Exploring Maya Ritual Fauna

Permanent Link: http://ufdc.ufl.edu/UFE0024821/00001

Material Information

Title: Exploring Maya Ritual Fauna caves and the proposed link with contemporary hunting ceremonialism
Physical Description: 1 online resource (123 p.)
Language: english
Creator: Anderson, Elyse
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009


Subjects / Keywords: archaeology, cave, maya, practice, ritual, zooarchaeology
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: EXPLORING MAYA RITUAL FAUNA: CAVES AND THE PROPOSED LINK WITH CONTEMPORARY HUNTING CEREMONIALISM By Elyse Anderson August 2009 Chair: Kitty F. Emery Major: Anthropology The ritual use of caves and associated landscape features has garnered increased archaeological attention. The thorough examination of cave faunal assemblages, however, is still relatively rare. These deposits are often deemed 'problematical' due to the difficulty in interpreting the presence of animal skeletal remains in cave contexts. New ethnozoological research by Linda Brown and Kitty Emery (2008) suggests that the key to understanding these cave deposits can be found in contemporary hunting ritual practices. Linda Brown (2005, 2006) has documented the practice of hunting ceremonialism among modern Maya hunters in the Guatemalan highlands. These practices involve the ritual caching of animal skeletal remains at sacred landscape features such as caves, rock shelters, and rock outcrops. Using the material signature of hunting ceremonialism developed by Brown and Emery (2008), this study tests the connection between archaeological caves deposits and hunting ceremonialism. This analysis examines in detail the zooarchaeological, spatial, and material characteristics of faunal assemblages from nine cave sites: Cueva de los Quetzales, Cueva de Rio Murcielagos, Aguateca Grieta, Naj Tunich, Eduardo Quiroz, Actun Balam, Actun Polbilche, Stela Cave, and Caves Branch Rockshelter. As part of a larger discussion, this study also analyzes the validity of previous studies that have attempted to identify pan-Maya ritual practices involving the deposition of faunal remains in cave contexts. This study challenges the idea of pan-Maya ritual practices and calls for a more in-depth, site-specific analysis of Maya ritual practices.
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 Elyse Anderson.
Thesis: Thesis (M.A.)--University of Florida, 2009.
Local: Adviser: Emery, Kitty F.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024821:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024821/00001

Material Information

Title: Exploring Maya Ritual Fauna caves and the proposed link with contemporary hunting ceremonialism
Physical Description: 1 online resource (123 p.)
Language: english
Creator: Anderson, Elyse
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009


Subjects / Keywords: archaeology, cave, maya, practice, ritual, zooarchaeology
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: EXPLORING MAYA RITUAL FAUNA: CAVES AND THE PROPOSED LINK WITH CONTEMPORARY HUNTING CEREMONIALISM By Elyse Anderson August 2009 Chair: Kitty F. Emery Major: Anthropology The ritual use of caves and associated landscape features has garnered increased archaeological attention. The thorough examination of cave faunal assemblages, however, is still relatively rare. These deposits are often deemed 'problematical' due to the difficulty in interpreting the presence of animal skeletal remains in cave contexts. New ethnozoological research by Linda Brown and Kitty Emery (2008) suggests that the key to understanding these cave deposits can be found in contemporary hunting ritual practices. Linda Brown (2005, 2006) has documented the practice of hunting ceremonialism among modern Maya hunters in the Guatemalan highlands. These practices involve the ritual caching of animal skeletal remains at sacred landscape features such as caves, rock shelters, and rock outcrops. Using the material signature of hunting ceremonialism developed by Brown and Emery (2008), this study tests the connection between archaeological caves deposits and hunting ceremonialism. This analysis examines in detail the zooarchaeological, spatial, and material characteristics of faunal assemblages from nine cave sites: Cueva de los Quetzales, Cueva de Rio Murcielagos, Aguateca Grieta, Naj Tunich, Eduardo Quiroz, Actun Balam, Actun Polbilche, Stela Cave, and Caves Branch Rockshelter. As part of a larger discussion, this study also analyzes the validity of previous studies that have attempted to identify pan-Maya ritual practices involving the deposition of faunal remains in cave contexts. This study challenges the idea of pan-Maya ritual practices and calls for a more in-depth, site-specific analysis of Maya ritual practices.
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 Elyse Anderson.
Thesis: Thesis (M.A.)--University of Florida, 2009.
Local: Adviser: Emery, Kitty F.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024821:00001

This item has the following downloads:

Full Text




2009 Elyse Anderson 2


To Nikko and Lily 3


ACKNOWLEDGMENTS This thesis would not have been possible with out the support of my colleagues, committee, friends, and family. Thanks to Dr. Linda Br own and Dr. Kitty Emery for allowing me to participate in this project and for their guida nce throughout the writing process. I am also grateful to Dr. Susan deFrance for her helpful advice and Dr. Donald Grayson for encouraging me to pursue zooarchaeology. A special thanks goes to th e zooarchaeology girls and boy. I'm grateful to Michelle LeFebvre for her support and encouragement. Meggan Blessing and Nicole Cannarozzi were great lab partners and always ke pt me entertained through endle ss hours of faunal analysis. Erin Kennedy-Thornton was a constant source of he lp for navigating the ins and outs of the zooarchaeology lab and graduate school in gene ral. I would especially like to thank Erol Kavountzis for being an amazing friend a nd for going on this journey with me. Finally I would like to acknowledge all of the people outside of academia that helped to keep me sane. Thanks to Dr. Meggen Sixbey for helping me get through the hard times and gain better perspective on my life. My parents, Scott and Cherese, and my sisters, Melinda, Cheren, and Marissa, were a constant source of encourag ement. Finally, thanks to Mary Warrick for being a loving and supportive companion. This project was funded by the Latin American Studies Tinker Grant from the University of Florida. 4


TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4LIST OF TABLES ...........................................................................................................................8LIST OF FIGURES .........................................................................................................................9ABSTRACT ...................................................................................................................................10 CHAPTER 1 INTRODUCTION ................................................................................................................ ..122 HUNTING CACHE RESEARCH ..........................................................................................15Contemporary Hunting Ceremonialism ..................................................................................15Contemporary and Ancient views of the Animal Guardian ...................................................16Jaloj-kexoj: bone and the rege neration of life from death .....................................................18Material Signature of Hunting Caches ...................................................................................19Spatial and Material Data .......................................................................................................22Summary .................................................................................................................................233 PRIOR RESEARCH IN MAYA CAVE ARCHAEOLOGY AND ZOOARCHAEOLOGY ..........................................................................................................24Maya Cave Archaeology ........................................................................................................24Maya Zooarchaeology and Ritual Deposits ............................................................................264 CAVE SITE BACKGROUND ...............................................................................................31Cueva de los Quetzales ...........................................................................................................31Cueva de Rio Murcielagos ......................................................................................................32Aguateca Grieta ......................................................................................................................33Naj Tunich ..............................................................................................................................34Caves Branch Rockshelter ......................................................................................................35Stela Cave ...............................................................................................................................35Eduardo Quiroz Cave .............................................................................................................36Actun Balam ...........................................................................................................................36Actun Polbilche ......................................................................................................................375 THEORETICAL APPROACH ..............................................................................................38Practice Theory .......................................................................................................................38A Practice Approach to Ritual ................................................................................................39Walker's Life History Approach .............................................................................................41 5


6 METHODS ..................................................................................................................... ........45Zooarchaeological Analysis ....................................................................................................45Quantification .........................................................................................................................46Other Sources of Bias .............................................................................................................48Testing Ritual "Signatures": The Conceptual Framework .....................................................49Formal Properties/Numerical Frequencies ......................................................................49Density .....................................................................................................................50Taxa ..........................................................................................................................50Diversity ...................................................................................................................51Age classes ...............................................................................................................51Elements, body portions, and skeletal completeness ...............................................52Sidedness ..................................................................................................................53Human modifications ...............................................................................................53Taphonomy ...............................................................................................................54Associations/Spatial Locations ........................................................................................54Location relative to settlement .................................................................................54Presence of ritual activity areas ................................................................................557 RESULTS ..................................................................................................................... ..........56Density ....................................................................................................................................56Taxonomy ...............................................................................................................................56Diversity .................................................................................................................................59Age Classes .............................................................................................................................59Elements, Body Portions, and Skeletal Completeness ...........................................................60Sidedness ................................................................................................................................61Human Modifications .............................................................................................................63Taphonomy .............................................................................................................................63Location relative to settlement ................................................................................................64Presence of ritual activity areas ..............................................................................................658 INTERPRETATION .............................................................................................................. 96Density ....................................................................................................................................96Taxonomy ...............................................................................................................................96Birds, reptiles, fish, and mollusks ...................................................................................96Mammals .........................................................................................................................97Diversity .................................................................................................................................99Age Distributions ..................................................................................................................100Elements, Body Portions, and Skeletal Completeness .........................................................101Sidedness ..............................................................................................................................103Human Modifications ...........................................................................................................104Taphonomy ...........................................................................................................................105Location relative to settlement ..............................................................................................106Presence of ritual activity areas ............................................................................................107Summary ...............................................................................................................................108 6


9 CONCLUSION .................................................................................................................. ...110LIST OF REFERENCES .............................................................................................................115BIOGRAPHICAL SKETCH .......................................................................................................123 7


LIST OF TABLES Table page 7-1 Site Density Values For Aguateca Grieta ..........................................................................70 7-3 NISP/MNI for Actun Polbilche, Eduardo Quiroz, Actun Balam, Stela Cave, and Caves Branch .....................................................................................................................73 7-4 Community Statistics for Hunting Shrines ........................................................................76 7-5 Community Statistics for Cave Sites .................................................................................76 7-6 Cueva de los Quetzales Species/Age Class Data (MNI) ...................................................78 7-7 Cueva de Rio Murcielagos Species/Age Class Data (MNI) ..............................................79 7-8 Aguateca Grieta Species/Age Class Data (MNI) ...............................................................80 7-9 Body Portion Frequencies (%NISP) ..................................................................................81 7-10 NISP values for left-sided elements by species at Cueva de los Quetzales .......................86 7-11 NISP values for left-sided elem ents by species at Rio Murcielagos .................................87 7-12 NISP values for left-sided elem ents by species at Aguateca Grieta ..................................88 7-13 NISP values for left-sided elements by species at Naj Tunich ..........................................89 7-14 NISP values for left-sided elem ents by species at Eduardo Quiroz ...................................90 7-15 NISP values for left-sided elements by species at Actun Balam .......................................91 7-16 NISP values for left-sided elements by species at Caves Branch Rockshelter ..................92 7-17 NISP values for left-sided elements by species at Stela Cave ...........................................93 7-18 Frequency of Taphonomic Markers ...................................................................................94 7-19 Fragmentation Analysis .....................................................................................................95 8


LIST OF FIGURES Figure page 7-1 Displays the relative proportion of mature vs immature individuals at Cueva de los Quetzales, Aguateca Grieta, Ri o Murcielagos, and Naj Tunich ........................................77 7-2 Age Class data based on element fu sion rates for the Petexbatn cave sites ....................77 7-3 The ratio of expected to observed body por tions for white-tailed deer at Cueva de los Quetzales, Aguateca Grieta, a nd Cueva de Rio Murcielagos. ...........................................82 7-4 The ratio of expected to observed body por tions for white-tailed deer at Naj Tunich, Caves Branch Rockshelter, Stela Cave Actun Balam, and Eduardo Quiroz. ...................82 7-5 The ratio of expected to observed bo dy portions for peccary at Cueva de los Quetzales, Aguateca Grieta, Cueva de Rio Murcielagos, and Naj Tunich ........................83 7-6 The ratio of expected to observed bo dy portions for peccary at Cueva de los Quetzales and Aguateca Grieta. .........................................................................................83 7-7 The ratio of expected to observed bo dy portions for Cueva de los Quetzales and Aguateca Grieta. ................................................................................................................84 7-8 The ratio of expected to observed bo dy portions for opossum at Naj Tunich, Caves Branch Rockshelter, Stela Cave, and Eduardo Quiroz. .....................................................84 7-9 The ratio of expected to observed body portions for paca at Cueva de los Quetzales and Aguateca Grieta. ..........................................................................................................85 7-10 The total number (NISP) of left and right sided elements for Cueva de los Quetzales, Cueva de Rio Murcielagos, Aguateca Grieta Stela Cave, Caves Branch Rockshelter, and Eduardo Quiroz. ..........................................................................................................85 9


Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts EXPLORING MAYA RITUAL FAUNA: CA VES AND THE PROPOSED LINK WITH CONTEMPORARY HUNTING CEREMONIALISM By Elyse Anderson August 2009 Chair: Kitty F. Emery Major: Anthropology The ritual use of caves and associated landscape features has garnered increased archaeological attention. The thorough examinati on of cave faunal assemblages, however, is still relatively rare. These deposits are often deemed "problematical" due to the difficulty in interpreting the presence of animal skeletal remains in cave contexts. New ethnozoological research by Linda Brown and Kitty Emery (2008) suggests that the key to understanding these cave deposits can be found in contemporary hun ting ritual practices. Linda Brown (2005, 2006) has documented the practice of hunting ceremon ialism among modern Maya hunters in the Guatemalan highlands. These practices involve the ritual caching of animal skeletal remains at sacred landscape features such as caves, rock shelters, and rock outcrops. Using the material signature of hunting ceremonialism developed by Brown and Emery (2008), this study tests the c onnection between archaeological caves deposits and hunting ceremonialism. This analysis examines in detail the zooarchaeological, spatial, and material characteristics of faunal assemblages from nine cav e sites: Cueva de los Quetzales, Cueva de Rio Murcielagos, Aguateca Grieta, Naj Tunich, Edua rdo Quiroz, Actun Balam, Actun Polbilche, Stela Cave, and Caves Branch Rockshelter. As part of a larger disc ussion, this study also analyzes the validity of previous studies that have attempted to identify pan-Maya ritual practices 10


11 involving the deposition of faunal remains in cave contexts. This study challenges the idea of pan-Maya ritual practices and calls for a more in -depth, site-specific analysis of Maya ritual practices.


CHAPTER 1 INTRODUCTION Recent archaeological investiga tion of caves and other sacred landscape features in the Maya region have stimulated interest in anci ent Maya ritual practices. Archaeologists have drawn upon both historical and contemporary Maya accounts of rituals carried out in cave contexts in order to understand ancient ritual practices. New ethnoarch aeological work by Linda Brown (2005, 2006) and Kitty Emery (Emery et al 2007, Brown and Emery 2008) has proposed a connection between the material remains of contemporary Maya hunting rituals and archaeological deposits found in caves throughout the Maya Lowlands. Brown has documented the practice of hunting ceremonialism among Tz 'utujil and Kaqchikel Maya hunters around Lake Atitln in the Guatemalan Highlands. This prac tice involves the negotiation with a supernatural animal guardian who must be placated before and after hunting expeditions in order to ensure both hunting success and safety for the hunter. Nego tiations take place at hunting shrines which are places in the landscape, usually caves, rocksh elters, or rock outcrops, where the guardian is believed to reside. Hunting ceremonialism results in a unique material si gnature that includes a particular location and context, a distinctive faunal assemblage, and three ritual activity areas. Brown and Emery (2008) and Emery et al (2007 ) argue that hunting ceremonialism has deep roots in Maya history and that the material signature of hunting ceremonialism could be used to possibly identify hunting ceremonialism in the ar chaeological record. Th is study tests that assertion. The material signature of hunting ceremonialis m is largely composed of zooarchaeological characteristics because this practice involves th e careful curation and deposition of skeletal remains from wild animals at the hunting shrine. As a result, hunting shrines often contain a faunal cache, or an accumulation of animal remain s, that are unique in terms of the overall 12


density, species, diversity, age, skeletal completeness, body side, taphonomy, and human modifications of the remains. In combination, these characteristics can be used to identify hunting ceremonialism in the archaeological record. Some of these ch aracteristics both contradict and overlap with previous zooarchaeologi cal investigations (Pohl 1983, Brady 1989, Emery 2002) of lowland cave archaeologica l deposits. None of these studies however, argued that these characteristics reflect hunting ce remonialism. For example, Mary Pohl (1983) argued that cave deposits had a preponderance of juvenile animal specimens, a characteristic that was also found at the hunting shrines investig ated by Brown and Emery (2008). According to Pohl, however, this characteristic of juvenile individuals is reflective of pan-Maya, cave-specific rituals of accession and fertility. Therefore, this study also addresses a much larger question: Are lowland cave faunal assemblages relatively uniform and ther efore reflect pan-Maya cave ritual behavior or are these assemblages highly variable reflectin g different ritual type s, regional differences, etc. To answer these questions, this study examin es faunal assemblages from nine cave sites located in the Maya Lowlands. Three caves, Cuev a de los Quetzales, Cueva de Rio Murcielagos, and Aguateca Grieta, are located in the Petexba tn region of Guatemala. Animal remains from these caves were analyzed by the author and form the core of this analysis. The Belizean caves of Eduardo Quiroz, Actun Polbilche, Actun Balam, Stela Cave, and Caves Branch Rockshelter and the Petn Guatemalan site of Naj Tunich were also chosen as comparative samples because published faunal reports were available for each of these sites. This study demonstrates the great variability found in faunal assemblages located in Maya cave sites. There are a number of similaritie s and differences between contemporary hunting shrines and lowland cave deposits and there is great variation be tween the cave sites themselves. 13


14 This study demonstrates the futility of attempting to define Maya ritual practice as a type of uniform activity throughout time and space and suggests that an in-depth, site specific analysis of cave deposits is a more useful archaeological endeavor.


CHAPTER 2 HUNTING CACHE RESEARCH Written accounts document the widespread belief in a supernatural guardian of animals among the Maya people stretching back to the Classic Period. Although th ese accounts refer to hunting rituals performed in order to appease the Guardian of the Animals, archaeological examples of such rites have not been documen ted. Recent ethnoarchaeological work by Linda Brown (2005, 2006), resulted in the discovery of many contemporary and nineteenth century hunting shrines used by Tzutujil and Kaqchi kel Maya hunters around La ke Atitln in the Guatemalan Highlands. Brown (2005) argued that these faunal caches might hold the key for understanding archaeological fauna l deposits found in caves and associated landscape features throughout the Maya region. In 2007, Linda Br own and Kitty Emery (2008) led a detailed investigation of three of thes e hunting shrines, which resulted in a material signature of hunting ceremonialism. Within this signature is a set of zooarchaeological characteristics reflective of the faunal deposits f ound within hunting shrines. A nu mber of these characteristics overlap with the markers of Maya ritual assemblages proposed by Mary Pohl (1983), James Brady (1989), and Kitty Emery (2002). Therefore, th ese three studies were integrated into this analysis which seeks to answer two questions: 1) Are lowland Maya cav e sites reflective of hunting ceremonialism and/or 2) are these s ites comparable to other Maya cave ritual assemblages? Contemporary Hunting Ceremonialism Among contemporary Tzutujil and Kaqchikel hunters, conducting preand post-hunt rituals are vital for ensuring pers onal safety and success in the pursu it of wild game. Both rituals mostly take place at sacred locations on the land scape, usually caves or rockshelters, where the Guardian of the Animals is believed to reside (Brown and Emery 2008:31 5). At these locations, 15


hunters bring offerings to the Guardian with the hope that he will grant permission to take the lives of animals that reside in his forest domain. At one particular huntin g shrine, PaRuchiAbaj, local oral history claims that when the site was in use, hunters perfor med pre-hunt ceremonies every 20 days (Brown and Emery 2008:312). People brought gifts of roosters, sheep, or beef and left them at the base of the rock outcrop wh ere upon midnight the animal guardian would take the offerings. If the gifts were acceptable, he would send the hunters a dream detailing how many animals they could take (Brown and Emery 2008:312). Under the direction of the animal guardian, the prey would then present themselves to the hunter in an act of self-sacrifice. After a successful hunt, the bones from the animal carca ss are carefully curated and returned to the hunting shrine as part of the post-hunt ritual. The process of butcheri ng, handling, and returning animal bones to the hunting shrine in an appropriate manner is part of the social obligation that hunters have to the animal guardian (Brown and Emery 2008:314). This process also allows hunters to prove they have followed the instructio ns of the animal guardian and have not killed more animals than given permission to. Furthe r, these bones contain the life force of the deceased animal and need to be carefully curated and returned undamaged to the hunting shrine (Brown and Emery 2008, Brown 2005). The animal guardian then uses the life force to repopulate the forests which ensure a steady, sustainable supply of forest game. Both the belief in a Guardian of the Animals and the concept of bone-soul and the cycl e of regeneration in which life arises from death is believed to have deep roots in the Maya region. Contemporary and Ancient views of the Animal Guardian Ethnographic, ethnohistorical, a nd iconographic/epigraphic ev idence suggests that the belief in an animal guardian has great historical depth throughout Mesoamer ica. Below is a brief overview of this evidence largely, but not excl usively, drawn from Brown (2005). Like any other cultural belief or practice, the detailed beliefs concerning an animal guardian varies across both 16


