early History of the Sandwich Islands Kingdom, University of Michigan Press, Ann Arbor. Sahlins, M. (1985a). Islands of History, University of Chicago Press, Chicago. 360

PAGE 361

Sahlins, M. (1985b). Hierarchy and humanity in Polynesia. In Hooper, A., and Huntsman, J. (eds.), Transformations of Polynesian Culture, The Polynesian Society, Auckland, pp. 195-217. Sahlins, M. (1990). The Political Economy of Grandeur in Hawaii. In Ohnuki-Tierney, E. (ed.), Culture Through Time: Anthropological Approaches, Stanford University Press, Stanford, pp. 26-56. Sahlins, M. (1991). The return of the event, again; with reflections on the beginning of the Great Fijian War of 1843 to 1855 between the kingdoms of Bau and Rewa. In Biersack, A. (ed.), Clio in Oceania: Toward a Historical Anthropology, Smithsonian Institution Press, Washington D.C., pp. 37-99. Sanders, W. T., Parsons, J. R., and Santley, R. S. (1979). The Basin of Mexico: Ecological Processes in the Evolution of a Civilization, Academic Press, New York. Sanders, W. T., and Price, B. J. (1968). Mesoamerica: the Evolution of a Civilization. Random House, New York. Saxe, A. A. (1970). Social Dimensions of Mortuary Practices, Unpublished Ph.D. thesis, University of Michigan. Scholes, F. V., and Adams, E. B. (1952). Proceso contra Tzintzicha Tangaxoan, el caltzontzin. Porra y Obregon, Mexico City. Sebeok, T. A. (2001). Signs: an introduction to semiotics, University of Toronto Press, Toronto. Seeger, A. (1975). The Meaning of Body Ornaments: a Suya Example. Ethnology 14(3):211-224. Seler, E. (1993). The ancient inhabitants of the Michuacan region. In Collected Works in Mesoamerican Linguistics and Archaeology, 2 nd edition, vol. 4, pp. 3-66. Labyrinthos, Culver City. Shanks, M., and Tilley, C. (1982). Ideology, symbolic power and ritual communication: a reinterpretation of Neolithic mortuary practices. In Hodder, I. (ed.), Symbolic and Structural Archaeology, Cambridge University Press, Cambridge, pp. 129-154. Simpson, L. B. (1966[1950]). The Encomienda System in New Spain: the Beginning of Spanish Mexico, University of California Press, Berkeley. Smith, M. E. (2001). The Aztec Empire and the Mesoamerican World System. In Alcock, S. E., DAltroy, T. N., Morrison, K. D., and Sinopoli, C. M. (eds.), Empires: Perspectives from Archaeology and History, Cambridge University Press, Cambridge, pp. 128-154. 361

PAGE 362

Smith, M. E. (2003). Economic Change in Morelos Households. In Smith, M. E., and Berdan, F. F. (eds.) The Postclassic Mesoamerican World, University of Utah Press, Salt Lake City, pp. 249-258. Smith, M. E. (2008). Tlahuica Ceramics: the Aztec-Period Ceramics of Morelos, Mexico. Institute of Mesoamerican Studies Monograph 15, Albany, New York. Smith, M. E., Wharton, J. B., and Olson, J. M. (2003). Aztec Feasts, Rituals, and Markets: Political Uses of Ceramics Vessels in a Commercial Economy. In Bray, T. L. (ed.), The Archaeology and Politics of Food and Feasting in Ancient States and Empires, Kluwer Academic, New York, pp. 235-269. Smith, M. E., and Berdan, F. F. (1992). Archaeology and the Aztec Empire. World Archaeology 23(3):353-367. Stanislawsky, D. (1947). Tarascan political geography. American Anthropologist 49(1): 46-55. Stark, B. L. (1992). Ceramic Production in Prehistoric La Mixtequilla, South-Central Veracruz, Mexico. In Bey, G. J. III, and Pool, C. A. (eds.), Ceramic Production and Distribution: an Integrated Approach, Westview Press, Boulder, CO, pp. 175-204. Stawski, C. J. (2008). Residential Zoning at Prehispanic Tzintunztzan, Mexico Revisited: a Quantitative Analysis. Unpublished MA Thesis, Michigan State University. Stone, C. (2004). In Place of Gods and Kings: Authorship and Identity in the Relacin de Michoacn, University of Oklahoma Press, Norman. Strathern, M. (1979). The Self in Self-Decoration. Oceania 49(4):241-257. Strathern, M. (1988). The Gender of the Gift: Problems with Women and Problems with Society in Melanesia, University of California Press, Berkeley. Strathern, M. (1991). One man and many men. In Godelier, M., and Strathern, M. (eds.), Big Men and Great Men: Personifications of power in Melanesia, Cambridge University Press, Cambridge, pp. 197-214. Tainter, J. A. (1975). Social inference and mortuary practices: an experiment in numerical classification. World Archaeology 7:1-15. Tainter, J. A. (1978). Mortuary practices and the study of prehistoric social systems. Advances in Archaeological Method and Theory 1:105-141. Tcherkzoff, S. (1987). Dual Classification Reconsidered: Nyamwezi Sacred Kingship and Other Examples, Cambridge University Press, Cambridge. 362