time and space, yet certain aspects appear to show up time and time again. These are the association of an animal guardian who resides within a cave or simila r earth opening and who must be placated in order to ensure hunting success. Within the Maya region, ethnographic accounts of an animal guardian was found among the Yukatek, Mopan, Huastec, Kiche, Tzelta l, and the Tzutujil Maya (Brown 2005:138). The Yukatek Maya provide maize offerings to the fo rest spirit known as Tzip, the supernatural protector of deer (Redfield and Villa Rojas 1934). Among the Hu astec Maya of Veracruz, prehunt rituals include the construction of a corral with an enclosed deer skull. This corral is then ceremonially opened which signifie s the release of a deer from the "master of the animals" (Alcorn 1984:88, Brown 2005:318). In many contem porary accounts, the animal guardian is believe to reside inside a mountai n where he tends to a large floc k of wild animals penned inside a corral. At Zinacantan, the Tz otzil call this mountain Bankilal MukTa Viz (Senior Large Mountain) and it contains a series of supern atural corrals that hous e every wild animal companion or animal co-spirit ( way ) of the Zinacantecos (Vogt 1976: 18). In Santiago Atitln, the Tzutujil Maya continue to make offe rings to the animal guardian in the cofrada system that blends both indigenous Maya and Catholic beliefs. In the Cofrada San Juan, they have transformed a colonial-period stat ue of Saint John the Baptist into the Guardian of the Animals by painting jaguar spots and whiskers on the lamb he is carrying (Carlsen 1997:98). The belief in an animal guardian is a widespread Mesoamer ican phenomenon that is found throughout Mexico an well into Honduras as well (Brown 2005:138). Historical sources from the colonial period and iconographic/ epigraphic sources from the Classic Period also provide intrig uing clues that refer to both animal guardians and hunting rituals which together hint at the antiquity of these beliefs a nd practices. In the Guatemalan 17


highlands, Archbishop Pedro Cortes y Larraz (1 958:119-120) referred to hunting rituals where burned offerings were made to a deer lord na med Xaqui Coxol. In the Yucatan lowlands, the work of Friar Diego de Landa also describes hunting rituals associated with a monthly festival. He described a hunter's ceremony held on the da y 7 Zip that featured a large dance where participants would hold arrows a nd deer skulls that had been painted with blue bitumen. Along with the dance, hunters performed bloodletting ri tes where they pierced their ears and tongues using a maguey spine (Tozzer 1941:155). Two Mayan epigraphers, Gabrielle Vail (1997) and Pierre Colas (2006) found evidence of this huntin g ritual in the pages of the Madrid codex, which also depicts a hunting deity known as God Y sporting a deer headdress and dancing with an arrow. In Classic Period iconography, the deer guardian T'zip was depicted as an old man who possessed deer attributes like deer ears and antlers (Brown 2005:139). Known depictions of Tzip were found on Classic Period ceramic vessel s and in the painted murals of Bonampak (Taube 1997, 2003). Epigraphic sources about Tzip are found in stone inscri ptions at Copan and Dos Pilas and in the Dresden and Paris codices as well (Taube 2003). Jaloj-kexoj: bone and the regen eration of life from death Hunting rituals are also informed by jaloj-kexoj, a cyclical concept of time and regeneration that has a known historical depth in the Maya area (Carlsen and Prechtel 1991). Traditionally, the Maya believe that life arises from death. The term jal refers to the change manifested in the individual life-cycle as one transitions to life through birth, through youth and old age, and finally back to death. In contrast, Kex refers to generational change and in concept is similar to reincarnation and the process of ma king the new out of the ol d (Carlsen and Prechtel 1991:26). Together these two types of change form a system of transformation and renewal. This conceptual framework is reflected in the prac tice of hunting ceremonia lism where a hunter takes an animal's life (jal ) and then must assure the regeneration of the soul ( kex ) by placing the 18


remains in an appropriate location so that the animal guardian can maintain the propagation of the species (Brown 2005:140). This concept is often reflected in the conflation of human/animal lives with agricultural metaphors. In Santiago Atitln, the cyclical life stages of the maize plant (seed, sprout, mature plant, dried husk, etc) is considered commensurate to human life stages (Carlsen 1997:54-56). Carlsen and Prechtel (1991:30-36) found evidence at the site of Palenque and from the Kiche creation myth, the Popol Vuh that suggests the antiquity of the jaloj-kexoj concept. At Palenque, the royal sarcophagus of Pakal displa ys sophisticated depictions of vegetation sprouting forth from the heads of human figures ; of maize and trees growing from skull-like deities; and ears of maize with transposed human faces (Carlsen and Prechtel 1991:34). They argue these depictions reflect the belief in ancestral regenera tion according to the vegetation model. Atitecos also retain considerable knowledge of the Popul Vuh because jaloj-kexoj is central to the origin myth and forms the founda tion of Atiteco religion (Carlsen and Prechtel 1991:34). The conflation of bone, seed, and rege neration pervades the Popol Vuh and is specifically reflected when the mother of the He ro Twins is impregnated by a skull and when the Hero Twins are reborn in the Underworld from their ground bones. Based on the evidence above, Carlsen and Prechtel (1991:33) ar gue that the concept known as jaloj-kexoj among the residents of Santiago Atitln has helped to guide the Maya culture ever since the Classic Period. Material Signature of Hunting Caches Although Brown (2005:132-136) had noted some material properties of these sites previously, such as their associ ation with rock formations and the presence of exclusively wild species, there were many questions left to be answered. In the summer of 2007, an ethnozoological project led by Brown and Emery investigated th ree communal hunting shrines in order to produce a detailed material signature of hunting ceremonialism. Communal shrines, one 19


of the two types of hunting shrines documented by Brown (2006), are used by multiple hunters and/or a whole community. Brown (2006) also documented shrine s used by individual hunters, however, these shrines were not incl uded in the ethnozoological project. Kitty Emery and three zooarchaeologist graduate students (author included) analyzed a total of 6,671 faunal remains from the communal hun ting shrines of PaRuchiAbaj, Pa Sak Man, and PaZiguan. Each skeletal specimen was id entified to taxon, element, element portion, completeness, side, age, cultural modifications (cut marks and burning), taphonomy (animal gnawing, root etching) and patholog ies. Due to the highly sacred na ture of these sites, we were not allowed to remove any bones from these locati ons for analysis. Therefore, all identifications had to be done on site using a detailed digital photographic archive of Maya animal skeletal elements (Emery et. al. 2007). Luckily, most of these remains were whole elements making this analysis relatively easy. This analysis resulted in a detailed description of the material remains of modern Maya hunting ceremonialism (Bro wn and Emery 2008, Emery et. al. 2007). Using the description of the animal remain s from the modern hunting caches, Emery developed a material signature that can be used to identify hunting caches and therefore hunting ceremonialism in the archaeological record (Emery et. al 2007). The char acteristics of this signature are outlined in detail below. 1. High density bone assemblages in defined spatial clusters : The communal hunting shrines studied in 2007 each c ontained a large cache of hunted animal remain. The largest cache, Pa' Ruchi' Abaj, had an estimated total of 600,000 bones. Hunting shrines used by individual hunters, however, often contained a few animal remains and therefore do not contain high density bone assemblages. This material signature is reflective of the communal shrine pattern. 2. Very high OR very low (single or severa l species) taxonomic diversity emphasizing animals valued as food and then those of economic/symbolic value : A total of 23 discrete taxa were identified over these three si tes, with the most diverse assemblage at Pa Sak Man containing 19. The most common sp ecies were white-tailed deer, peccary, armadillo, paca, coati, rabbit, and small agouti. There was also a fair number of opossum, tapir, brocket deer, raccoon, and turkey re mains. Turkey represents the only non20


mammalian taxa recovered. Large and medium size cats, canids, monkeys, tamandua, and pocket gophers were represente d by only one species. Brown (2005) has also documented hunting shrines that contain only one species of faunal remains. 3. Possibly high juvenile frequencies: The ratio of immature to juvenile individuals among all taxa was higher than expected. This pa ttern was especially prominent among commonly hunted species like the white-tailed deer (Brown and Emery 2008:321). 4. Low element fragmentation rate s and possibly low butcher rates: Overall, element fragmentation was low. Over 60% of the asse mblage was composed of complete elements and 25% from one-half complete elements. Bu tchering rates were also found on relatively few remains although it is higher than most archaeological assemblages. Emery suggests that this is consistent with the modern use of metal butchering tools ra ther than stone tools (Brown and Emery 2008:322). 5. Low frequencies of high-heat burns ubiquito us over taxa and elements, but spatially discrete: Most surface deposits at the 19 known hun ting shrines show low levels of burning (Brown 2006). The number of burned faunal remains was identified at two of the three communal hunting shrines. About 24% of re mains at Pa' Ruchi' Abaj and 40% at Pa Sak Man were burned. The fact that burning was ubiquitous over taxa and elements and was the result of high temperatures, suggested that this resulted fr om on-site, rituallyrelated fires and not from cooking or ot her household activities (Brown and Emery 2008:322). 6. Correlation of weathering/animal alterations with local taphonomy: The effect of taphonomic factors such as animal gnawing a nd weathering correlate d with the physical characteristics of each hunting shrines. For exam ple, shrines that were more exposed to the elements had a higher rate of weathering wh ich suggests in-situ pr ocesses (Brown and Emery 2008:322). 7. Relatively high body-proportion completeness: Interviews conducted by Brown (2006:27-28) revealed three ca tegories of symbolically important bones that should undergo specialized discard at hunting shrines: 1) crania and mandible, 2) bones of large animals, and 3) all bones of hunted mammals These three categories reflect different community practices. For instance, Kaqchikel communities emphasized the importance of returning crania and mandibles. The three hun ting shrines analyzed in 2007 are located near the Tz'utujil towns of San Pedro la Laguna and San Juan la Laguna. Residents of these communities emphasized the importance of re turning every bone of the any hunted mammal. Most skeletal elements were presen t from each taxa at the three hunting shrines, therefore, these shrines we re characterize by high body-portion completeness. This patterning reflects verified what the Tz'utujil informants described. Although most skeletal elements and body proportions were found for each taxonomic group, there were some notable exceptions. For instance there was an absence of most cranial vault segments and antlers for deer, cranial elements for opossum and peccary, and a predominance of distal elements for tapirs (Brown and Emery 2008:318). 21


In addition to the zooarchaeological material signature, Brown and Emery (2008:314) recorded spatial and material data that included the ty pes of landscape features and the various activity areas associated with hunting ceremonialism. Spatial and Material Data Hunting shrines were located at freestandi ng volcanic boulders, exposed outcrops, rock shelters, and caves. With one exception the vast majority of hunting shrines were located at landscape features outs ide of town boundaries (Brown and Emery 2008:315). The Cofrada San Juan served as the only hunting shrine that was located within a community. This cofrada house, however, is strongly linked with mountain and cave symbolism (Christenson 2001). The house is conceived and decorate d as a sacred mountain with the doorway giving access to the cave interior (Christenson 2001:1). A number of stuffed wild animals are hung from the rafters of the cofrada house because local hunters bring offerings such as peccaries, coatis, raccoons, and mountain cats (C hristenson 2001:7). Three main activity areas were identified at the hunting shrines: offering areas, cleared spaces, and discard areas. The ritual offering area contained altars, sacrificial offering hearths, and bone caches. Ritual practiti oners would communicate with th e animal guardian via altars while simultaneously burning offerings in sacrificial hearths. Offerings commonly given included copal, flowers, candles, herbs, chocolate, fruits, tobacc o, and perfumed waters (Brown and Emery 2008:317). Many of the hunting shrines also contained cleared areas which were multi-purpose locations for both gathering spectators and performers involved in rituals, dances, and ceremonial drinking, Finally, the accumulation of debris from previous ceremonies were often swept clean and deposited in middens found in discard areas adjacent to the hunting shrines (Brown and Emery:2008:317). 22


Summary This study tests the proposed connecti on between contemporary hunting caches and lowland cave deposits using both th e zooarchaeological signature and the spatial/material data collected by Brown and Emery (2008). Therefore, in this chapter the connection between the modern hunting ceremonial beliefs and those recorded for the Col onial and more ancient Maya past are discussed. In addition, the specific archaeological markers that have been proposed as indicative of hunting ceremonialism the remains left as a result of hunting ceremonies are detailed because these are the markers that s hould be found in the ar chaeological record. In addition, this project incorporates a more in-dep th analysis of Maya ritual cave assemblages because it is recognized that caves may have been th e site of many other types of rituals and it is important to recognize the material markers of gene ral cave-related rituals in order to effectively separate these from those markers specific to hunting ceremonialism. 23


CHAPTER 3 PRIOR RESEARCH IN MAYA CAVE ARCHAEOLOGY AND ZOOARCHAEOLOGY Mary Pohl (1983), James Brady (1989), a nd Kitty Emery (2002) have all proposed zooarchaeological characterist ics of Maya ritual deposits drawing largely from cave assemblages. Mary Pohl (1983) proposed zooarc haeological markers that reflect cave-specific rituals of accession and fertilit y. James Brady (1989) proposed markers indicative of general ritual behavior carried out in caves that is distinctive of everyday subsistence activities. Lastly, based on the comparison between cave and surface site faunal assemblages, Emery (2002) suggested zooarchaeological marker s that might be used to reflect public, exclusionary rites, a particular type of ritual Emer y (2002) suggests were held at Cu eva de los Quetzales. A number of the characteristics proposed by Pohl, Brady, and Emery both overlap with and contradict many of the markers of hunting ceremonia lism proposed by Brown and Emery (2008). The question remains of whether any of the proposed zooarchaeological characteristics are truly reflective of pan-Maya cave cerem onies, generalized ritual behavior or hunting ceremonialism. This chapter provides a brief hi storical overview of Maya cave archaeology followed by a more in-depth discussion of the three zooarchaeological inves tigations of ritual cave deposits by Pohl (1983), Brady (1989), and Emery (2002). Maya Cave Archaeology The recent popularity of Maya cave studies have resulted in a number of lengthy discussions on this topic that can be found elsewhere (Brady a nd Prufer 2005b, Brady and Prufer 2005c). The history of cave research in Maya ar chaeology is often traced back to two of the disciplines founding figures, John L.Stephens and Frederick Catherwood (Brady and Prufer 2005b, Ishihara 2007), who described the caves of Bolonchen and Gruta de Chac. Early descriptive reports of Maya caves were plentiful with influentia l studies like Henry C. Mercers 24


(1975) The Hill-Caves of Yucatan Edward H. Thompsons (1897) Caves of Loltun, Yucatan and George Gordons (1898) Caverns of Copan, Honduras Although early descri ptive reports were available, an interpretive model of caves was not presented in the scholarly literature until J. Eric Thompsons The Role of Caves in Maya Culture in 1959 (Brady and Prufer 2005b). In 1975, this article was republished in Mercers The Hill-Caves of Yucatan where Thompson drew upon Maya ethnography, ethnohistory, and archaeology to outline a number of proposed uses for caves by the Maya people. Thompson (1975) conc luded that caves were used for sources of drinking water, sources of virgin water for religious rites, religi ous rites, burials art galleries, deposition of ceremonially discarded utensils, places of refuge, and other uses. Thompsons article stressed the ritual importance of cave s ites, however, many archaeologists at the time felt that habitation must have served as the majo r cave function, and thus denied the religious significance of caves (B rady and Prufer 2005b:3). Over time, the religious importance of cav es gained support in the archaeological community. Doris Heyden (1975, 1981) published a series of articles about her investigation of the Pyramid of the Sun at Teotihuacan th at emphasized the importance of caves in Mesoamerican cosmology. This majo r architectural feature was built directly on top of a cave system. Meanwhile, Barbara MacLeod and Dennis Pulestons published Pathways into Darkness: The Search for the Road to Xiba lba in 1978 based largel y upon their fieldwork investigating cave sites in Belize and the link between caves and the underworld. All of these major works eventually led to ar rival of cave archaeology in the 1980s as a key area of study in the Maya world (Brady and Prufer 2005b:5). After the 1980s, cave specialists began to emerge (Brady and Prufer 2005b:8). James Bradys (1989) dissertation, Juan Louis Bonor Villarejos (1989) Las Cuevas Mayas: Simbolismo y Ritual and Andrea Stones (1995) Images of the 25


Underworld: Naj Tunich and the Tradition of Maya Cave Painting were all key publications that stimulated interest in cave investigations and their gradual ac ceptance by mainstream archaeology. Maya Zooarchaeology and Ritual Deposits Until the work of Mary Pohl (1983) re latively little had been published on zooarchaeological characterist ics of Maya ritual assembla ges. In 1983 there was enough literature available to allow Pohl to investigat e patterning in Maya cere monial deposits. Using zooarchaeological data from supposed ritual cont exts (burials, caches, caves, and cenotes), she compared the ritual assemblages against midden deposits and drew her interpretations largely from animal depictions from ancient Maya art and the Dresden and Madrid codices. Her analysis developed three key characteristics of Maya ritual faunas: (1) th e differential use of taxa in comparison with midden deposits, (2 ) the prevalence of left-sided elements, and (3) a preference for juvenile individuals. These ch aracteristics are believed to reflect the use of animal remains in cave ceremonies associated with accession, calendrical, and fertili ty rituals (Pohl 1983, Pohl and Pohl 1983). Pohl (1983:100) argued that the procurement of fauna for ritu al purposes was a specialized activity in that animals specifically selected for ce remonial offerings tended to be taxa that were rare in midden deposits. This di fference, she argued, was most evident in the case of amphibians, snakes, birds, and fishes which occurred more fr equently in ceremonial contexts. Pohls second ritual characteristic came directly from her in terpretation of the faunal assemblage at Eduardo Quiroz cave in Belize. Among the avian fauna, 11 of the 14 skeletal elements were from the left side of the body. This occurrence paralleled Pohls previous res earch at the Guatemalan Petn site of Seibal (Pohl 1976) which revealed a statis tically significant number of left elements from white-tailed deer within elite and ceremonial contexts. Pohls suggestion about the ritual 26


importance of juvenile specimens was drawn primar ily from excavations at the site of Cuello on the Atlantic coast of Belize (P ohl 1983:62). Excavations uncove red a cache, designated Cache 363, which contained a large collection of white-t ailed deer mandibles in which 20 of the 30 minimum numbers of individuals were juvenile. Pohls work set the standard for all future inve stigations on the ritual use of animal remains in the Maya world. In fact, most subsequent i nvestigation used Pohls characteristics as the foundation of their analysis. Brady's (1989) investig ation of the faunal asse mblage of Naj Tunich is largely a critique of Pohls work. Brady (1989:371) noted that no consistent preference for a particular body side had ever been found for the entire faunal collection of any cave at that time nor had any patterning emerged for a particular species at more than one site. At Naj Tunich specifically, of the 24 bird bones recovered only 11 were from the left side and 13 from the right. Looking at white-tailed deer specimens, 57 of the 99 sideable elements were from the left side. Brady (1989:373-374) also argued that no special preference was shown for immature animals at Naj Tunich. Among mammals specifically, only 15 of the 44 minimum numbers of individuals were immature. Lastly, Brady (1989 :376) also rejected Pohls asse rtion of the difference between ritual and midden deposits. The taxonomic compos ition of the Naj Tunich assemblages was more similar to assemblages recovered from surface sites like Barton Ramie and Altar de Sacrificios than other cave sites like Actun Balam, Eduard o Quiroz, and Actun Polbilche. Although Bradys analysis was largely a critique of Pohl, he di d suggest three other characteristics of ritual assemblages based on the Naj Tunich assembla ge. Brady (1989:378) found that a lack of butchering, the relative intactness of skeletal specimens, and the minimal evidence of cooking was more suggestive of ritual, as opposed to subsistence, activity. 27


Kitty Emery (2002) investigated one third of the faunal assemblage at Cueva de los Quetzales and she also proposed several zooarc haeological characteristics of Maya ritual deposits. Emery (2002:101) argued that even seem ingly ritual contexts and taxa did not necessary provide evidence of ritually derived animal a ssemblages and that clear zooarchaeological correlates for di fferent ritual activities or categor ies were necessary in order to identify ritual deposits. Emery based her analys is on three ritual categories defined in the archaeological literature (Drennen 1976, Flannery 1976, Goody 1982, Hayden 1995, Dietler 1996, Hayden 1996, Mennell 1996): private rituals, public exclusionary rituals, and public inclusionary rituals. These categories were then linked with animal-specific formalized behaviors and material correlates drawn from Maya ethno graphic analogies, iconographic analyses, and ethnohistoric documents (Emery 2002:104-105). Private rituals were characterized by the use of sacred specific or rare individua ls, sacrifices of individual anim als, and the use of individual performative or transformational paraphernalia. Public exclusiona ry rituals included the use of lineage or totem species, the use of exotics or high status goods in competitive displays, a show of control of managed species through sacrif ice of special individuals, and small scale performances. Public inclusionary rituals incl uded large-scale feasting, with sacrifices that emphasized quantity not quality of goods, and elaborate, theatrical performances. To test for correlates of these three ritual activity categories in the zooarchaeological assemblage of Cueva de los Quetzales, Em ery (2002) looked for patterning in taxonomy, population characters, body portions and sides and compared such patterns to the non-ritual faunal assemblage recovered from the nearby su rface site of Arroyo de Piedra. The remains found in Cueva de los Quetzales had a greater number of species that Emery defined as potentially sacred including bats, marine toads, and opossums. Looking specifically at managed 28


species, Emery found that both the Cueva de los Quetzales and Arr oyo de Piedra faunal assemblages had roughly equal proportions of bot h deer and peccary, while those from Cueva de los Quetzales had significantly more dog and turkey (Emery 2002:107). In her comparison of individuals of different ages (mature vs. immature) Emery (2002:108) found that although the overall proportions were similar between surface and cave site assemblages, the proportion of immature managed and sacred species was much higher in the cave asse mblage. These results indicated the selective deposition of specific imma ture individuals of managed and sacred animal species which she concluded repres ented a pattern indica tive of public exclusionary rituals in the cave. Overall skeletal completeness for both medium and large mammals was also higher in the cave assemblage. This pattern held for specific ta xa as well, most notably deer and dog. Emery (2002:109) also found differential use of body sides at Quetzales, showing that all artiodactyls, especially deer, were more frequently represente d by left-sided skeletal elements. Based on these results, Emery (2002:111) suggested that Maya ca ve ritual animal use was characterized by an emphasis on symbolic underworld species, left-sid ed elements, juvenile managed species, and whole body carcasses of managed species. Taken as a whole, these three studies appear to present an unclear pi cture of Maya ritual deposits. What do these zooarchaeological patterns actually mean? Both Pohl's and Brady's work suggested blanket characteristics of pan-Maya r itual behavior. They failed to take into account the variability in cave faunal assemblages throug hout the Maya region and variability in ritual practices that must have produced those assemb lages. Emery recognized the need to understand the diversity in potential ritual activities in caves but did entertain the possibility of different ritual categories. The questions remains whether any of these proposed zooarchaeological 29


characteristics are truly reflective of pan-Maya cave ceremonies, a specific ritual category like hunting ceremonialism, or so me other unexplored factor. 30


CHAPTER 4 CAVE SITE BACKGROUND Nine cave sites were incorporated into this analysis. Animal remains from Cueva de los Quetzales, Cueva de Rio Murcielagos, and Aguateca Grieta were analyzed by the author and form the core of this analysis. The remaining six cave sites include Naj Tunich, Eduardo Quiroz, Actun Polbilche, Actun Balam, Stela Cave, and Caves Branch Rockshelter. These sites were chosen as comparative samples to provide a br oad range of assemblages and because they are available in the published liter ature with sufficiently detail ed data to allow comparison. Background information pertaining to the histor y of occupation, excavation, and analysis was provided for each cave if it was available. Although this comparative analysis treats all of the cave sites as similar in terms of occupation period, basic structure, excavation history, etc. these characters do vary between sites and these variations will affect this analysis. An in-depth discussion on the potential effects of this va riation is provided in the methods section. Cueva de los Quetzales Cueva de los Quetzales is part of a cave system that runs under the site of Las Pacayas, an elite political center occupied from 300 BC to AD 500. Excavations were carried out by the Atlas Arqueolgico de Guatemala under the direc tion of Dr. Juan Pedr o Laporte and Hector Escobedo. Previous zooarchaeological analysis wa s conducted at this site by Kitty Emery from 1999 to 2002 which covered roughly a third of th e total faunal assemblage (Emery 2002). These remains are included in the analysis presented in this study. Three caves make up the system that runs under Las Pacayas, Cueva de los Quetzales is the principle one. The surface of Las Pacayas was constructed on a natural hill that was transformed into a large stepped platform. A heavy concentration of construction was found atop this hill including pyramids and plaza groups. Th e main ceremonial complex incorporates a 31


skylight entrance into Quetzales at the center of the plaza and to the side of the largest pyramid (Brady 1997:608). Next to this entrance excavators found a round a ltar. A conical deposit three meters deep and ten by four meters wide compos ed of faunal material, ceramics, and many other artifacts was found directly under the skylight opening (Brady and Rodas 1995). Due to the shape of the deposit and its location relative to ceremonial architecture, Brady and Rodas (1995) argue that offerings from ceremonies held on th e surface of Las Pacayas were dropped down into the cave. The cave deposit was trowel excavated in ten centimeter arbitrary levels and the matrix was water-screened using a 4 mm gauge mesh (Emery 2002:103). All materials were considered to form one large unstratified pr ovenience in this analysis due to a lack of cultural or natural subdivisions. In terms of taphonomy, there was no evidence of water activity at Quetzales during or after the occupation at Las Pacayas (Brady and Rodas 1995), therefore it is unlikely that materials have been transported or eroded by water. Due to its distinct morphological characteristics, Kitty Emery (2002:104) argued agai nst a significant role for intrusive animals as inclusions in the deposit or as taphonomic agents. Since the cave entrance is located in the center of bustling administrative center, it would have been largely inaccessible to wild animals. Cueva de Rio Murcielagos Rio Murcielagos is one of the 22 caves discove red in the immediate area of Dos Pilas by the Petexbatn Regional Cave Survey direct ed by Dr. James Brady (Brady 1997). The Petexbatn Regional Cave Survey was a four-yea r long subproject of Vanderbilt Universitys Petexbatn Regional Archaeological Project dire cted by Dr. Arthur De marest. The project investigated the role of caves in prehistoric Maya sacred ge ography and specifically sought to determine whether the placement of surface architecture depended on the location of cave features (Brady 1997:604). Major ce remonial structures at the Do s Pilas site, such as the El 32


Duende complex and the Bat Palace, were found by Brady's team to be spatially linked with underlying cave features. Cueva de Rio Murcielagos is located 2.5 km southeast of the main plaza at Dos Pilas. Mapping and surveying of the cave was conduc ted in 1993 (Brady et. al. 1991). A 1 x 2 m test pit was excavated at the entrance of the cave which uncove red evidence of three plastered floors. Early Classic Period ceramics were recove red from the uppermost floor, while Preclassic ceramics sherds were found in the bottom two floors (Brady et. al. 1991:668). Emery completed analysis of 25% of the remains from this site. These were included in this analysis. Aguateca Grieta The Grieta Principal is a deep and long chasm running northeast -southwest through the center of the site of Aguateca lo cated along the Petexbatn River in the Pasion River drainage of El Peten, Guatemala (Ishihara 2007:4 3). Aguateca is a Classic Period Maya center that served as the dual capital of the Petexbat n kingdom along with Dos Pilas. The Grieta was investigated during the second phase of the Aguateca Archae ological Project under the direction of Dr. Daniela Triadan, Dr. Takeshi Inomata, and Licenciado Erik Ponciano. Excavations were conducted during 2004 and 2005 by Dr. Reiko Is hihara. Approximately 860 m long, 10-70 m deep, and 1.5-15 m wide, the chasm has major cons truction located on either flank. Both palace and elite residential groups ar e located on the eastern flank and the Plaza Principal and elite residential construction can be f ound to the west (Ishihara 2007). Five primary areas (suboperations) were mapped, surface collected, and excavated during the two year project. These regions were de signated the Hidden Jar Area, the Two Owls Area, the Southern Entrance Area, Chill Hill Area, a nd Windy Valley. In addition, Grieta Rincon is at the southeastern end of the site and adjacen t to an area known as Barranca Escondida. The 33


Barranca was excavated in 1999 by Markus Eberl. Rincon consists of two areas: an open Lshaped grieta component and a cave (Ishihara 2007:48). Each of the five primary suboperation areas of the Grieta Principal, the Grieta Rincon, and Barranca Escondida contained faunal material that was analyzed fo r this project. A total of 33 excavation units were opened with 29 units in the Grieta Principal and 4 in the Grieta Rincon (Ishihara 2007:107). Excavation was conducted using both natural and cultural units. The majority of the matrix was wet clay, therefor e, screening was limited. Excavators checked the matrix by hand prior to its placement in the backdirt pile. Previous zooarchaeological analysis of the Aguateca Grieta material was analyzed by Dr. Kitty Emery and Michelle LeFebvre and published in Ishihara's (2007:Table 8.1) dissertation. Their identifica tions were incorporated into this analysis. Naj Tunich Naj Tunich is located in the limestone hills of the Maya Mountains in the southeast corner of El Peten, Guatemala. It resides in the munici pality of Poptun, Peten wi thin one kilometer of the Belize border. From 1980-1989, the site unde rwent numerous investigations including extensive mapping of its passages, photogra phy of its wall art, and excavation of the archaeological material. Brady (1989:436) suggests that Naj Tunich primarily served as a pilgrimage from the Late Preclassic period to the Late Classic Period. Its most intense use occurred during the Early Classic Period. Th e cave contains exte nsive architectural modifications, tombs, and nearly 500 hieroglyphic inscriptions. These inscriptions suggest that the cave was visited by elite members of Maya so ciety (Stone 1995). Zooarc haeological analysis was conducted by Susan Colby and presented in Brady's (1989: Appendices 9.1-9.3) dissertation. 34


Caves Branch Rockshelter Caves Branch Rockshelter is situated in the Caves Branch River Valley of Western Belize. It served primarily as mortuary ritual locati on used by a local, rural village population and is estimated to hold at least 150 individuals (Glassman and Vill arejo 2005, Wrobel 2008). Interred skeletons consisted of a large number of infant s as well as adult men and women, which suggests that the cave was used by everyone in a small community adjacent to the site (Wrobel 2008). Initial excavation at the site was conducted in 1994 and 1995 by Juan Luis Bonor Villarejo. From 2005-2008, the Belize Valley Archaeological Reconnaissance Project resumed excavation of Caves Branch. Ceramic vessels, interred as gr ave goods, date from the Late Preclassic through Terminal Classic periods. Due to the extensive bioturbation as a result of grave digging, none of the sites natural stratigraphy was left in situ. Th e general matrix of the site consisted of faunal remains, chert, freshwater snails, carbon fleck s, and large amounts of ceramic sherds. Wrobel (2008) suggests that these artif acts represent surface offerings left in the rockshelter that was later disturbed and spread throughout the matrix by continuous bioturbation. Faunal identifications were conducted at this s ite by Erol Kavountzis in the Environmental Archaeological Program laboratory of the Fl orida Museum of Natural History. These identifications are used with his permission (Kavountzis 2009). Stela Cave Stela cave is found in the Macal River Valle y of western Belize. In the summer of 2003, Stela cave was investigated by the Western Beliz e Regional Cave Project under the direction of Dr. Jaime Awe (Ishihara and Griffin 2004). Ma pping and excavations were conducted to examine the extent and nature of the architectural features and thei r role in the use of space in the cave (Ishihara and Griffith 2004:57). Ceramics recove red from the site span Maya prehistory, but the majority of the material is Late Classic. In terms of taphonomy, it is a dry cave and easily 35


accessible. Evidence of modern hum an activity, including looters pits and modern garbage, was found. Faunal identifications we re conducted by Erol Kavountzis a nd are used here with his permission (Kavountzis 2009). Eduardo Quiroz Cave Eduardo Quiroz Cave is located in the southe rn part of the Cayo District of Belize known as Chiquibul, northeast of the site of Caracol (Pendergast 1971). No surface site was found in its immediate area. It lies at the top of a tall us slope about 17 meters high and surrounding by a limestone cliff rising above the entrance. It was excavated in 1961 by David Pendergast and are assumed to have been hand excavated. Faunal identifications were made by Howard Savage of the Royal Ontario Museum and can be found in Pendergast (1971: Appendices A, B). Based on the ceramic evidence, Pendergast (1971) argues th at activities within the cave were mostly carried out during the Late Classic Period. The presence of one sherd from an Early Classic ceramic vessel, however, has led him to apply this date with some caution. The cave consists of six chambers, four of which have some kind of artificial construction such as masonry walls and plastered floors. Mollusca remains dominated the faunal assemblage and were found in every chamber except Chamber 5. Non-molluscan remain s, however, were restricted to Chamber 1 near the entrance of the cave. Actun Balam Like Eduardo Quiroz, Actun Balam is also located in Belize in the Cayo District of Chiquibul near the Maya Mountains (Pendergas t 1969). It was excavated by David Pendergast in 1964. There were no surface sites found in the imme diate vicinity of Actun Balam. Pendergast (1969:58) suggests that Maya people traveled from great distances to make offerings as a form of religious pilgrimage. The cave is composed of five main chambers, however, all of the faunal materials was located in Chamber C in one large cone shaped mound underneath a chimney 36


37 opening in the caves ceiling (Pendergast 1969). The conical deposit contained polychrome cereamics, figures, items of odornment, and stone tools. Based on his ceramic analysis of over 22,000 ceramics sherds in Chamber C, Pendergast (1969:58) argued that this cave was primarily used during the Late Classic and Termin al Classic/Early Postclassic periods. Actun Polbilche Actun Polbilche is situated on the southern si de of the Sibun River in Belize in a karstic region in the eastern part of the Maya Mountai ns (Pendergast 1974). The cave consists of one main chamber, Chamber 1, and two main alcoves. Exploration of the cave was conducted by David Pendergast in 1971 and mostly resulted in surface collecting and mapping of ceramic vessels. Pendergast (1974:810\) argu ed that the bulk of the cera mic vessels placed the primary activity within the cave in the Terminal Classic and the beginning of the Post-Classic period. It should be noted, however, that ra diocarbon dates from pine splints found in Alcove 1 were dated to AD 735. Recently, these ceramics have been reanalyzed by Loez Varela in which he has also uncovered Late Formative and Early Classic cer amics (Peterson 2006). In total there were 5129 remains identified by Elisabeth Luther. These re mains were recovered from the surface and the top 6-7 cm of the soil from Alcove 1 and from a small niche at the north end of Alcove 1 in association with a broken ceramic vessel. Drilled mammalian teeth make up 897 of these remains.


CHAPTER 5 THEORETICAL APPROACH This analysis adopted a practice theory appr oach for investigating ritual. The chapter begins with a brief description of practice theo ry followed by a more detailed discussion on how practice theorists tend to approach the subject of ritual. Finally, there is an overview of William Walker's life-history approach which forms the methodological framework for this study. Practice Theory In its most simplistic form, practice theory l ooks at the relationship between agency and structure. This approach is an explicit rejection of functionalist interpretations of behavior concerned with why people behave in particular ways and symbolic approaches interpreting what behavior means. Contemporary practitioners prim arily draw from the work of two scholars, Pierre Bourdieu (1977, 1990) and Anthony Gi ddens (1979, 1984). Giddens emphasized the "duality of structure, how people both shape and are shaped by the structures and conditions in which they live. Bourdieu examined how social structures became internalized and replicated through individuals' habits or habitus Within this sort of theore tical framework, practices are seen as the embodiment of people's habits or disp ositions. Some of these dispositions have what Pauketat (2001:80), following Bour dieu, refers to as 'doxic' referents which can be unconscious, spontaneous, nondiscursive, practical or commonsensical forms of knowing. It is important to stress that although practices are shaped by di spositions, they are not held hostage to them. Practice theory puts a large emphasis on the notion of agen cy (Dobres and Robb 2000). Agents are powerful in the sense that they can influence the social, symbolic, and material conditions of their lives. At the same time, individual agency is tempered by these same structures. Nonetheless, a practice approach ac knowledges that agents motivations and actions, both conscious and unconscious, are extremely important. 38


Practice theory also attempts to understand and emphasis historical change by focusing on the constant interaction between agents and structure. All people enact, embody, or re-present traditions in ways that continuously alter those traditions through their daily practices (Pauketat 2001:79). The important of agency, practice, and hi storical change are highlighted in greater detail in the following section. A Practice Approach to Ritual There are a number of nuances that disti nguish variability in how practice theorists approach ritual; however, Catherin e Bell (1997) suggests that in ge neral they tend to share four characteristics. First, ritual is seen as a historical rather than stagnant process, where practices can be reproduced, reinterprete d, and transformed. Second, practice theorists are concerned with what rituals do rather than what they mean. A third characteristic is the acknowledgment of individual agency. Finally, many practice theories suggest that ritual as a category should be removed from analysis in orde r to investigate the particular logic and strategy of cultural practices. Despite the fact that rituals often present th emselves as time-honored customs, practice theorists emphasize that rituals do change over time. Functional and symbolic approaches to ritual tend to view it as a repe titive and stagnant action. Ritual s give the impression of endurance through time which provides a sense of legitima cy and protects it from potential frivolous alteration (Bell 1997:211). The hist orical roots of ritu al theory help to explain this phenomenon. According to Bell (1997), most theories of ri tual are rooted in et hnographic observation of nonliterate societies, which tend to de-emphasize hi storical change. Yet as archaeologists studying ritual note, (Kyriadkidis 2007, Marc us 2007) the continuity of ritu al practice does not imply the continuity of ritual belief and vice versa. 39


Symbolic approaches to ritual tend to pl ace an emphasis on what rituals mean, while practice theory is more concerned on what ri tual does. Bell (1992, 1997) shows that discourse about ritual is largely based on the initial bifu rcation of thought and action. In this framework, ritual is seen as a particular type of action that expresses i nner beliefs. This action is often characterized as routine, habitual obsessive or mimetic. Thus, "rit ual as action" is placed in a secondary position, the mere physical expression of logically prior thoughts (Bell 1992:19). Within archaeology, this view mani fests itself as a tendency to e quate ritual with meaning and symbolism. Material evidence of ritual is seen as a physical e xpression of past belief systems, and thus we attempt to analyze these objects in terms of what they meant or symbolized to people in the past. This is in sharp contrast to how we view more quotidian objects, which are often analyzed in terms of th eir utility and function (Brck 1999, Stahl 2008, Walker 1998, Bell 1992). In archaeological inference, mo st often an objects physical fo rm is used to identify its function and by extension the function of the places or cont exts in which it was found. For example, the presence of u tilitarian-looking cooking vessels is used to infer domestic architecture or household activit ies (Walker 1998:246). It is ofte n assumed that practical-looking objects (i.e. have known utilitarian forms) were not used in past ritual activities. The distinction between ritual and quotidian objects is far less intuitive. As Walker (1998:249) has pointed out, even the most mundane cooking pot can be symbo lic and practical-looking objects are often used in ritual activities. A practice a pproach to ritual recognizes th at an artifacts form does not necessarily determine its function. The brief description of practice theory refe rred to the emphasis placed on agency. Human behavior is not seen as a st atic manifestation of underlying cultural rules. Rather, human 40


practices are strategic and allow for both the reproduction and reorganization of social and cultural environments (Bell 1997). This also applies to ritual pr actice. Bourdieu characterizes ritual as strategic practices for transgressing and reshuffling cultura l categories in order to meet the needs of real situ ations (Bell 1997:78). A lthough the agency of human bodies are more commonly discussed in the literature, there is a growing concern with the animacy and agency of non-human objects ( Gell 1998, Latour 1993, Ingold 2000) and its material manifestations in the archaeological record (Brown and Walker 2008, Brown and Emery 2008, Mills and Ferguson 2008, Zedeno 2008, Walker 2008). A strictly defined category of ritual is a di fficult concept for many practice theorists. Many have proposed to get rid of the term entirely (Bell 1997). Practice th eory, in general, tends to shy away from any kind of universal definition. The cr eation of an essentialized category of action known as "ritual" often results in a pursuit for universal, cross-cultural comparisons. Bell (2007) argues that such an approach detracts attention from the comparison of different ritual activities within a culture. This conforms to what most practice theorists tend to emphasize; that ritual should be analyzed and understood in its real context, i.e. within a particular cu lture at a given point in time (Bell 1997). Walker's Life History Approach William H. Walker's (1995, 1998, 1999) life hist ory model attempts to investigate interactions between people, objects and the supernatural realm. It is an approach that integrates ethnographic and archaeological data in order to interpre t the material remains of past ritual behaviors. An objects life-history, from a beha vioral archaeology perspective, refers to the unique path an object follows from creation or ac quisition to its eventual disposal and entrance into the archaeological record (Schiffer 1972, Brown 2004). As the primary unit of analysis, Walker advocates using the life-history approach in order to identify past ri tual technologies. It is 41


argued that all objects go through a series of stages in their life history that include manufacture, distribution, use, reuse, discard, and post-deposition. Walker ( 1999) argues that each of these stages can leave a distinctive trac e that can manifest itself in any or all of these stages. The behaviors that comprise the life histories of objects affect four dimensions of artif act variability: numerical frequency, formal attributes, associat ion with other objects, and spatial locations (Walker 1998:247, Walker 1999). In aggregate, thes e measures can distinguish the function of objects, including a ritual function. Identifying "function" is a particularly important component for Walker in analyzing ritual. This approach has its roots in behavioral ar chaeology, originally defined as the study of the relationships between human behavior and artif acts in all times and places (Walker 1995:71). This is a material-focused approach that includes a broad definiti on of technology. Technology refers to the study of the use char acteristics an object possesses. Th is allows the analysis of both quotidian and ritual objects as components with in technologies. Like practice theory, this approach also rejects the initial bifurcation of thought and action di scussed previously. In Walker's 1995 study of ceremonial trash, he adm onished archaeologists for reducing ritual assemblages to a collection of ar tifacts without uses. He argued that once archaeologists focused on ritual behaviors, rather than beliefs, the ma terials actions involving ac quisition, use, control, and discard of artifacts can be examined. Such an approach challenges pre-conceived notions of utilitarian vs. non-utilitarian objects. In the life-history approach, a ll interpretations of artifact function should be grounded in known interactions between people and objects, preferably from ethnogr aphic and historical sources (Walker 1998). Under this appro ach, ethnography forms the foundation for understanding the material patternin g in the archaeological record. In Walker's (1998) analysis of 42


ceremonially ordered deposits of Southwestern pithouses and kivas, he drew upon ethnographic sources to suggest that these deposits held th e remains of persecuted witches. Walker (2008:144145) approaches the importance of ethnography a nd oral tradition in wa y that parallels his approach to ritual. He argues that oral tradition is easy to dismi ss as inaccurate because it is often conceptualized as a thing of id eas much like ritual is conceptu alized as a thing of beliefs. Treating oral tradition as a form of practice, the accounts become hist orical events and the practice of oral traditioning contributes to the making of history. Walker (2008:149) argues origin stories and oral traditi ons can be used to both explai n past actions and simultaneously provide contexts for future actions. Patterns in th e archaeological record may represent indirect traces of past oral traditions. In 2008, Walker expanded his in terpretation to suggest that dog remains found within these same settings were also considered witches or perhaps as guardians. Thus like practice theory, his approach recognizes both human and non-hu man agency. In animistic or relational societies, ritual events often involve the interaction between hu man and non-human social actors. Animism is defined as an ontology in which objects and other non-human beings possess souls, life-force and qualities of personhood (Tylor 195 8, Brown and Walker 2008:1). Under this ontology, many objects such as animals, plants, rocks, landscape features, houses, etc. possess both animacy and agency. Animism and object agency has been an important topic of discussion of both social theory and ethnography (Gell 1998, Latour 1993, Ingold 2000), but has recently spurred interest in the archaeological commun ity as well (Brown and Walker 2008, Brown and Emery 2008, Mills and Ferguson 2008, Zede no 2008, Walker 2008). Brown and Walker (2008:298) describe object agency as the causal consequences objects have on the course of human activity, and this includes animate objects as well as the pe rformance characteristics of 43