PAGE 363

Thomas, J. (1999). An Economy of substances in earlier Neolithic Britain. In Robb, J. E. (ed.), Material Symbols: Culture and Economy in Prehistory, University of Southern Illinois Press, Carbondale, pp. 70-89. Thomas, N. (1991). Entangled Objects: Exchange, Material Culture, and Colonialism in the Pacific, Harvard University Press, Cambridge. Tilley, C. (1991). Material Culture and Text: the art of ambiguity. Routledge, New York. Tilley, C. (1999). Metaphor and Material Culture. Blackwell, Malden, MA. Toren, C. (1999[1994]). All Things Go in Pairs, Or the Sharks Will Bite: the antithetical nature of Fijian chiefship. In Toren, C., Mind, Materiality, and History: Explorations in Fijian Ethnography, Routledge, New York, pp. 163-181. Turner, T. S. (1977). Narrative Structure and Mythopoiesis: A Critique and Reformulation of the Structuralist Concepts of Myth Narrative and Poetics. Arethusa 10(1):103-164. Turner, T. S. (1980). The Social Skin. In Cherfas, J., and Lewin, R. (eds.), Not Work Alone: a Cross-cultural View of Activities Superfluous to Survival, Temple Smith, London. Turner, T. S. (1984). Dual Opposition, Hierarchy, and Value: Moiety Structure and Symbolic Polarity in Central Brazil and Elsewhere. In Galey, J.-C. (ed.), Diffrences, Valeurs, Hirarchie: Textes Offerts Louis Dumont, Editions de lEcole des Hautes Etudes en Sciences Sociales, Paris, pp. 335-370. Turner, T. S. (1985). Animal Symbolism, Totemism, and the Structure of Myth. In Urton, G. (ed.), Animal Myths and Metaphors in South America, University of Utah Press, Salt Lake City, pp. 49-106. Urton, G. (1990). The History of a Myth: Pacariqtambo and the origin of the Inkas, University of Texas Press, Austin. Versluis, V. (1994). The Iconography of the Protohistoric Tarascan State of Western Mexico: The Material Expression of the State Ideology. M.A. Thesis, Department of Anthropology, Michigan State University, East Lansing, Michigan. Wagner, R. (1975). The Invention of Culture, Prentice-Hall, Englewood Cliffs, NJ. Wagner, R. (1977). Analogic kinship: a Daribi example. American Ethnologist 4(4):623642. Wagner, R. (1991). The fractal person. In Godelier, M., and Strathern, M. (eds.), Big Men and Great Men: Personifications of power in Melanesia, Cambridge University Press, Cambridge, pp. 159-173. 363

PAGE 364

Wallerstein, I. (1974). The Modern World-System: Capitalist agriculture and the origins of the European world-economy in the sixteenth century. Academic Press, New York. Warren, J. B. (1963). The Caravajal Visitation: First Spanish Survey of Michoacn. The Americas 19(4):404-412. Warren, J. B. (1970-71). Fray Jernimo de Alcal: Author of the Relacin de Michoacn? The Americas 27:306-327. Warren, J. B. (1977). La Conquista de Michoacn, Fimax Publicistas, Morelia, Mexico. Warren, J. B. (1984). Los estudios lingisticos en Michoacn en el siglo XVI: una expression del humanismo cristiano. In Paredes Martnez, C. (ed.), Humanismo y ciencia en la formacin de Mxico, El Colegio de Michoacn and Conacyt, Zamora and Guadalajara, Mexico, pp. 113-124. Warren, J. B. (1985). The Conquest of Michoacan, University of Oklahoma Press, Norman. Webmoor, T. (2007). What about one more turn after the social in archaeological reasoning: Taking things seriously. World Archaeology 39(4):563-573. Weiner, A. B. (1992). Inalienable Possessions: the Paradox of Keeping-While-Giving, University of California Press, Berkeley. West, R. C. (1947). Cultural Geography of the Modern Tarascan Area, Institute of Social Anthropology, Smithsonian Institution, Washington D.C. Winter, M. C., and Pires-Ferreira, J. W. (1976). Distribution of Obsidian among Households in Two Oaxacan Villages. In Flannery, K. V. (ed.), The Early Mesoamerican Village, Academic Press, New York, pp. 306-311. Yoffee, N. (2005). Myths of the Archaic State, Cambridge University Press, Cambridge. 364

PAGE 365

BIOGRAPHICAL SKETCH David Haskell was born in Columbus, Ohio in 1979. He graduated from Reynoldsburg High School in 1997. David earned a Bachelor of Arts from The Ohio State University in 2000, majoring in anthropology and graduating Summa Cum Laude. After studying anthropology in graduate level courses at the University of Illinois at Urbana-Champaign for one academic year, he decided to follow his advisor, Dr. Susan Gillespie, to the University of Florida at Gainesville. There David earned his MA in anthropology in 2003. His Masters thesis was entitled History and the Construction of Hierarchy and Ethnicity in the Prehispanic Tarascan State: a Syntagmatic Analysis of the Relacin de Michoacn. David has made the archaeological and ethnohistoric investigation of the prehispanic Tarascan kingdom the focus of his academic pursuits since 2001, participating in and co-directing research at the site of Erongarcuaro, Michoacn, Mexico over the years. He now lives in Columbus, Ohio with his wife Emmy and son Siler.