44 material things. Animacy is similar in that all animate objects possess agency, but this agency is seen as autonomous, purposeful, and deliberate. It arises from sentient qualities possessed by the object, like consciousness or life-force (Brown and Walker 2008:2). Once we recognize that objects can possess both animacy and agency, th e lines between object and subject become blurred. The realization of how animistic practices might have shaped the materials we find in the archaeological record goes largely ignored and re legated to their categorization as beliefs and therefore either irrelevant to our interpretati on or impossible to investigate. Under a practice approach, however, an understanding of non-obj ect agency is fundamentally important for archaeological interpretation because the archaeol ogical record possesses material remains that often are the result of interactions be tween human and non-human social actors.


CHAPTER 6 METHODS The following chapter outlines the methodological approach incorporated in this analysis. Basic zooarchaeological procedures referring to identification and basic measures are discussed first. These procedures were used to identi fy the faunal material from the primary cave assemblages, Cueva de los Quetzales, Cueva de Rio Murcielagos, and Aguateca Grieta. Comparative cave sites were compiled from th e published archaeological reports as well to provide a more representative sample of lowla nd cave assemblages. The exact location of each of faunal reports compiled for this analysis wa s provided in the site background chapter. This chapter also discusses the potenti al sources of bias for this st udy and the specific calculations used to compare the sites for possible ritual activities. These calculations are presented in a conceptual framework that allows for the consideration of these assemblages in terms of the behaviors that may have created them. Zooarchaeological Analysis Identification and analysis of the primary faunal assemblages (Cueva de los Quetzales, Cueva de Rio Murcielagos, and Aguatca Grieta) followed standard zooarchaeological procedures (Reitz and Wing 1999). Identifications were ma de using the modern comparative materials housed in the Environmental Archaeology and Ma mmalogy Collections of the Florida Museum of Natural History. The following information was recorded for each skeleton element when applicable: taxa, element, element portion, comp leteness, side, age, cultural modifications, taphonomy, and pathologies. Age determinations we re based on the degree of epiphyseal fusion, the closure of cranial sutures, and dentiti on eruption (Reitz and Wing 1999:161-162). Cultural or human modifications that were examined include signs of butch ering, burning, and artifactual modifications. The following taphonomic markers were also recorded: animal gnawing, water 45


damage, weathering, and concretion (calcium car bonates or dried clay). These methods of identification are the same as those adopted in the hunting cache research (Brown and Emery 2008:308-309) to facilitate comp arison between assemblages. The zooarchaeological assemblages were s ub-sampled in the interest of time and comparability of the archaeological samples w ith the hunting cache assemblages. All mammalian specimens equal to or larger than the size of a pocket gopher were identif ied and all bird, fish, reptile, amphibian, and shell remains were excluded from analysis. This subsample is appropriate because it corresponds to the material signat ure of the hunting caches where only mammalian remains were found (except turkey) and the sm allest identified taxon was the pocket gopher. Although this strategy provides a more comparable data set, it does bias interpretation of the archaeological sites. It is entirely possible that some material markers of ritual activity included the use of non-mammalian and small mammal species. Mary Pohl (1983) ha s suggested that the difference between ritual and subsistence faunas is most evident in the case of amphibians, snakes, birds, and fishes which are more commonly identified in ritual deposits. Unfortunately, this study will not be able to evaluate these mark ers but later studies should expand the analysis to include all taxa. Quantification All counts were calculated at the site level which means that differences in provenience, stratigraphy, or time periods were not taken into account. This method was adopted due to the unique nature of cave stratigraphy and the difficulty of obtaining accurate dates for these assemblages. Basic quantifications were base d on two measures, the number of identified specimens (NISP) and the minimum number of individuals (MNI). NISP is a count of the number of identified fragments per taxon. NISP valu es for this analysis refers to all mammalian specimens identified to at least the class leve l (ie. Mammalia) or lower. NISP is generally 46


presented as a relative frequency to allow compar ison between samples of different sizes. This measurement is subject to a number of biasing fact ors. Identifiability is one of these factors. Some specimens are easier to iden tify than others while some species simply have more bones in their skeletons than others. Fragmentation of sk eletal elements is another issue because it can result in inflated NISP values, and the degree to which elements are susceptible to fragmentation varies between species, elements, and age classe s. Despite these potentia l sources of bias, NISP is still a good measure for comparison between faunal assemblages since it is a relatively objective and unmodified count that refers directly to the original faunal assemblage. Therefore, it is easy to calculate and re plicate (Grayson and Frey 2004). MNI is an estimate of the minimum number of individuals within an assemblage. Like NISP, MNI is also affected by site formation pr ocesses, recovery techniques, and laboratory procedures (Reitz and Wing 1999:195). MNI takes in to account other attributes such as element, side, age, and provenience. It is not subject to fragment ation issues or differences in the number of skeletal elements between sp ecies. Nonetheless, there are drawbacks to using MNI values. MNI can overemphasize the im portance of rare taxa (Grayson 1984, Reitz and Wing 1999:197198). The presence of a single specimen of a speci es will be weighted as heavily as several specimens of a more common species (Grayson 1978). MNI is also subject to aggregation affects in that MNI values are influenced by how data from archaeological proveniences are aggregated during analysis. MNI values will differ if subdivisions like time periods, excavation units, etc. are taken into account in calculating MNI values More conservative MNI estimates will result from the aggregation of separate faunal samp les into one analytical unit (Reitz and Wing 1999:197). Due to the potential for aggregation eff ects and the lack of te mporal control of the cave assemblages under analysis, al l MNI counts were calculated at the site level for this study, 47


the most conservative assessment. MNI calcula tions took into account taxa, element, portion, side, and age. Despite the potential drawback s of both NISP and MNI, many of these biasing factors can be overcome by careful evaluation of the data, cautious zooarc haeological analysis, and presentation of both counts to allow comparison of the re sults (Reitz and Wing 1999:199). Other Sources of Bias Differences between cave assemblages in terms of taphonomic conditions, sampling methods, excavation methods, and zooarchaeology methods introduce some degree of bias into this analysis. Each cave had its own unique taphonomic histor y which affects the overall character of the faunal assembla ges housed within. For instance, the caves differ in terms of exposure to the elements and ease of accessibility to both human and animal visitors. Therefore, some sites have conditions better suited for preservation of skeletal specimens than others. Recovery methods were also not uniform. Many of the sites were formally excavated, while others were only subject to surface collecting. In addition, most of the caves sites were not screened. Whether or not a site was subject to screening will imp act the chance of recovery of smaller taxa and thus favor larger specimens. This may not be a huge fact or in this particular analysis since the sampling st rategy eliminates small specime ns already. Nonetheless, the comparable assemblages obviously did not incorpor ate the same sampling strategy. The bias that results from eliminating non-mammalian taxa was addressed above, but other sampling concerns remain. It is unclear, among the ol der cave sites in particular, wh ether the faunal identifications from each individual cave site represents the complete site faunal assemblage or just a subsample. Lastly, differences in zooarchaeologi cal procedures can affect the results of this analysis. Identifications from the comparable assemblages were done by a variety of analysts who each had their own methods for approaching th e cave assemblages. The majority of the cave faunal reports did not outline their zooarchaeologi cal methods. It is probable that MNI/ NISP 48


values and age class determinations in particular were calculated differently than the approach used in this analysis. Testing Ritual "Signatures": The Conceptual Framework In order to test the cave faunal assemblages for material markers of hunting ceremonialism and other ritual markers proposed in the literature, a conceptual fram ework of analysis was drawn from Walker's life-history approach. Wa lker (1998, 1999) outlined four dimensions of artifact variability that are useful for analyz ing ritual deposits: formal properties, numerical frequencies, associations, and spatial locations. These dimensions are the foundation of this projects methodological framework in which two major questions are being asked. The first is whether the Petexbatn cave assemblages are comp arable to other Maya ritual assemblages, particularly other cave assemblages. This que stion was analyzed by co mparison to proposed zooarchaeological characteristic s of ritual assembla ges outlined in the work of Mary Pohl (1983), James Brady (1989), and Kitty Emery (2002).The second question is the one most central to this study and asks whether lowland cave assemblages reflect hunting ceremonialism. Formal Properties/Numerical Frequencies The categories of formal properties and nume rical frequencies are combined in this analysis and are defined according to Walker (19 99:385). Formal properties of artifacts refer to characteristics like completeness, shape, size, colo r, etc. Numerical freque ncy refers to counts of artifacts or characters. Most of the ritual markers proposed for ritual cave faunas could be placed in either category depending on the way the data is analyzed. For example, bones could be either burned or unburned (a formal property) or 10% of all bones could be burned (a numerical frequency). Therefore, these two categories were combined. 49


Density Density was measured as the number of bone fragments (NISP) per meter squared. Only two cave sites, Cueva de los Quetzales and Aguateca Grieta, had the necessary spatial data to test density. Cueva de los Quetzales was calculated as one large unit. Aguateca, on the other hand, had multiple excavations units, each of which wa s calculated separately to provide a range of density values for the site. Taxa Using relative NISP and MNI counts, the di stribution of taxa from both primary and comparative cave assemblages were compared with those from Brown and Emerys research on the Atitln hunting caches. I followed curre nt taxonomic nomenclature as provided on www.itis.gov [June 2, 2009]. All taxonomic identifica tions from the previously analyzed comparative archaeological assemblages use th e taxonomic names provided in the original reports; none were changed to reflect modern taxonomic revisions. Subgroupings of related taxonomic units were not created. Although this may have eased comparative interpretation, it would have altered the original data. The NISP and MNI counts were also left as is, however the relative proportions of each of these measur es (%NISP, %MNI) were recalculated for this study in order to reflect the elimination of non-mammalian taxa from analysis. MNI values were not calculated for Actun Polbilche, Eduardo Quir oz, and Actun Balam by the original analyst. Therefore, MNI had to be calcula ted using the faunal identification information provided in each of the cave reports (Pendergast 1969:56-57, Pendergast 1971:Appendix A-B, Pendergast 1974:65-79).The comparison of MNI and NISP c ounts should show the ubiquity and abundance of various mammalian species in the cave assemblages. 50


Diversity Measures of diversity (or heterogeneity), richness, ev enness, and dominance were calculated for each site. Heterogeneity measures combine the concepts of species richness and equitability. Richness refers to the number of taxa within an assemblage, while equitability measures the degree to which species are equally abundant (Reitz and Wing 1999:103). Simpson's (1949) index of heterogeneity and Odum's (1971) richness measure were used for this analysis. Simpson's formula is D= {N(N-1)/ n(n-1)} where D= hete rogeneity, n= specimen count/species, N= total specimen count. Odum's Richness formula is R= (s-1/logN) where s= number of taxa and N= total specimen count. Equ itability (i.e. evenness) was calculated as the similarity in abundance of several taxa within a sample. The relative dominance values of the most frequently occurring species in each assembla ge was calculated as the ratio of the first to the second species in rank-ordered distribution. A ll community statistics were based on absolute MNI counts. MNI and not NISP values were used because community statistics are measurements of individuals and not measures of body parts. MNI values allow for more effective comparison among taxa between the various cave sites. Using MNI, however, introduces some bias because it is an interpre tive measure and is more heavily influenced by small sample sizes. Age classes Both Pohl and Emery (Pohl 1983, Emery 2002) s uggest that ritual assemblages contain higher proportions of juvenile speci mens. To test whether this is the case at the Petexbatn sites, the relative percentage of each age class was calculated in terms of MNI. As previously mentioned, the age of each skeletal elemen t was determined by epiphyseal fusion, cranial sutures, and dental eruption where applicable Mammals, and other animals with determinate growth, tend to have similar ra tes of skeletal fusion. Certain elements tend to fuse early among 51


all mammals, usually within the first 12 months of life, while other elements are known to fuse much later. Using comparative charts of m odern fusion and eruption rates (Reitz and Wing 1999:76) skeletal elements were placed in th e following age classes: juvenile, immature, immature+, subadult, and adult. The following general characteristics were used to classify the materials. Juvenile specimens are unfused at bot h the early and late epip hyses and the diaphyses usually have juvenile cortex. Juvenile specime ns can also be identified by the presence of deciduous teeth. Immature specimen s are unfused at the late ep iphysis and either fused or bridged at the early epiphysis. Specimens are cat egorized as subadult when long bones are fused at early epiphyses and bridged at late epiphyses or when te eth are erupted but unworn. Adult specimens are either fused at bot h early and late epiphyses or ha ve adult teeth that are erupted and worn. Finally, immature + is a category that refers to specimens that are fused at the early epiphysis but the condition of the late epiphys is is unknown (usually because the specimen is fragmented or incomplete). This specimen could be either immature, subadult, or adult. These age classifications were then used to determin e how many individuals were represented by each taxon and their relativ e ages at death. Elements, body portions, a nd skeletal completeness Specific skeletal elements a nd body proportions as a whole we re examined in order to elucidate patterning in the cave assemblages. NI SP and %NISP values were determined for each species per body portion. Body portions were divi ded into the following categories: axial, cranial, distal, upper front limb, lower front limb, upper hind limb, lower hind limb, and distal. Axial elements are all ve rtebrae, ribs, and sternum elements. Cranial elements are the skull, the mandible, and antlers if present. Distal elements refer to the me tapodials, carpals, tarsals, and phalanges. The upper front limb contains the humer us and scapula and the lower front limb is composed of the radius and ulna. The femur and innominate make up the upper hind limb and 52


lower hind limb contains the ti bia, fibula, and patella. Skelet al completeness was measured by comparing the observed number of archaeologi cal specimens to the number expected in a complete skeleton. This equation, d= log X l og Y, was taken from Reitz and Wing (1999:212). D refers to the log ratio, X is the percentage of each skeletal portion id entified to a particular taxon in the archaeological collectio n, and Y is the percentage of this same portion in a complete skeleton. Two skeletal models, ge neral artiodactyla and general car nivore, were adopted for the expected values. This calculations determines whether specific elements/portions or whole bodies were deposited in the various cave assemblages. Sidedness The relative proportion of left and right elemen ts was calculated in terms of NISP values for all sideable elements at the site level. Many of the specimens could not be assigned to a particular side, so they were not included in these calculations. Also, left and right are designated from the perspective of the individual animal. In formation on sided elements at the site level could only be gathered from Stela Cave, Ca ves Branch Rockshelter, and the primary assemblages. The number of left sided skeletal el ements was also examined in terms of both taxa and body portion. Although most of th e comparative reports did not pr ovide site totals for left and right sided elements, all of them had side data for the more a bundant species at each individual site. These two tests will determine wh ether there was a differen tially usage of either left or right sides elem ents in cave deposits. Human modifications Rates of burning and butchering were measured for cave sites for which appropriate data was available. The total number of skeletal specimens (NISP) that showed signs of burning was calculated for each site and presented as a relati ve proportion of all spec imens. All degrees of burning (browned, blackened, calcined) was consider ed in these calculati ons. Signs of butchering 53


evidence taken into consideration included cut marks, hack marks, and sawing. These taphonomic markers reflect the processing of animal carcasses for use. Like burning, the rate of butchering is a simple measure of the relative proportion of the number of skeletal elements exhibiting signs of butchering marks. The qua ntity of marks found on each element was not calculated. Rather, butchering evidence was si mply presented as present or absent. Taphonomy Natural taphonomic factors taken into consid eration included weathering, water damage, animal gnawing, concretion, and element fragme ntation. The presence of each of these taphonomic markers was calculated at the sites where the appropriate data was available. Each marker was calculated by the relative proportion of the number of skeletal elements in which these markers were present. Fragmentation rate s were determined by calculating the %NISP of all whole and all ha lf-complete bones ( 50% complete) per site. The results are compared to the values obtained by Brown and Emery (2008), who adopted a similar strategy for testing fragmentation. Associations/Spatial Locations Associations and spatial locations are two of the four dimensions of artifact variability that Walker (1998, 1999) uses to analyze ritual deposits in his life-his tory approach. This study looks at the spatial location of each cave site in relation to settlements or communities as well as within site spatial organization. Location relative to settlement Most of the documented hunting shrines were located at landscape features outside of town. Using the published excavation reports from each of the cave sites, each of the sites was found to be located within or outside their co mmunity. A site is considered outside of a community if it is not located w ithin the boundaries of or immediat ely adjacent to a surface site. 54


55 The actual physical distance of most of these cav es sites in relation to known surface sites has not been recorded in almost all of the excavation reports. Therefore, what is and is not outside a community has to be loosely defined. Presence of ritual activity areas The presence of maintained cleared areas, ri tual offering areas, and discard areas was tested at each cave sites. Spatially discrete, dense faunal deposits were sought first. This methodology was outlined above in the density sec tion. Next, excavation field reports and other sources were used to examine whether there was any evidence for ritual offering areas, maintained cleared areas, and discard areas in the proximity of dense faunal deposits. The following markers were considered indicative of offering areas: presence of altars, ceramic remains, symbolically potent items, and sacrific ial hearths. Evidence of sweeping, characterized as a sparse but even distribution of small cha rred and other materials across a generally cleared area, and the presence of compacted earth with embedded trampled objects were the two makers considered for maintained cleared areas. Finally discard areas were identified by the presence of sweeping boundaries and middens.


CHAPTER 7 RESULTS This chapter presents the results of the an alyses discussed in the proceeding methods chapter. These results address whether or not th e faunal assemblages of the nine cave sites are reflective of hunting ceremonialism activities and/or whether these caves contain the zooarchaeological markers proposed to be characteristic of Maya ritual deposits. After analyzing density, taxonomy, diversity measures, age cla sses, skeletal complete ness, sidedness, human modifications, taphonomy, location relative to settlement, and location rela tive to other ritual areas it is clear that there is great variability between cave sites. Density Density values were calculat ed as the number of mammalian specimens (NISP) per meter squared for both Cueva de los Quetzales and Agua teca Grieta. Quetzales, representing one large deposit, was estimated to be 10m x 4m in si ze by Brady and Rodas (1995). One thousand seven hundred and sixty mammalian bones were identified in this deposit after sampling. Therefore the density of Quetzales is 44 specimens per meter square (1760 specimens / (10m*4m)). In contrast to Quetzales, Aguateca Grieta represents a numbe r of spatially discrete deposits (i.e. excavation units). The bone density of each excavation unit was also calculated in terms of NISP/meter squared (Table 7.1). Unit 14 from the Two Owls sub-operation had the highest bone density overall with 399 specimens per square meter. Unit 11 from the same sub-operation had the second highest density values with 256 specimens per square meter. Taxonomy All specimens were identifie d to the lowest possible taxonomic level. Some were identifiable only at the level of class, order, family, or genus. All identifications regardless of taxonomic level are presented in tables 7-2 and 7-3. In general, each of thes e sites tend to share a 56


common group of taxa that includes dog, white-tai led deer, opossum, peccary, agouti, and paca. At Cueva de los Quetzales, the most comm on species in terms of both NISP and MNI is Odocoileus virginianus with 185 (10.51%) specimens a nd 14 (25.93%) individuals. Other noteworthy taxa include the paca with 158 speci mens (8.98%) and 8 (14.81%) individuals, the domestic dog with 94 specimens (5.34%) and 7 (12.96%) individuals, and the opossum with 83 specimens (4.72%) and 4 (7.41%) individuals. Although the nine-banded armadillo was the second most abundant taxon in terms of NISP (1 83 specimens, 10.40%), the bulk of the remains were from the dermal skeleton so the MNI of 2 is a more accurate reflection of the armadillo abundance at this site. White-tailed deer is also the most abundant species in the Cueva Rio Murcielagos assemblage with 56 specimens (14.97%) and 4 (23.53%) individuals. Armadillo is the second most abundant (N=39, 10.43%) taxon, but again this due to the larg e number of dermal scutes and a better estimate is the MNI of 1. Peccar y is represented by 14 specimens (3.74%) and 1 individual (5.88%) and dog is represented by 12 specimens (3.16%) and 2 individuals (11.76%). Aguateca Grieta is somewhat different in co mparison to the first two caves. The first interesting note is that white -tailed deer is not the most abundant taxon. According to NISP values, peccary is the most abundant with 182 specimens (8.02%). I believe these values are skewed because these bones were highly fragmented. The MNI of peccary is 3. Other top contributors include agoutis (142 NISP, 11MNI), domestic dog (135 NISP, 9 MNI), and opossum (167 NSIP, 7 MNI). The second interesting thing about Aguateca Grieta is the presence of a number of rare or unique taxa such as jaguar, kinkajou, howler m onkey, and spider monkey. Turning to the comparative cave data, it beco mes apparent that cave sites are highly variable in terms of taxa and their relative abundances. The Naj T unich (Table 7.2) assemblage is 57


the most similar to the Pete xbatn caves in terms of ta xa and relative abundances. Odocoileus virginianus is the most abundant species making up 24.26% of the NISP (N=156) and 15.22% of the MNI (N=7) followed by agouti, peccary, brocket deer, and opossum This is similar to the group of common cave taxa I identified above, except for the relative lack of dog specimens (NISP=5, MNI=1). On the other hand, the Belizean caves sites (Table 7-3) are highly variable in terms of their mammalian content. Actun Polbilche is heavily dominate d by ones species, the domestic dog, contributing 91.69% of the total NISP (N=1103) and 93.94% of the MNI (N=124) At Eduardo Quiroz and Actun Balam, th e total mammalian NISP values are low with 175 specimens at Eduardo Quiroz and 60 at Actun Balam Only three mammalian species were identified at Actun Balam (jaguar, white-tailed deer, and cottontai l rabbit). Cottontail rabbit is the most abundant in both NISP and MNI with 38 identified specimens (63.33%) and at least 4 individuals (66.67%). Both deer and jaguar ha d only one individual identified from the assemblage. Although 20 different taxonomic categorie s were identified in the Eduardo Quiroz assemblage, none of these groups are particul arly abundant. With MN I values of 3 each, squirrels and forest rabbits are the top ranked taxa in terms of MNI. Stela Cave and Caves Branch Rockshelter show a similar distribution of taxa. Overall, mammalian specimens are few in terms of both NISP and MNI values. The mo st abundant specimens are small mammals like armadillo, pocket gophers, squirrels, pacas, and opossums. Dasypus novemcinctus the ninebanded armadillo, is the most abundant taxa at both sites in terms of NISP values with 108 specimens at Caves Branch and 81 specimens at Stela Cave. At Stela Cave there were a fair number of opossum specimens (NISP=13) and this species had the highest MNI value at the site with 3. Although there are not many white-tailed deer specimens at either site, the few present suggest at least 2 individuals were de posited in both Stela and Caves Branch. 58


Diversity Diversity was measured as richness (number of taxa), evenness (distribution of the specimens among taxa), heterogeneity (a combined value representing richness and evenness), and dominance (the extent to which the first tw o ranked species dominate the assemblage. These values are presented in Table 7-4 and 7-5. Aguate ca Grieta was the most rich cave assemblage with a value of 11.54 followed by Naj Tunich, Ca ves Branch Rockshelter, and Eduardo Quiroz which have richness values ranging from 10.1410.60. The diversity values, however, indicate that Caves Branch Rockshelter and Eduardo Quir oz represent the most diverse assemblages. Although both Naj Tunich and Aguateca Grieta have a high number of taxa, they are not equally abundant. Caves Branch and Eduardo Quiroz, on th e other hand, have the lowest evenness values of all the caves sites with .52 and .85. These two site s, therefore, have a large number of taxa that are evenly distributed. This is in sharp contrast to the community statistics calculated for Actun Polbilche. Polbilche had the lowest richness ( 2.36) and diversity (2.5 ) values and highest evenness (2497.2) and dominance (62.0) values of all the cave sites. Age Classes I analyzed age class data for the three Pete xbatn sites and Naj Tuni ch. The Belizean cave publications lacked age class data. Overall, matu re animals (including adults and subadults, both reproductively fertile groups) are more common than immature animals (juveniles and immatures) in this analysis (Figure 7-1). Naj Tunich has the highest percentage of mature animals with 69.57%. Rio Murcielagos is the exception. Immature specimens represent 52.94% of the total MNI at the site. A closer look at the ungrouped age class data gives a more detailed representation of age classes from the Petexabtun sites (Figure 7-2). The %MNI for adults at Quetzales is 27.78%, 25.93% at Aguateca Grieta, and 17.65% at Rio Mu rcielagos. The %MNI for subadults is 5.56% 59


at Quetzales, 16.67% at Aguateca Grieta, and 5.88% at Rio Murcie lagos. Quetzales also has the largest percentage of indivi duals in the category immature+ with 25.93% followed by 11.76% in Murcielagos and 11.11% in the Aguateca Grieta. Immature sp ecimens represent 9.26% at Aguateca Grieta, 5.56% at Quetzales, and 5.88% at Rio Murcielagos. Finally, Rio Murcielagos has the highest juvenile individuals with 47.06% follwed by 11.11% at Aguateca Grieta, and 3.7% at Quetzales. When the data is broken down by taxa for each of the sites (Table 7-6-7.8), it becomes apparent that no particular age class is dominant. There appears to be a fairly even distribution of age classes fo r each taxon with a few possible exceptions. At Cueva de los Quetzales both dog and deer have more mature i ndividuals than immature (Table 7-6). Five of seven identified dogs are mature and 11 of 14 identified deer indi viduals are mature. Proportionally more mature dogs are also found in the Aguateca Grieta as well (N=6 of 9 individuals) (Table 7-8). Elements, Body Portions, and Skeletal Completeness Elements from the axial portion of the body (v ertebra, ribs, and sternum) composed the majority of the specimens at the Petexbatn caves as well as Caves Branch Rockshelter and Stela Cave (Table 7-9). Caves Branch, however, appear s to have proportionately more axial elements overall with 49.07%. The proportion of cranial elements is fair ly similar among all the sites except for Cueva de los Quetzales. The other four sites have a range of about 18-26% cranial elements, but Quetzales only had 9.92% cranial elements. The two Belizean sites, Stela Cave and Caves Branch, appear to have fewer limb bone elements (front and hind) than the Petexbatn sites. Front limb elements are the most co mmon at Quetzales with 16.56% while hind limb elements are most common at Murcielagos with 27.94%. At the taxonomic level it is difficult to discern any kind of ove rall patterning of body portions by site (Figure 7-3 to 7-9). It was difficult to compar e sites by taxa and body portions 60


because many of the sites did not possess the same taxa. Despite these limitations some patterning did emerge among two distinct groups. Group 1 (Quetzales, Rio Murcielagos, Aguateca Grieta, Naj Tunich) and group 2 (Actun Polbilche, Actun Balam, Eduardo Quiroz, Stela Cave, and Caves Branch Rockshelter) tend to follow distinct patte rning in terms of body portions. Group 1 tended to have higher than expect ed values for limb elements across taxa. This is reflected the best in the white-tailed deer (Figure 7-3,7-4) and opossum (Figure 7-7, 7-8) values but can also be seen in the peccary and dog body portion graphs. On the other hand, limb elements were relatively rare in group 2 at both the site and indi vidual species level. The results suggest that whole skeletons we re not deposited at the cave si tes in group 2. Finally, a clearer picture of Quetzales presented itself when body portions were broken down to the taxonomic level. Cranial elements are rare or absent fo r the larger game animal s like deer and peccary (Figure 7-3, 7-5), but are presen t for the smaller mammals like pacas and opossums (Figure 7-7, 7-9). Sidedness Element sides at the site level could only be calculated for the Petexbatn sites and three Belizean sites, Stela Cave, Caves Branch Ro ckshelter, and Eduardo Quiroz. There are no dramatic differences in terms of right vs left element at any of the five sites when the full assemblage is considered (Figure 7-10). At Aguateca Grieta, I identified 328 (49.77%) left specimens and 331 (50.23%) right specimens. Ri o Murcielagos has 103(53.37%) left and 90 (46.63%) right while Quetzales has 238 (53.72%) le ft and 205 (46.28%) right. Stela Cave has 55 (55%) left elements and 45 (45%) right and Caves Branch has 41(47.67%) left and 45 (52.32%) right elements. At Eduardo Quiroz, 19 (41.31%) left elements were recovered and 27 (58.69%) right elements. Therefore, when all elements of all taxa within a single cave are considered, there are relatively equal proportions of right and left elements. 61


Although data for sidedness at the site level wa s not available for every cave, side values for the most common mammalian species was availa ble for each site. The amount of left and right elements of a single taxon with in a single cave is evaluated first. Any species with a sample size over 10 and a value of greater or equal to 65% of either right or left elements is considered to represent a differential usage of sided elements. Under these cr iteria, there is no evidence for the differential use of right or left elements in terms of all elements within a single taxon at any of the cave sites (Tab le 7-10 to 7-17). Taken one step further, patterning was exam ined for the presence of sided elements by body portion for each taxa within each individual cave site. The same criteria of proportions and sample size were used as cited in the paragraph above. Only three sites ha d data that met these criteria, Cueva de los Quetzales, Aguateca Grieta and Naj Tunich. At Quetzales there appears to be a large proportion of left sided elements from the front limb and pelvic region of white-tailed deer specimens (Table 7-10). 72.22% of skeletal specimens from the upper front limb (i.e. humerus) were left (N= 13 of 18) and 68% of sp ecimens from the lower front limb (i.e. radius and ulna) were also left (N= 17 of 25). In a ddition, 66.67% (N= 10 of 15) of white-tailed deer pelvic elements were left as well. It also appears that the preval ence of left sided elements from the upper front limb extended to domestic dog sp ecimens as well. 70.58% (N=12 of 17) were from the left side of the animal. The final pa ttern that emerged at Quetzales was the large proportion of right-sided elements form the upper hind limb (i.e. femur) among paca specimens. Similar patterns emerged at Aguateca Grieta (T able 7-12), which also had proportionately more left-sided elements coming from the front limb region of both white-tailed deer and dog specimens. All white-tailed deer specimens (N=11) from the upper front limb were from the leftside and 66.67% (N=10 of 15) of dog specimens from the lower front limb were also lefts. There 62


also patterning of right-sided elements for both white-tailed d eer and domestic dogs. 63.63% of dog specimens from the distal hind limb and 80% (8 of 10) of deer pelvic elements are rights. Finally, at Naj Tunich (Table 713)we also find proportionately more left-sided elements (N=10 of 13, 76.92%) from the lower front limb of white-tailed deer specimens. Human Modifications Evidence for human modification is limited in the Petexbatn cave sa mples (Table 7-18). Burning rates are low. At both Quetzales and Ag uateca Grieta less than 2% of all identified specimens show signs of burning. Rio Murciela gos has the largest proportion of burned bone (5.79%). The spatial location of the burned bone within the sites was also examined for the Petexbatn caves. Quetzales is one large deposit so all burned bone occurred in one location. At Aguateca Grieta the largest c oncentration of burned bone occurs in Suboperation 31B Unit 14 (N= 9 of 28, 32.14%). This also happens to be the place with the dens est concentration of mammalian remains at the site. At Rio Murcielagos, 10 of 22 burned specimens (45.45%) were recovered from Operation 1, Unit 5 which represen ts the densest concentration of both burned bone and mammalian faunal remains within the cave. Butchering is even rarer. All three sites ha ve less than 1.5% of specimens with butchery marks. Following the general trend, animal gnawing is sparse. Aguateca Grieta has the highest rate of animal gnawing with just 1.68%. The ove rall lack of burning, butchering, and animal gnawing suggests these bones were handled with some care. In comparison with two Belizean caves, I found that burning rates we re higher especially at Stela Ca ve which had a burning rate of 18.23%. Taphonomy In general, taphonomic markers are fairly sp arse throughout the Petexbatn sites (Table 718). In terms of natural taphonomic markers, w eathering, concretion, water damage, and animal 63


gnawing was examined at each site. Rio Murciela gos has the most water-damaged bone with 6.58%. This also corresponds with the concretions which were al so the most common at Rio Murcielagos (5%). Concretion refers to hardened calcium carbonate attached to the remains that have a concrete-like appearance. Weathering ra tes are low with Aguateca Grieta showing the highest amount of exposed bone with 3.23%. Animal gnawing was al so low with all sites having values less than 2%. Weathering, water damage animal gnawing, and concretions were not recorded at Stela Cave and Caves Branch Rockshelter. Fragmentation rates were examined in term s of the proportion of complete or whole specimens and half-complete specimens. I was ab le calculate fragmentation for the Petexbatn sites and two Belizean caves sites (Table 7-19). The Petexbatn assemblages are significantly more complete than the Belizean assemblage s ranging between 43-50% complete or halfcomplete specimens. The Quetzales assembalge is the most "complete" with 49.13% (34.49% whole). On the other hand, the Stela Cave and Caves Branch assemblage s are just under 20%. Location relative to settlement There was a diverse mixture of caves sites in th is analysis that were either located inside the boundaries of a surface site or situated away from major settlement. Cueva de los Quetzales, Rio Murcielagos, and Aguateca Grieta were all lo cated inside their resp ective communities. As previously mentioned, Cueva de los Quetzales is loca ted in the elite, cerem onial core of the site of Las Pacayas. The grieta of Aguateca runs thr ough the center of the site with elite ceremonial and residential architecture fla nking either side. Rio Murcielagos is within the boundaries of Dos Pilas, although not centrally located with el ite ceremonial architectur e like other Petexbatn sites. It was mostly likely accessible to both elite and commoner members of society (Kitty Emery personal communication). Naj Tunich re presents the only Guatemalan cave in this 64


analysis that is not associated with a surface site. Brady (1989 ) argues the cave functioned as a pilgrimage site for the Maya elite. In contrast to the Guatemalan sites, most of caves found within the modern boundaries of Belize appear to be located away from surface sites. Pendergast (1971:7) mentions that no surface site was reported in th e immediate vicinity of Edua rdo Quiroz as of 1971. In his description of Actun Polbilche, Pendergast (1974:7) notes that the only archaeological site known in the area was another cave site called Batty's Cave. Batty's Cave reportedly had a number of associated mound features near the cave entrance, however, Pendergast appears to have been unaware of the caves precise locati on in relation to Actun Polbilche. Known surface sites in the area of Actun Balam, includes both Caracol, Tzimin Kax and Maria Camp, however, all of them are miles away. Tzimin Kax appears to be the closest site which is approximately eight miles north (Pendergast1969:6). Caves Bran ch Rockshelter appears to be the only Belize cave site in this analysis located immediately next to a surface site. There are at least five mounds currently found in front of Caves Br anch Rockshelter (Glassman and Villarejo 2005:286). The site served primarily as a cemetery for the local population. Presence of ritual activity areas As one large deposit, Cueva de los Quetzale s contains the primary characteristic of communal hunting shrines, a large faunal cache. It is located in the heart of the ceremonial/political center. Br ady (Brady 1997, Brady and Rodas 1995 ) argues the deposit was formed as artifacts were thrown down into the cave during cerem onies held on the surface. Thus we have a separation between the performance a nd discard areas. In addition, a round altar is located on the surface next to the skylight entrance into the cave. More than 40% of the ceramics recovered from the surface collection of Quetzal es consisted of painted pottery (Brady and Rodas 1995:3) and over 300 sherds came from polychrome ceramic drums. Other musical 65


instruments recovered from Quetzales include a number of whistl es and a bone rasp. In addition, over 100 sherds came from modeled or applique d incensarios. Although Cueva de los Quetzales contains a number of characteri stics typical of three ritual ar eas defined by Brown and Emery (2008), these characteristics are often overlapping. An altar, cer amics, and symbolically potent items typical of ritual offering areas were found, however, there doesn't appear to be any sacrificial hearths or evidence of burning in general. It is also difficult to assess whether the deposit represents the ritual offering area or the di scard area since it could just as well represent items discarded after ceremonies conducted on the surface. At Aguateca Grieta there are two units that contain high density faunal assemblages or possible "caches". The first two are located in th e area of the Grieta Principal designated as the Two Owls Area (Operation 31B). Unit 11 has 525 identified mammalian specimens and unit 14 has 798, which represent the two de nsest units throughout the Grieta Principal. Ishihara (2007) suggests that lot 11-1-1, which contained 488 identified mammalian specimens, is primarily composed of construction fill or the result of sw eeping of previous ritual activities conducted in the area. The vast majority of this assemblage is composed of unidentified mammal specimens and opossum remains. Lot 11-1-3 is a cultural deposit consisting of the remains of several ritual activities. Ishihara (2007:317) suggests this lot contained an offering hearth where bone was deposited as part of the burning of offering hearths as sacrifices Unit 14 is located to the south of Unit 11 and contains a large number of unide ntified mammalian specimens (539 specimens), peccary (167 specimens), dog (42 specimens) and white-tailed deer (18 specimens). Carbonized material was found in unit 14 yet it was more epheme ral, scattered, and in smaller pieces leading. Ishihara (2007:321) suggests unit 14 is a midden where remains from previous rituals performed in Passage 1 (lot 11-1-3) were swept clean and dumped. 66


In terms of associated artifacts, the Two Owls Operation contained a high frequency of ceramic sherds and symbolica lly important artifacts. Opera tion 31B contained 56.6% of all ceramic sherds found in the Grieta Principa l with 45.1% coming from Unit 14. The ceramics found in Unit 11 primarily consisted of the Sa xche-Palmar Orange Polychrome type which Ishihara (2007:324) argues acted as individual containers for li quid foods like atole. Due to the characteristics of the ceramic assemblage and th e physical size of area, Ishihara suggests smallscale household level ceremonies were held in Passage 1 in which small bowls of good were offered to the gods. The remains of these bowls were then deposited around Unit 14 which was outside the central locus of r itual activity (Ishihara 2007:326). A number of musical instruments were also deposited here such as whistles and flutes (n=6 of 38) as well two-thirds of all drum sherds (Ishihara 2007:328). The excavation information for Rio Murcielagos is fairly brief. In fact, none of the proposed markers for three ritual areas were men tioned in excavation reports from the site except for the presence of ceramics and ritually potent items (Brady et. al. 1991). Operation 1 also happens to be the densest area of the cave with 307 of the 380 identified specimens at the site. Other notable finds include 2 figurines (1 whis tle figurine), 3 perforated animal canines, 3 pendants, jade beads, 10 pyrite disks, and a number of bone and stone tools. None of the remaining six sites have eviden ce of spatially discrete, dense bone deposits, which was the key marker for distinguishing co mmunal shrines used for hunting ceremonialism. Markers suggesting ritual offering areas, maintained clear areas, and discard areas might still be indicative of other rituals though b ecause they are common areas at ot her ritual sites. Most of the sites have evidence of ritual offering areas includ ing hearths, altars, dens e deposits of ceramics 67


and ritually potent items. On the other hand, char acteristics of maintain ed cleared areas and discard areas are not easy to find in the excavation reports. David Pendergast's excavation report of Act un Polbilche gave no indication of ritual activity areas. His work at Eduardo Quiroz, on th e other hand, may suggest the presence of some ritual activity area markers. The primary hu man activity and subsequent excavation was conducted on Chamber 1 of Eduardo Quiroz. The ch amber is split in two by an artificial stone wall. Next to the wall Pendergast found a bed of burned earth and charcoal that included a number of ceramic sherds in and above the ash la yer. This may represents an offering hearth, but Pendergast (1971:13) suggests that it results from some other, un-named human activity. The bulk of the faunal material was found in Chambe r 1, but it was primarily composed of small rodents. Naj Tunich is a large cave system and evidence of potential ritual offering areas is numerous. Altars were found in Operation 3, 4, and 6. Operation 4 lot 53 contained a stone covered with large amounts of ceramic, char coal, and animal bone (Brady 1989:109). Hearths were found associated with structures built insi de the cave like in lot 36 of Operation IV and often contained animal remains. For instance, Operation 6 lot 3H cont ained a large hearth associated with large faunal remains and two mammiform vessel supports (Brady 1989:157). At Stela Cave there are two chambers that have markers of distinct ritual areas. Chamber 1 contains a monolithic square platform roughly 2 m on a side. Ishihara and Griffith (2004:58) suggest two possible functions for the platform as either a perf ormance area or an altar where offerings were placed. Neither suggestions are gi ven more weight nor are offerings said to be associated with the platform. Chamber 3 houses a unique stela-like limesto ne slab that lies adjacent to concentration of carbon, ash, and asso ciated faunal material on the cave floor. 68


Chamber 1 contains a few markers indicative of r itual offering areas or perhaps a performance area while Chamber 3 may represent a ritual discard area. Finally, Caves Branch is a unique situation in that it served primarily as a cemetery. Excavators did find a large amount of ceramic sher ds, faunal remains, chert, and carbon flecks in the general matrix of the rockshelter which su ggests to them that surface offerings were commonly left in the cave. Constant bioturbatio n related to grave digg ing, however, disturbed these offerings spreading them th roughout the matrix (Wrobel 2008:12). 69


Table 7-1. Site Density Values For Aguateca Grieta Operation Unit NISP Size (m2) NISP/m AG31A 9 40 4 10.00 10 4 4 1.00 11 95 3.92 24.23 12 12 1.08 11.11 13 11 3.06 3.59 14 2 0.92 2.17 AG31B 6 28 4.89 5.73 7 48 4.65 10.32 8 14 2.7 5.19 9 5 1.2 4.17 10 35 1 35.00 11 525 2.05 256.10 12 26 1.5 17.33 13 40 0.92 43.48 14 798 2 399.00 AG31C 3 75 5 25 7 1 2.6 0.38 8 57 2.4 23.75 9 1 3.17 0.32 10 28 11 190 1.1 172.73 AG31D 1 11 3 3.67 2 1 1 1.00 3 3 1 3.00 4 10 2.65 3.77 5 13 1.6 8.13 6 4 0.68 5.88 8 2 1.05 1.90 AG31E 1 6 3.05 1.97 70


Table 7-2. NISP/MNI for Cueva de los Quetzale s, Aguateca Grieta, Cueva de Rio Murcielagos, and Naj Tunich Quetzales Aguateca Grieta Rio Murcielagos Naj Tunich Taxa %NISP %MNI %NISP %MNI %NISP %MNI %NISP %MNI Mammalia 5.97 25.91 30.21 Mammalia, very large 0.17 0.62 Mammalia, large 31.31 3.44 13.37 39.04 Mammalia, large/intermediate 18.07 0.27 3.89 Mammalia, intermediate 11.53 9.43 7.75 6.07 Mammalia, intermediate/small 1.67 0.16 Didelphidae 4.72 7.41 0.26 12.96 Didelphis sp. 7.10 0.53 5.88 Didelphis marsupialis 0.04 1.85 0.27 5.88 1.71 8.70 Chironectes minimus 0.09 1.85 Philander opossum Marmosa mexicana Dasypodidae 0.04 Dasypus novemcinctus 10.40 3.70 0.22 1.85 10.43 5.88 0.93 4.35 Alouatta/Ateles 0.47 Alouatta sp. 0.79 1.85 0.31 2.17 Ateles sp. 0.97 1.85 Ateles geoffroyi 0.31 2.17 Carnivora 0.06 0.04 Carnivora, large 0.11 Carnivora, intermediate 0.06 0.04 Carnivora, small 0.27 5.88 Canidae/Felidae 0.23 Canidae 0.28 0.04 Canidae, small Canis sp Canis lupus familiaris 5.34 12.96 5.95 16.67 3.21 11.76 0.78 2.17 Urocyon cinereoargentus 0.06 1.85 0.22 1.85 Procyonidae 0.11 Procyon lotor 0.13 1.85 Nasua narica 0.06 1.85 0.04 1.85 1.40 2.17 Poto flavus 0.04 1.85 0.16 2.17 Mustelidae 0.06 1.85 Spilogale/Conepatus sp Felidae 0.06 Felidae, large 0.23 1.85 0.31 2.17 Felidae, intermediate 0.40 1.85 Felidae, small 0.53 5.88 Leopardus pardalis 0.45 1.85 0.44 1.85 Leopardus weidii Herpailurus yagouroundi 0.16 2.17 Panthera onca 2.91 1.85 Tapirus bairdii 2.02 2.17 Artiodactyla 0.68 0.22 1.07 71


72 Table 7-2. Continued Artiodactyla, large 0.06 0.18 2.67 Artiodactyla, intermediate 0.53 Tayassuidae 0.85 3.70 7.40 5.56 2.67 5.88 Tayassu sp. 0.53 1.07 5.60 8.70 Tayassu tajacu 0.09 0.27 0.31 Cervidae 0.91 3.21 1.40 6.52 Odocoileus virginianus 10.51 25.93 3.17 5.56 14.97 23.53 24.26 15.22 Mazama sp. 3.69 3.70 0.09 1.85 1.07 5.88 2.95 6.52 Rodentia, large Rodentia, intermediate 0.06 Sciuridae 0.35 1.85 Sciurus sp. 0.06 1.85 0.16 2.17 Geomyidae 0.18 1.85 0.53 Orthogeomys sp 0.11 1.85 0.09 Orthogeomys hispidus 0.04 1.85 1.60 17.65 1.09 4.35 Coendu sp. 0.04 1.85 Agoutidae/Dasyproctidae 0.28 0.97 1.60 Dasyprocta puncata 1.31 7.41 6.26 20.37 1.87 5.88 2.33 6.52 Agouti paca 8.98 14.81 2.25 5.56 3.27 15.22 Sylvilagus sp. 0.91 5.56 0.18 3.70 0.16 2.17 Sylvilagus floridanus 0.04 0.16 2.17 Sylvilagus brasiliensis Totals 100.00 100.00 100.00 100 100.00 100.00 100.00 100.00 Total NISP/MNI 1760 54 2269 54 374 17 643 46


Table 7-3. NISP/MNI for Actun Polbilche, Eduardo Quir oz, Actun Balam, Stela Cave, and Caves Branch Actun Polbilche Eduardo Quiroz Actun Balam Stela Cave Caves Branch Taxa %NISP %MNI %NISP %MNI %NIS P %MNI %NISP %MNI %NISP %MNI Mammalia 1.58 14.86 41.52 17.32 Mammalia, very large Mammalia, large 4.57 26.17 22.44 Mammalia, large/intermediate 1.66 12.57 Mammalia, intermediate 15. 43 12.29 37.93 Mammalia, intermediate/small Didelphidae 1.25 0.76 0.57 4.17 1.60 11.11 0.98 Didelphis sp. Didelphis marsupialis 1.71 4.17 Chironectes minimus Philander opossum 0.57 4.17 Marmosa mexicana Dasypodidae 4.17 Dasypus novemcinctus 13.71 4.17 9.95 7.41 13.17 10.00 Alouatta/Ateles Alouatta sp. Ateles sp. Ateles geoffroyi Carnivora Carnivora, large Carnivora, intermediate Carnivora, small Canidae/Felidae Canidae Canidae, small Canis sp. 0.25 3.70 0.37 5.00 Canis lupus familiaris 91.69 93.94 Urocyon cinereoargentus 0.25 3.70 Procyonidae 2.29 4.17 0.12 3.70 0.49 5.00 73


Table 7-3. Continued Procyon lotor 0.12 5.00 Nasua narica 0.57 4.17 0.12 3.70 Poto flavus 0.12 3.70 Mustelidae 8.33 0.12 3.70 Spilogale/Conepatus sp. 1.71 4.17 Felidae 0.25 3.70 Felidae, large 0.50 1.52 Felidae, intermediate Felidae, small Leopardus pardalis Leopardus weidii Herpailurus yagouroundi Panthera onca 10.00 16.67 Tapirus bairdii 2.86 4.17 Artiodactyla 0.86 3.70 0.49 5.00 Artiodactyla, large Artiodactyla, intermediate Tayassuidae 0.25 3.70 Tayassu sp. 0.25 3.70 Tayassu tajacu 0.37 3.70 0.49 5.00 Cervidae 10.86 8.33 0.98 0.00 0.12 5.00 Odocoileus virginianus 6.29 4.17 26.67 16.67 0.61 7.41 1.10 10.00 Mazama sp. 1.41 1.52 2.86 8.33 0.49 3.70 0.12 5.00 Rodentia, large Rodentia, intermediate Sciuridae 2.29 12.50 0.49 7.41 0.12 5.00 Sciurus sp. Geomyidae 1.23 7.41 1.10 10.00 Orthogeomys sp. 0.24 5.00 Orthogeomys hispidus 0.12 5.00 Coendu sp. Agoutidae/Dasyproctidae 0.57 0.12 Dasyprocta puncata 4.17 0.12 5.00 74


75 Table 7-3. Continued Agouti paca 1.66 1.52 4.17 1.11 7.41 3.05 10.00 Sylvilagus sp 0.37 3.70 0.12 5.00 Sylvigalus floridanus 63.33 66.67 0.12 3.70 Sylvilagus brasiliensis 0.25 0.76 5.14 12.50 Totals 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 Total NISP/MNI 1203 132 175 24 60 6 814 27 820 20


Table 7-4. Community Statis tics for Hunting Shrines Hunting Shrine Richness Di versity Evenness Dominance Pa' Ruchi'Abaj 6.6 5.77 39.58 1.57 Pa Sak Man 6.95 4.42 233.48 1.97 Pa Ziguan 3.51 1.62 209.24 10 Table 7-5. Community Sta tistics for Cave Sites Site Richness Diversit y Evenness Dominance Quetzales 8.66 9 12.4 1.75 Rio Murcielagos 8.13 13.6 1.07 1.33 Aguateca Grieta 11.54 11.63 7.97 1.22 Naj Tunich 10.41 14.33 3.9 2 Actun Polbilche 2.36 1.13 2497.2 62 Eduardo Quiroz 10.14 25.01 0.54 1 Actun Balam 6.43 2.5 3 4 Stela Cave 7.97 14.57 0.85 2 Caves Branch 10.6 27.08 0.52 1.5 76


Figure 7-1. Displays the relative proportion of mature vs immature individuals at Cueva de los Quetzales, Aguateca Grieta, Rio Murcielagos, and Naj Tunich Figure 7-2. Age Class data based on elemen t fusion rates for the Petexbatn cave sites 77


Table 7-6. Cueva de los Quetzales Species/Age Class Data (MNI) Taxa Adult Subadult Immature+ Immature Juvenile Unknown Total Didelphidae 1 1 2 4 Dasypus novemcinctus 2 2 Canis lupus familiaris 3 2 2 7 Urocyon cinereoargenteus 1 1 Nasus narica 1 1 Mustelidae 1 1 Felidae, large 1 1 Felidae, intermediate 1 1 Leopardus pardalis 1 1 Tayassuidae 1 1 2 Odocoileus virginianus 2 2 7 3 14 Mazama sp. 1 1 2 Sciurus sp. 1 1 Orthogeomys sp. 1 1 Dasyprocta punctata 1 1 1 1 4 Agouti paca 1 1 1 3 2 8 Syvilagus sp. 2 1 3 Total 15 3 15 2 17 2 54 78


79 Table 7-7. Cueva de Rio Murcielagos Species/Age Class Data (MNI) Taxa Adult Subadult Immature+ I mmature Juvenile Unknown Total Didelphis sp. 1 1 Didelphis marsupialis 1 1 Dasypus novemcinctus 1 1 Carnivora, small 1 1 Canis lupus familiaris 1 1 2 Felidae, small 1 1 Tayassuidae 1 1 Odocoileus virginianus 1 1 2 4 Mazama sp. 1 1 Orthogeomys hispidus 1 2 3 Dasyprocta puncata 1 1 Total 3 1 2 1 8 2 17


Table 7-8. Aguateca Grieta Sp ecies/Age Class Data (MNI) Taxa Adult Subadult Immature+ I mmature Juvenile Unknown Total Didelphidae 1 2 2 1 1 7 Didelphis marsupialis 1 1 Chironectes minimus 1 1 Dasypus novemcinctus 1 1 Alouatta sp. 1 1 Ateles sp. 1 1 Canis lupus familiaris 5 1 1 2 9 Urocyon cinereoargentus 1 1 Procyon lotor 1 1 Nasua narica 1 1 Poto flavus 1 1 Leopardus pardalis 1 1 Panthera onca 1 1 Tayassuidae 1 2 3 Odocoileus virginianus 1 1 1 3 Mazama sp. 1 1 Sciuridae 1 1 Geomyidae 1 1 Orthogeomys hispidus 1 1 Coendu sp. 1 1 Dasyprocta puncata 2 1 1 3 4 11 Agouti paca 1 2 3 Sylvilagus sp. 2 1 Total 14 9 6 5 14 6 54 80


Table 7-9. Body Portion Frequencies (%NISP) Quetzales Rio Murcielagos Aguateca Grieta Caves Branch Stela Cave Proportion %NISP %NISP %NISP %NISP %NISP Axial 46.64 28.21 33.03 49.07 38.04 Pectoral 1.85 0.28 1.38 0.23 2.02 Pelvic 4.54 3.35 6.00 0.23 2.02 Cranial 9.92 17.60 21.94 23.60 25.65 Distal 8.24 7.82 10.01 13.08 13.83 Front limb, distal 2.61 2.23 0.72 0.23 0.58 Front limb, lower 7.31 8.38 4.44 2.80 4.90 Front limb, upper 6.64 4.19 4.02 2.10 2.31 Total front limb 16.56 14.80 9.18 5.13 7.79 hind limb, distal 3.45 5.31 3.66 3.74 4.32 Hind limb, lower 3.45 8.94 6.35 3.27 3.75 Hind limb, upper 5.38 13.69 8.45 1.64 2.59 Total hind limb 12.28 27.94 18.46 8.65 10.66 Total 100.00 100.00 100.00 100.00 100.00 81


Figure 7-3. The ratio of expected to observed body portions for white -tailed deer at Cueva de los Quetzales, Aguateca Grieta, a nd Cueva de Rio Murcielagos. Figure 7-4. The ratio of expected to observed body portions for wh ite-tailed deer at Naj Tunich, Caves Branch Rockshelter, Stela Cave Actun Balam, and Eduardo Quiroz. 82


Figure 7-5. The ratio of exp ected to observed body portions for peccary at Cueva de los Quetzales, Aguateca Grieta, Cueva de Rio Murcielagos, and Naj Tunich Figure 7-6. The ratio of exp ected to observed body portions for peccary at Cueva de los Quetzales and Aguateca Grieta. 83


Figure 7-7. The ratio of expect ed to observed body portions fo r Cueva de los Quetzales and Aguateca Grieta. Figure 7-8. The ratio of expect ed to observed body portions fo r opossum at Naj Tunich, Caves Branch Rockshelter, Stela Cave, and Eduardo Quiroz. 84


Figure 7-9. The ratio of expect ed to observed body portions for paca at Cueva de los Quetzales and Aguateca Grieta. Figure 7-10. The total number (NISP) of left and right sided elements for Cueva de los Quetzales, Cueva de Rio Murcielagos, Aguateca Grieta, Stela Cave, Caves Branch Rockshelter, and Eduardo Quiroz. 85


Table 7-10. NISP values for left-sided elem ents by species at Cueva de los Quetzales Portion opossum white-tailed deer dog Pacas Cranial Axial Pectoral 2 of 2= 100% 1 of 3= 33.33% 1 of 2= 50% 3 of 6= 50% Front Limb, Upper 5 of 8= 62.5% 13 of 18= 72.22% 5 of 10=50% 5 of 8= 62.5% Front Limb, Lower 3 of 6= 50% 17 of 25= 68% 12 of 17=70.58% 7 of 12= 58.33% Front Limb, Distal 4 of 11= 36.36% 2 of 9= 22.22% Pelvic 2 of 5= 40% 10 of 15= 66.67% 6 of 8= 75% 3 of 8= 37.5% Hind Limb, Upper 0 of 4= 0% 6 of 11= 54.54% 1 of 6= 16.67% 4 of 12= 33.33% Hind Limb, Lower 0 of 1= 0% 3 of 3 = 100% 1 of 1= 100% 4 of 8= 50% Hind Limb, Distal 2 of 4= 50% 5 of 14= 35.71% 2 of 9= 22.22% Distal 2 of 4=50% Unknown Total 12 of 26 58 of 94 33 of 67 28 of 63 Total% 46.15 61.70 49.25 44.44 86


87 Table 7-11. NISP values for left-sided elements by species at Rio Murcielagos Portion dog white-tailed deer peccary paca/agouti Cranial 3 of 5= 60% 5 of 6= 83.33% 1 of 2= 50% 1 of 4= 25% Postcranial Axial Pectoral Front Limb, Upper 1 of 2= 50% Front Limb, Lower 3 of 3= 100% 4 of 7= 57.14% 1 of 1= 100% 2 of 3= 66.67% Front Limb, Distal 1 of 2= 50% Pelvic 0 of 1= 0% 1 of 1= 100% Hind Limb, Upper 6 of 9= 66.67% 0 of 2= 0% Hind Limb, Lower 1 of 4= 25% 4 of 8= 50% 0 of 1= 0% 0 of 1=0% Hind Limb, Distal 5 of 11= 45.45% 1 of 4= 25% Distal Unknown Total 7 of 12 26 of 45 3 of 9 4 of 11 Total % 58.33 57.78 33.33 36.36


Table 7-12. NISP values for left-sided elements by species at Aguateca Grieta Portion opossum white-tailed deer d og agouti paca peccary Cranial 9 of 21= 42.86% 1 of 1 11 of 27= 40.74% 19 of 37= 51.35% 7 of 18= 38.89% Postcranial Axial 0 of 0= 0% Pectoral 1 of 1= 100% 3 of 3= 100% 0 of 2= 0% 2 of 3= 66.67% Front Limb, Upper 4 of 6= 66.67% 11 of 11= 100% 5 of 8= 62.5% 7 of 15= 46.67% 0 of 1= 0% 3 of 8= 37.5% Front Limb, Lower 6 of 9= 66.67% 7 of 12=58.33% 10 of 15= 66.67% 4 of 11= 36.36% 1 of 1= 100% 5 of 12= 41.67% Front Limb, Distal 1 of 1= 100% 0 of 1= 0% 5 of 6= 83.33% Pelvic 3 of 9= 33.33% 2 of 10= 20% 4 of 7= 57.14% 5 of 12= 41.67% 0 of 3= 0% 4 of 6= 66.67% Hind Limb, Upper 12 of 22= 54.55% 1 of 5= 20% 8 of 19= 42.11% 10 of 22= 45.45% 1 of 3= 33.33% 5 of 11= 45.45% Hind Limb, Lower 6 of 16= 37.5% 1 of 5= 20% 8 of 18= 42.11% 6 of 12= 50% 3 of 5= 60% 6 of 9= 66.67% Hind Limb, Distal 1 of 2= 50% 5 of 7= 71.43% 4 of 11= 36.36 6 of 8= 75% 2 of 4= 50% 5 of 9= 55.55% Distal 1 of 1= 100% 1 of 3= 33.33% Unknown Total 43 of 87 29 of 55 53 of 109 57 of 119 14 of 35 35 of 54 Total % 49.43 52.73 48 .62 47.90 40.00 64.81 88


Table 7-13. NISP values for left-sided elements by species at Naj Tunich opossum white-tailed deer brocket deer agouti paca peccary Portion L of sided L of sided L of sided L of side L of Sided L of Sided Cranial 3 of 6= 50% 3 of 9= 33.33% 0 of 1= 0% 1 of 3= 33.33% Axial Pectoral 1 of 2= 50% 4 of 7= 57.14% 0 of 2= 0% 2 of 3= 66.67% Front Limb, Upper 1 of 2= 50% 5 of 8= 62.5% 1 of 2= 50% 1 of 3= 33.33% 0 of 1= 0% Front Limb, Lower 10 of 13= 76.92% 2 of 4= 50% 1 of 6= 16.67% Front Limb, Distal 3 of 4= 75% 1 of 2= 50% Pelvic 4 of 6= 66.67% 0 of 1= 0% 1 of 1= 100% Hind Limb, Upper 5 of 13= 38.46% 0 of 1= 0% 0 of 1= 0% 3 of 4= 75% 0 of 1= 0% Hind Limb, Lower 5 of 10= 50% 1 of 3= 33.33% 3 of 3= 100% 0 of 1= 0% Hind Limb, Distal 6 of 11= 54.54% 1 of 1= 100% 1 of 1= 100% Distal 3 of 4= 75% 0 of 2= 0% Unknown Total 5 of 10 48 of 85 4 of 14 6 of 12 6 of 9 3 of 11 Total % 50.00 56.47 28 .57 50.00 66.67 27.27 89


Table 7-14. NISP values for left-sided elements by species at Eduardo Quiroz opossum deer rabbit paca Portion L of sided L of side L of Sided L of sided Cranial 1 of 2= 50% 0 of 4= 0% 1 of 1= 100% Postcranial Axial Pectoral 0 of 1= 0% Front Limb, Upper 3 of 7= 42.86% Front Limb, Lower 0 of 1= 0% 5 of 7= 71.43% Front Limb, Distal Pelvic 1 of 1= 100% Hind Limb, Upper 0 of 1= 0% Hind Limb, Lower 1 of 3= 33.33% Hind Limb, Distal Distal Unknown Total 1 of 3 1 of 9 2 of 2 8 of 14 Total % 33.33 11.11 100.00 57.14 90


91 Table 7-15. NISP values for left-sided elements by species at Actun Balam panther deer rabbit Portion L of sided L of sided L of sided Cranial 1 of 1= 100% -2 of 6= 33.33% Postcranial --Axial -Pectoral -Front Limb, Upper 1 of 1= 100% -Front Limb, Lower --Front Limb, Distal Pelvic -Hind Limb, Upper -Hind Limb, Lower -Hind Limb, Distal -Distal --Unknown Total 1 of 1 1 of 1 2 of 6 Total % 100.00 100.00 33.33


Table 7-16. NISP values for left-sided elements by species at Cave s Branch Rockshelter opossum armadillo white-tailed deer Portion L of sided L of sided L of sided Cranial 1 of 4= 25% 2 of 2= 100% Axial ---Pectoral 0 of 1= 0% Front Limb, Upper 0 of 1= 0% 1 of 1= 100% 0 of 3= 0% Front Limb, Lower 1 of 4= 25% 1 of 2= 50% Front Limb, Distal Pelvic 1 of 1= 100% Hind Limb, Upper Hind Limb, Lower 0 of 3= 0% 1 of 1= 100% Hind Limb, Distal 2 of 3= 66.67% Distal Unknown Total 2 of 7 4 of 11 4 of 8 Total % 28.57 36.36 50.00 92


Table 7-17. NISP values for left-sided elements by species at Stela Cave opossum armadillo dog white-tailed deer brocket deer Portion L of sided L of sided L of sided L of sided L of sided Cranial 4 of 6= 66.67% 0 of 1= 0% 0 of 1= 0% Axial --Pectoral 0 of 1= 0% Front Limb, Upper 1 of 2= 50% 1 of 1= 100% Front Limb, Lower 2 of 3= 66.67% 1 of 1= 100% Front Limb, Distal 0 of 1= 0% Pelvic 1 of 2= 50% Hind Limb, Upper 0 of 1= 0% Hind Limb, Lower 4 of 5= 80% Hind Limb, Distal 1 of 2= 50% 1 of 3= 33.33% 0 of 1= 0% Distal -0 of 1= 0% Unknown Total 4 of 6 9 of 15 1 of 2 1 of 5 1 of 4 Total % 66.67 60.00 50.00 20.00 25.00 93


Table 7-18. Frequency of Taphonomic Markers Quetzales Aguateca Grieta Rio Murcielagos Stela Cave Caves Branch Taphonomic Marker NISP %NISP NISP %NISP NISP %NISP NISP %NISP NISP %NISP Burning 29 1.65 28 1.23 22 5.88 157 19.29 80 9.76 Butchering 11 0.63 26 1.15 3 0.80 3 0.37 0 0.00 Animal Gnawing 26 1.48 39 1.72 2 0.53 25 3.07 1 0.12 Weathering 36 2.01 75 3.31 0 0.00 ----Concreted 3 0.17 35 1.54 19 5.08 ----Water Damaged 0 0.00 12 0.53 25 6.68 ----94


Table 7-19. Fragmentation Analysis Site %Whole % 50% Combined Aguateca Grieta 28.70 18.9 47.6 Quetzales 34.49 14.64 49.13 Rio Murcielagos 20.99 22.74 43.73 Stela Cave 9.41 9.41 18.82 Caves Branch 8.69 10.78 19.47 95


CHAPTER 8 INTERPRETATION This study questions whether nor not faunal assemblages in lowland Maya caves have similar properties, numerical frequencies, associ ations, and spatial dist ributions as those of modern hunting caches and might therefore also be reflective of hunting ceremonial activities. A secondary question concerns whether the lowlan d caves sites are comparable to other Maya ritual cave assemblages. Zooarch aeological characteristics of Maya ritual deposits were proposed by Mary Pohl, James Brady, and Kitty Emery usi ng cave data. This discussion begins with a comparison of the formal propert ies and frequencies of the hun ting caches and cave deposits and follows with an evaluation of th e spatial and associative patter ns between caches, shrines, and settlements and cave faunal assembla ges, artifacts, and settlements. Density A thorough investigation of the density values fo r the majority of the nine caves sites was not possible. Therefore, it is difficult to assess whether or not lowland cave sites also have dense faunal caches similar to the highland hunting shri nes. Nonetheless, spatial information was available for two sites, Cueva de los Quetzales and Aguateca Grieta. Ag uateca Grieta had two dense deposits, units 11 (256 NISP/square meter) and 14 (399 NISP/square meter) of the Two Owls Suboperation (31B). Cueva de los Quetzales contained one conical deposit about 10 meters by 4 meters wide. Superficially, Quetzales appears to match the definition of a large faunal caches, however, it was not very dense (44 NISP/square meter). Taxonomy Birds, reptiles, fish, and mollusks The modern Atitln hunting cache deposits we re almost exclusively composed of mammalian species (Brown and Emery 2008). This was also the pattern for caches at all the 96


shrines identified by Brown in her broad region al surveys (Brown 2005). Therefore, Brown and Emery have both suggested that this is a charact eristic of the material assemblages associated with hunting ceremonialism. The majority of the archaeological cave deposits studied here was also composed of mammalian fauna with the ex ception of Eduardo Quiroz and Actun Balam. Only 4% of the total faunal assemblage at Actun Balam and 20% at Eduardo Quiroz were mammalian specimens. These two sites were heavily dominated by marine and freshwater mollusks. On the other hand, mammalian specimens represented 97% of the remains at Actun Polbilche, 88% at Stela Cave, 81% at Caves Br anch, between 60-70% at the Petexbatn caves, and 56% at Naj Tunich. Non-mammalian remains, therefore, are regularly represented in archaeological cave deposits and must play an im portant role in ritual activities. In comparison with the Atitln caches, the relative abundance of non-mammalian remains represents an obvious disjunct. This may reflect, in part, a difference in environment especially in the case of certain reptilian species. Environmental differences al one, however, cannot expl ain these differences. Current residents situated close to Lake Atit ln definitely rely on the freshwater, aquatic resources that the lake provides. Fish remain s, however, were not de posited at the hunting caches. In fact, fish remains may have been pur posely deposited elsewhere at other shrines specifically designated for fish remains (Linda Brown, personal communication). This suggests that the disposal of non-mammalian remains have changed over time in relation to sacred landscape features. Mammals An emphasis on wild game mammals was s uggested as a characteristic of hunting ceremonialism (Brown and Emery 2008). At each of the three hunting shines the same taxa were found in the same general proportions. White-taile d deer was the top ranked species followed by 97


armadillo, peccary, coati, paca, rabbit, and sma ll agouti. The second highest ranked species included the opossum, tapir, and raccoon (Brown and Emery 2008:318). Within the archaeological cave deposits, similar taxa were found but not in the same proportions. By far, white-tailed deer was the most ubiquitous species in th e archaeological cave sites. It was found at all caves with the excep tion of Actun Polbilche. MNI and NISP values indicate, however, white-tailed deer were the most abundant species at only three sites, Quetzales, Rio Murcielagos, and Naj Tunich. Theref ore, white-tailed deer are not the top ranked species at all cave sites like the hunting caches Armadillo, abundant in highland caches, are relatively infrequent in the cave assemblages. Pe ccary is fairly well-represented at Naj Tunich and Aguateca Grieta, but is larg ely absent from the remaining seven caves. Coati, a high-ranking species in the hunting caches, was extremely rare in the cave assemblages. Naj Tunich had the most identified coati specimens with a NISP of 9, while all other sites had either zero or one specimen. On the other hand, the paca and small a gouti are two of the most ubiquitous species in the cave assemblages. They are found in every site except Actun Balam. Finally, both the opossum and the dog were well represented in th e cave assemblages. Neither of these species were found in great abundance in the highland caches, especial ly the domestic dog. Overall, it appears that the cave assemblages and the hunt ing caches are different in terms of faunal composition. White-tailed deer are undoubtedly centr al to the activities carried in both the highland shrines and the lowland cave sites. On the other hand, dog, opossum, agouti, and paca remains are more common in the lowland caches, while coati and peccary are less common in the assemblages found in the hunting shrines. Emery (2002:110) proposed that an emphasis on symbolic underworld species is a key marker of Maya cave rituals in general. Based on her analysis of the Cueva de los Quetzales 98


fauna, Emery argued for the importance of sacred or ritually important taxa like felines, opossums, dogs, and deer. The taxonomic result s of the nine cave sites support Emery's proposition. Deer, domestic dog, and opossums were identified in many of the cave sites and were often the most abundant species in terms of both NISP and MNI values. Dogs were particularly abundant at Aguat eca Grieta, Actun Polbilche, and Cueva de los Quetzales. For instance at Actun Polbilche, dog remains represented about 92% of the total NISP and 94% of the total MNI. The prevalence of dog remains in cave contexts is not surprising given their suggested ancient role as the "guardian of souls" through the underworld (Tozzer and Allen 1910, Pohl 1983). In addition, opossums were found at every site with the exception of Actun Balam. Although their ritual significance may no t be as well known as other species, it is apparent that the opossums were associated with Maya New Year ceremonies (Pohl 1983:79, Thompson 1970:277 Tozzer 1941:137-141). Emery ( 2002:105) suggests that New Year rituals might have taken place inside cav es. Feline specimens, which were symbolically associated with rulership and the underworld (Saunders 1994), were f ound at six of the eight caves sites, but their numbers were relatively few. Despite the feline specimens, it does appear that the lowland cave sites put an emphasis on symbolic species particularly in regards to deer, domestic dogs, and opossums. Diversity The community statistical measurements of diversity, richness, evenness, and dominance were highly variable among the cave sites. In general, the cave sites had higher richness and diversity values. This indicates that the cave sites tend to have a larger number of species that are more evenly distributed in the assemblage. W ith the exception of Actun Polbilche, evenness values were dramatically higher in the hunting caches. These valu es indicate that hunting caches 99

PAGE 100

are more heavily dominated by a few key species Overall, these measurements do not support a connection between highland hunting caches and lowland archaeological deposits. Age Distributions A high proportion of juvenile i ndividuals as compared to adults was one of the markers that Emery identified as a possibly characterist ic of hunting ceremonialism (Brown and Emery 2008:319). Both Emery (2002, Emery et. al. 2007) a nd Pohl (1983) have al so suggested that ritual deposits are char acterized by a high proportion of juveni le individuals and the Petexbatn cave data as well as the modern cache da ta appears to support this assertion. In the modern cache deposits, the proportion of juvenile individuals as a ratio of all identified mammalian individuals ranged from a bout 2 to 15% (9% mean). White-tailed deer juveniles ranged from 2 to 20% (12% mean) of a ll white-tailed deer indi viduals in particular. Brown and Emery (2008:319) argue that these valu es indicate a higher th an expected proportion of very young individuals compared to modern hunted populations. The proportions of immature specimens from the Petexbatn sites were 25.93% at Aguateca Grieta, 31.48% at Quetzales, and 47.06% at Rio Murcielagos. These values are much higher than those found at the hunting cache sites, indicating a disproportiona te number of juvenile specime ns within the Petexbatn cave fauna. The large proportion of juvenile specimens ma y reflect the ritual importance of these locations. Young individuals appear to be commo n in both elite and ritu al contexts (Carr 1996, Pohl 1983, Wing and Scudder 1991). The importance of juvenile specimens in Maya ritual activities was discussed in both the Popol Vuh a nd the writings of Diego de Landa (Tozzer 1941, Pohl 1983). It is also possible th at a large number of juvenile sp ecimens is the direct result of environmental degradation and overhunting. This explanation, however, is probably unlikely. Although evidence of increas ed environmental toil and overhunting is present in some regions of 100

PAGE 101

the Maya lowlands (e.g. Copan, a nd Peten Lakes region), there is little evidence for it in the Petexbatn (Emery 2004). Elements, Body Portions, and Skeletal Completeness Modern Tzutujil Maya hunters emphasized the importance of returning complete skeletons to the hunting shrines (Brown 2005). Z ooarchaeological analysis of the cache deposits confirmed that most skeletal elements and body portions were found for each taxonomic group (Emery et. al. 2007). The few exceptions include d the absence of antle rs and cranial vault segments for deer, cranial elements for peccary and opossum, and all but the distal elements for cats (Emery et. al. 2007). Brady (1989) also sugg ested that some ritual fauna may have been interred as complete carcasses in cave deposit s and that this was associated with cave ceremonialism. Measurements of skeletal completeness and body portions were variable among the nine cave sites. Total NISP values were calculated in terms of body portion for Cueva de los Quetzales, Rio Murcielagos, Aguateca Grieta, Cave s Branch Rocksheleter, and Stela Cave. The axial body portion, representing the ribs, vertebra, a nd sternum, was the most abundant at each of these sites. Cranial elements were the second-mo st identified body portion at every site with the exception of Cueva de los Quetzales. When the body portion data for the site was examined in tandem with the body portion data fo r specific species at each site, patterning for two distinctive groups emerged. Group 1 (Naj Tunich, Quetzales Rio Murcielagos, and Aguateca Grieta) had most skeletal elements and body portions re presented for each taxonomic group. One exception was mentioned briefly already. Cranial elements we re rare at Cueva de lo s Quetzales, especially among the larger game species like white-tai led deer and peccary. Although the body portion graphs (Figures 7-3 to 7-9) sugge st that both axial and distal elem ents are rare or absent for the majority of species at the Petexbatn sites, th is is largely the result of not being able to 101

PAGE 102

confidently identify most axial and distal elements beyond the level of class. Each of these sites had a large number of axial a nd distal elements from large, intermediate, and small mammals. Group 2 (Actun Polbilche, Actun Balam, Edua rdo Quiroz, Stela Cave, and Caves Branch Rockshelter), on the other hand, tended to be co mposed of axial and cranial elements and few limb bone elements. These results suggest whole bodies or whole skeletons were more frequently deposited at the cave sites associated with group 1 and animal parts or body portions were deposited in group 2. The results on skeletal comp leteness are mixed. Naj Tunich, Quetzales, Rio Murcielagos, and Aguateca Grieta match the pattern found at the hunting caches and Eduardo Quiroz, Actun Balam, Actun Polbilche, Stela Cave, and Caves Branch Rockshelter do not. Status differences in cave use may be responsib le for these two distin ct patterns. The three Petexbatn sites are all located wi thin a site's boundaries and associ ated with elite architecture. Naj Tunich is not associated with an elite si te, but it is believed to have functioned as a pilgrimage center largely for elite members of Ma ya society (Brady 1989). It is more difficult to discern who was using the cave sites in group two. Caves Branch Rockshelter is believed to have been used by all members (infants, youths, adul t males/females) of the small, local non-elite population (Wrobel 2008). Eduardo Quiroz, Actun Balam, Actun Polbilche, and Stela Cave are not associated with any particul ar community nor are they located close to a major Maya site. For instance, Actun Balam is probably the closest to any major site from this group being at least 13 miles to the southeast of Caracol (Penderg ast 1969:6). Differences in the deposition of skeletal body portions may be the result of differential use of bone as a raw material source between social classes. Bone is a valuable raw material source especially for tool manufacture. It is possible that elite members of Maya society could afford to be more generous in their 102

PAGE 103

offerings of skeletal material than non-elite members of society. The non-elite sites lacked limb bone elements that are often used sp ecifically for bone tool manufacture. Conversely, this difference in body part comp leteness might simply be the result of transportation costs. The Maya ha d to travel great distances and over difficult terrain to get to many of these cave sites and trav eling with a partial animal sk eleton is less strenuous. Although the latter explanation is possible, the fact that whole skelet ons were found at Naj Tunich appears to discredit this reasoning. It is also possible that these caves were used for different purposes or merely represent different local traditions of what skeletal elements are deposited at sacred sites. Sidedness Mary Pohl (1983) suggests that ritual assembla ges are characterized by a preference of left sided elements. Emery (2002) also found evidence of the prevalence of left-sided deer elements at Cueva de los Quetzales lending support to Pohl 's claims. The hunting caches did not favor one side or the other since the emphasis was on de positing the complete skeleton (Brown and Emery 2008). In terms of site totals, the number of left elements was not significantly higher than right elements at Quetzales, Rio Murcielagos, Agua teca Grieta, Stela Cave, Caves Branch, and Eduardo Quiroz. At Naj Tunich, Brady (1989) looked at sided elem ents for both avian specimens and white-tailed deer and found no evidence for the differential use of side. Data on the remaining caves sites could not be obtained. Patterning was examined for the differential use of either left or righ t elements at both the species and the body portion level. Th is patterning would suggest wh ether elements of particular species or certain skeletal elements came from on e side more often than the other. The criteria for patterning was a sample size equal to or greate r than 10 and a proportion of either right or left sided elements at 65% or greater. Patterning emer ged at three sites, Quetzales, Aguateca Grieta, and Naj Tunich that appeared to complement one another. A disproportionate amount of left103

PAGE 104

sided elements are from the front limbs of both white-tailed deer specimens and domestic dogs were found. Symbolic and ideological principl es of left and right sides, in regard to the body, has a long history in Maya culture (Palka 2002). Investigations of Maya iconography have demonstrated that figures orientated to thei r left or those who use their le ft hands are often deities or individuals depicted in the Ot herworld (Palka 2002:226). Joel Palka (2002:231) suggests the association of left with ritual and the Otherw orld is a common practi ce among indigenous groups of North America and is particularly evident in Mesoamerica. The patterns that emerged in this analysis suggest that left-sided elements may represent a key ritual mark er, but not one that is found among all species, elements, and ritual contex ts. This might be a marker of particular, as opposed to all, Maya ritual prac tices. It is also intriguing that this pattern was found among front limb elements as opposed to the hind limb. Mary Pohl (1976) had found patterning for left-sided elements of deer hind limb in elite site surf ace contexts. She has also suggested that the hind limb is the preferred form of sacrificial offe ring largely based upon her observation of deer offerings depicted in the Madrid Codex. It is possible that lim b portions, either front or hind, were both considered valu able ritual offerings. Human Modifications A lack of burning and butchering evidence wa s cited as a character istic of both hunting ceremonialism (Brown and Emery 2008:322) and ritual cave assemblages (Brady 1989:378). Signs of burning were found on 24% of the remains at Pa' Ruchi' Abaj' to just over 40% at Pa Sak Man. The value for Pa Sak Man, however, was high because sampling of this site occurred specifically within a burned feature (Kitty Emery, personal communication). In comparison, burning rates were also low for the lowland cave sites. Burning rates were 19.29% at Stela Cave, 9.76% at Caves Branch, 5.88% at Rio Murcielago s, 1.65% at Quetzales, and 1.23% at Aguateca 104

PAGE 105

Grieta respectively. Marks indicative of butchering and skinni ng activities were also not common at the hunting caches. Butchering rates ranged from 5-10% for the two hunting shrines where this modification was recorded (E mery et. al. 2007, Brown and Emery 2008:322). Aguateca Grieta had the highest rate of butchering marks with 1.23%, while Quetzales, Rio Murcielagos, Stela Cave, and Caves Branch were all below 1%. Every site had lower burning and butchering rates than the hunting caches. Al though specific data was not presented for Naj Tunich and Eduardo Quiroz, both Brady (1989:377) and Pendergast (1971:79) mentioned a lack of butchering and burning evidence as well. This suggests that the ancien t Maya took care in the handling of animal bones meant for disposal in cave deposits, a characteristic that supports the hunting cache-lowland cave connection. This pattern of cave disposal is different th an what is presently known about typical Maya garbage handling. Emery and Brown (2008) studi ed animal material disposal patterns among contemporary Itzj Maya. Through interviews they found that disposal by burning was one of the most common animal refuse discard methods (E mery and Brown 2008:6). Animal remains were primarily burned in kitchen fires, on patio areas, and maintained, but forested, areas away from the patio. This suggests that a lack of burning an d butchering evidence is a good marker of ritual activity in cave assemblages. A direct comparis on of the rates of human modifications between cave and surface sites would take this analysis a step further and be a worthwhile study in the future. Taphonomy The remains found at the hunting caches were primarily composed of complete bones. Over 60% of the total hunting cache assemblage was complete elements and an additional 25% were one-half complete. Although the proportion of complete bones was not nearly as high as in the hunting caches, the Petexbatn cave sites had a high percentage of complete or one-half 105

PAGE 106

complete specimens ranging from 44 to 49%. Given the age of these assemblages and the numerous taphonomic processes they must have gone through, it is probably safe to assume that the original deposit was much more intact than even these values suggest. Fragmentation could not be calculated for Naj Tunich, however, Brady (1989:377) wrote that he found an unusually high percentage (39%: 42 of 108) of complete or nearly complete long bones. The cave assemblages from Stela Cave and Caves Branch were more fragmented in comparison. Stela cave had 18.82% and Caves Branch had 19.47% co mplete or one-half complete specimens. At the hunting shrines there was a correlation of weathering and animal alterations with local taphonomic conditions (Emery et. al. 2007). For instance, Pa Sak Man had high weathering rates because it was an exposed site (Emery et. al. 2007:6). It was not the case that bones were left exposed to the elements w ithin the confines of the househol d and then taken to the shrine. Weathering and animal gnawing rates could only be determined for the Petexbatn caves. Weathering rates were less than 5% for each of the caves and animal gnawing was less than 2%. Overall taphonomic markers were low for the Petexbatn caves (Table 6-18 to 6-19). This suggests two things. First, that the ancient Maya at Dos Pilas, Aguateca, and Las Pacayas did not leave animals remains destined for cave disposal laying out on the surface. And second, the enclosed nature of these caves protected the remains from exposure to the elements and disturbance by scavenging animals. These finding s reinforce the idea that animal remains deposited in cave contexts were handled with great care. Location relative to settlement Hunting shrines were located at places in the landscape outside of towns (Brown and Emery 2009:315). The distance between the commun ity and the hunting shrines ranged from 200 m to 2.3 km away. Precise information about the distance between th e lowland caves sites 106

PAGE 107

analyzed and possible surface sites is largely unknown. These measurements were not included in the excavation reports. Therefore, a cave is dete rmined to be outside of their community if it was not located within or direc tly adjacent to a surface site. Us ing this criterion, Quetzales, Aguateca Grieta, Rio Murcielagos, and Caves Branch were all found within a community. Eduardo Quiroz, Naj Tunich, Stel a Cave, Actun Balam, and Actun Polbilche were considered outside of a community because no surface site s were found in the vicinity of these caves. These results present an inte resting conundrum. The sites that have shown the greatest degree of similarity with the hunting caches are the caves that are located within a community boundary like Quetzales, Rio Murcielagos, and Aguateca Grieta. The fact that modern hunting shrines are placed in the wilderness may be a reaction to colonialism and the need to maintain secrecy around traditional ritual activities. Among the ancient Maya the location of a cave in the wilderness or forest may not have been an intrinsic feature of their ritual practice. It is also possible that elite members of Ma ya society appropriated powerful wilderness features to help legitimize their posit ion in society. Presence of ritual activity areas Brown and Emery (2008:316-318) demarcated thr ee activity areas within a hunting shrine. The first was the ritual offering area consisting of sacrificial hearths, alta rs and the bone cache itself. Usually placed in the most ritually important sections of the site, altars can be both natural and constructed features. Altars are often decorate d with organic items such as flowers, copal, pine needles, and food gifts. In addition, they found ceramics, candles, carvings/figurines, and wooden crosses. Many of these offerings were burnt in sacrificial h earths placed near the altar. The second activity area, maintained or cleare d spaces, was also found at hunting shrines and serve as performance and gathering areas. Vegetation in these areas was deliberately cut back 107

PAGE 108

while the floors are ritually swept with a broom fashioned from branches and leaves located in the area. Finally, there are disc ard areas usually composed of sweeping boundaries and middens. The majority of the cave sites did not contain ev idence of all three activity areas typical of hunting shrines. Although most of the sites cont ain a number of markers indicative of ritual offering areas (hearths, ceramics, potent symbolic ite ms), it is rare for a site to also possess the suite of characters indicative of maintained cleared areas and discard areas. This might simply be the result of excavators not looking for markers indicative of ritual areas. Drawing on the work of Brown (2004) Ishihara (2007) was looking specifically for s acrificial offering hearths and ritually swept areas at Aguateca Grieta. Therefore, it may not be surprising that Aguateca Grieta has the most evidence for the three ritual areas. Unit 11 of the Two Owls Operation may represent a ritual offering area containing both ceramic offerings and a sacrificial hearth. Available evidence also sugge sts unit 14 of the Two Owls Operation contained both a maintained performance area and a ritual discard midden. With regards to the presence of activity areas, Aguateca Grieta appears to be the only site that supports a connection with hunting ceremonialism. Locating maintained cleared areas, performance areas, and ritual discard piles, however, is not immediately obvious to most excavators especially those who ar ent looking for it. The fact th at ritual offering hearths are fairly common features of cave de posits suggests that these other two features might be present as well. Future research may be able to tell whether or not this is the case. Summary These results do not present a clear ca se for the proposed connection between contemporary hunting ceremonialism and ancient Ma ya cave rites. There is variation not only between the hunting caches and cave faunal assemb lages but also betwee n the cave assemblages themselves. Overall, the cave sites found in the modern day boundaries of Belize (Stela Cave, 108

PAGE 109

Actun Balam, Caves Branch Rockshelter, Eduardo Quiroz, Actun Polbilche) have very little in common with contemporary hunting shrines. Th ey contained different taxonomic groups, community statistic values, body portions, and taphonomic characte ristics. On other hand, the Petexbatun sites (Cueva de los Quetzales, Agua teca Grieta,Cueva de Ri o Murcielagos) and Naj Tunich shared more in common with the highland hunting caches, but were still very different. They had a similar distribution of age clas ses, body portions, huma n modifications, and taphonomic markers. At the same time, the Pete xabtun sites were unique in that they had a disproportionate amount of left-sided elements from both white-tailed deer specimens and domestic dogs, a different taxonomic composition, and different richness and evenness values. 109

PAGE 110

CHAPTER 9 CONCLUSION Brown and Emery (2008, Emery et al 2007) sugg ested that archaeological faunal deposits found within caves in the Maya Lowlands were possibly the result of ancient hunting ceremonial practices akin to the rites performed t oday by contemporary Maya hunters in highland Guatemala. This study gathered zooarchaeological information from nine cave sites in the Maya Lowlands to test this assert ion. The archaeological cave faunal deposits were compared to the hunting caches that result from contemporary hun ting ceremonialism in terms of the following zooarchaeological categories: density, taxa, diversity, age cl asses, elements, body portions, skeletal completeness, sidedness, human modifications, and ta phonmy. In addition, the association between the cave faunal assemblages and their location relativ e to settlement sites and in relation to possible ritual ac tivity areas were also examined. The results suggest that ther e is not a simple one to one equation between contemporary hunting shrine assemblages and archaeological cave deposits in the Maya Lowlands. There are a number of similarities and differences between these two material phenomena. The Petexbatn caves were the most similar to the contemporar y communal Tzutujil hunting shrines. They were both characterized by high body portion completene ss suggesting the deposition of whole or nearly whole skeletons and high juvenile freque ncies. On the other hand, the Petexbatn sites were different in that they were located insi de their community and had different taxonomic compositions. With the exception of being locate d outside of a community, the caves found in the current geographical boundaries of Belize sh ared very little in common with the contemporary hunting shrines. 110

PAGE 111

This study also examined a much larger que stion of whether or not these cave faunal deposits represent pan-Maya cave ritual behavior s as has previously been suggested by Mary Pohl (1983), James Brady (1989) and Kitty Emery (2002). Each of three studies found support in this analysis. The presen ce of species symbolically associated with the underworld as suggested by Kitty Emery were found in most of the cave sites. In the case of Pohl's work, there was evidence for an increased number of juvenile specimens and left-sided elements at the Petexbatn sites in particular. On the other hand, Brady's asse rtions on low burning/butchering rates and the prevalence of high skeletal co mpleteness were also found. Yet none of these zooarchaeological markers were found at every single cave site or within every taxonomic group and body portion. The results of this study suggest that the ritual use of animal s in cave contexts was highly variable throughout both time and space in the Maya region. There is not a set of uniform panMaya ritual practices that archaeologists and/or zooarchaeologists can use to discern ritual contexts. This study suggests that Maya zooarch aeology in particular should take a different approach to investigating ritual. The search of zooarchaeological co rrelates of generalized ritual behavior fails to take into account the immense va riability in ancient Maya ritual practices that involved animals. Zooarchaeologist s should take a more in-depth l ook at both the similarities and differences that exist in the material correlates of Maya ritual animal use rather than search for generalizations. Although there appears to be a long-standing tradition in Maya culture of depositing animal remains in caves and associated landscape features, the exact nuances of this practice were likely different over time, between va rious groups of Mayan-speaking peoples, and between particular classes of society. Although th is study failed to prove the connection between 111

PAGE 112

contemporary hunting ceremonialism and ancient Ma ya cave rites, it remains entirely possible that these two practices are related in some fashion. Archaeologists of ten study ritual because they find comfort in the idea that rituals are routine actions th at dont change. Rituals tend to present themselves as static, time-honored customs, but in reality they ar e always in the process of changing. Even if contemporary hunting cere monialism and ancient Maya cave rites are indeed related, it is not a given th at either the practices and/or th e beliefs informing the practice have remained unchanged. Just as we cannot assume the continuity of ritual practice over time, we also cannot assume the same continuity in practice throughout the Maya region within a particular time period. The variability between nine cave sites in this study attests to this fact. To assume widespread continuity ignores the agen cy of people within Maya society. Although there was some sort of structure informing the practice of depositing animal remains in cave contexts, the exact nuances of this pract ice undoubtedly changed as Maya peoples attempted to reproduce, reinterpret, and transform thes e practices for their own ends. A more fine-grained, as opposed to broad-stroke, approach to investigating Maya cave ritual deposits would be more beneficial to future zooarcha eological studies. One possible avenue is to compare cave faunal deposits with faunal deposits found within associated surface sites. For instance, the caves of Dos Pilas could be analyzed in comparison with the midden assemblages found from elite and commoner reside ntial areas and from ceremonial centers at Dos Pilas. An analysis like this might help archaeologists understand how a particular Maya community, or segments therein, engaged in what Bell (1992, 1997) calls ritualization. Ritualization is a way of acting that is designed and orchestrated to distinguish and privilege what is being done in comparison to ever y-day, quotidian activities (Bell 1997:81-82). Ritualization is often used to distinguish certa in members of a community as being different 112

PAGE 113

from other members of the same community (e.g. elite vs. commoners). By comparing the faunal assemblages from a number of different contex ts within a particular site, a more thorough understanding of ritual can emerge. In combin ation with Walkers life-history approach, archaeologists can understand how animals remains destined for deposition in cave contexts are treated differently than those deposited elsewhere. As Bell (1992:91) point outs, the relative clarity and flexib ility of the boundaries that distinguish ritualized and non-ritualized behavior are highl y strategic and particular to a community. These differences, therefore, are best understood in a we ll-defined, concrete situation such as within a partic ular Maya site or community as opposed to a regional analysis. The following is a brief example of what this ki nd of approach might look like. Returning to the Petexbatn cave zooarchaeological data, it is apparent that th ese deposits have unique formal properties, numerical frequencie s, and associations. One striki ng characteristic of the faunal remains recovered from the Petexbatn caves site s, was the lack of taphonomic markers. It appears that animal remains destined for cave dis posal were handled with great care because they were relatively unburned and showed little si gns of taphonomic markers like animal gnawing, exposure to the elements, and fragmentation. It is possible that this care is unique to animal remains found in cave contexts as opposed to th ose remains found in residential middens, in fill material for buildings, or from other ritual cont exts like burials and caches within the same site. If this is the case, then the life history of fauna l remains destined for cave disposal is unique not only in the stage of discard, but also during their stage of use. The emphasis on maintaining the relatively pristine condition of thes e remains could be an example of ritualization. This practice might be a way of acting that distinguishes and privileges animals remains destined for ritual discard as opposed to animal remains discarde d during every-day quotidia n activities. An in113

PAGE 114

114 depth, site-specific analysis of all the different life histories of animal remains found within different contexts within a particular site w ould go a long way in helpin g both archaeologists and zooarchaeologists understand anci ent Maya ritual practices.

PAGE 115

LIST OF REFERENCES Alcorn, J. B. 1984 Huastec Mayan Ethnobotany. Univers ity of Texas Press, Austin. Bell, C. 1992 Ritual Theory, Ritual Practice. Oxford University Press, New York. 1997 Ritual: Perspectives and Dimensions. Oxford University Press, New York. 2007 Response: Defining the Need for a De finition. In The Archaeology of Ritual, edited by E. Kyriakidis, pp. 277-288. Cotsen In stitute of Archaeology, University of California, Los Angeles. Bonor, J. L. 1989 Las cuevas mayas: Simbolismo y ritual Universidad Complutense de Madrid. Bourdieu, P. 1977 Outline of a Theory of Practice. Ca mbridge University Press, Cambridge. 1990 The Logic of Practice. Stanford University Press, Stanford. Brady, J. E. 1989 An investigation of Maya ritual cave use with special reference to Naj Tunich, Peten, Guatemala, University of California. 1997 Settlement Configuration and Cosmology: The Role of Caves at Dos Pilas. American Anthropologist 99(3):602-618. Brady, J. E. and K. M. Prufer (editors) 2005a In the Maw of the Earth Monster. University of Texas Press, Austin. 2005b Introduction: A History of Mesamerican Ca ve Interpretation. In In the Maw of the Earth Monster, pp. 1-18. University of Texas Press, Austin. 2005c Introduction: Religion and the Role of Caves in Lowland Maya Archaeology. In Stone Houses and Earth Lords: Maya Religi on In the Cave Context, edited by K. M. Prufer and J. E. Brady. University Press of Colorado, Boulder. Brady, J. E. and I. Rodas 1995 Maya Ritual Cave Deposits: Recent Insi ghts from the Cueva de Los Quetzales. Institute of Maya Studies Journal 1(1):17-25. Brady, J. E., I. Rodas, L. Wright, K. Em ery, N. Lopez, L. Stiver and R. Chatham 1991 Proyecto Arqueolgico Regional de Cuevas Petexbatn. In Proyecto Arqueolgico Regional Petexbatn Inform e Preliminar #3, Tercera Temporada 1991, edited by A. A. Demarest, T. Inomata, H. Escobedo and J. Palka, pp. 652-748. Vanderbilt University Press, Nashville, TN. Brown, L. A. 2004 Dangerous Places and Wild Spaces: Creat ing Meaning With Materials and Space at Contemporary Maya Shrines on El Du ende Mountain. Journal of Archaeological Method and Theory 11(1):31-58. 115

PAGE 116

2005 Planting the bones: hunting ceremonialis m at contemporary and nineteenth century shrines in the Guatemalan highla nds. Latin American Antiquity 16(2):131-146. 2006 Planting the Bones: An Ethnoarchaeologi cal Exploration of Hunting Shrines and Deposits. Foundation For The Advancement of Mesoamerican Studies, Inc. Brown, L. A. and K. F. Emery 2008 Negotiations with the Animate Forest: H unting Shrines and Ho uses in the Maya Highlands. Journal of Archaeological Method and Theory 15:300-337. Brown, L. A. and W. H. Walker 2008 Prologue: Archaeology, Animism, and Non-Human Agents. Journal of Archaeological Method a nd Theory 15(4):297-299. Bruck, J. 1999 Ritual and Rationality: Some Problems of Interpretation In European Archaeology. European Journal of Archaeology 2(3):313-344. Carlsen, R. 1997 The War for the Heart and Soul of a Hi ghland Maya Town. University of Texas Press, Austin. Carlsen, R. and M. Prechtel 1991 The Flowering of the Dead: An Interp retation of Highland Maya Culture. Man 26:23-42. Carr, H. S. 1996 Pre-Columbian Maya Exploitation and Management of Deer Populations. In The Managed Mosaic: Ancient Maya Agricultur e and Resource Management, edited by S. L. Fedick, pp. 251-261. University of Utah Press, Salt Lake City. Christenson, A. J. 2001 In the Mouth of the Jaguar: Caves a nd Maya Cofrada Worship in Highland Guatemala. The PARI Journal 2(2):1-9. Colas, P. R. 2006 The Hunting Scenes in the Madrid Code x: A Planner for Hunting Rituals. In Sacred Books, Sacred Languages: Two Thousa nd Years of Ritual and Religious Maya Literature, edited by G. V. Rivera and G. L. Fort, pp. 81-92. 10 ed. Acta Mesoamerican. Markt Schawaben: Verlag Anton Saurwein. Cortes y Larraz, P. 1958 Descripcin Geogrfico-Moral de la Di cesis de Goathemala, Tomo 2. Tipografa Nacional, Guatemala City. 116

PAGE 117

Dietler, M. 1996 Feasts and commensal politics in the polit ical economy: food, pow er and status in prehistoric Europe. In Food and the Status Quest: An Interdisciplinary Perspective, edited by P. Wiessner and W. Schiefenhove l, pp. 86-126. Berghahn Books, Providence, RI. Dobres, M.-A. and J. Robb 2000 Agency in Archaeology. Routledge, London. Drennen, R. D. 1976 Religion and social evolution in formative mesoamerica. In The Early Mesoamerican Village, edited by K. V. Fl annery. vol. 345-368. Academic Press, New York. Emery, K. 2002 Animals from the Maya Underworld: R econstructing Elite Ma ya Ritual at the Cueva de los Quetzales, Guatmala. In Beha vior Behind Bones: Th e zooarchaeology of religion, ritual, status, and id entity, edited by S. J. O'Da y, W. Van Neer and A. Ervynk, pp. 101-113. Oxbow Books, Oxford. 2004 Environments of the Maya Collapse: Zooarchaeological Pe rspective from the Petebatn. In Maya Zooarchaeology: New Dire ctions in Method and Theory, edited by K. Emery. Cotsen Institute of Archaeology, University of Califor nia, Los Angeles. Emery, K. and L. A. Brown 2008 Ethnoarchaeological Studies of Animal Mate rial Disposal Patterns in the Southern Maya Lowlands and Implications for Maya Zooarchaeology. Paper presented at the Society for American Archaeol ogical meetings, Vancouver BC. Emery, K. F., L. A. Brown, E. M. A nderson, E. K. Thornton and M. LeFebvre 2007 Ethno-Zoology of Modern Maya Ritual An imal Bone Caches and Implications for Interpretations of Ancient Maya Diet and R itual. Paper presented at the University of Calgary Chac Mool Conference, Calgary. Flannery, K. V. 1976 The Early Mesoamerican Village. Academic Press, New York. Gell, A. 1998 Art and Agency: An Anthropological Theory. Clarendon Press, Oxford. Giddens, A. 1979 Central Problems In Social Theory: Acti on, Structure, and Contradiction in Social Analysis. University of California Press, Berkeley 1984 The Constitution of Society. Univers ity of California Press, Berkeley 117

PAGE 118

Glassman, D. M. and J. L. Bonor Villarejo 2005 Mortuary Practices of th e Prehistoric Maya From Caves Branch Rockshelter, Belize. In Stone Houses and Earth Lords: Ma ya Religion in the Cave Context, edited by K. M. Prufer and J. E. Brady, pp. 285-296. University of Colorado Press, Boulder. Goody, J. 1982 Cooking, Cuisine and Class: A Study of Comparative Sociology. Cambridge University Press, Cambridge. Gordon, G. 1898 Caverns of Copan, Honduras: Report on Explorations by the Museum, 1896-97, Memoirs of the Peabody Mu seum of American Archaeo logy and Ethnology, Vol. 1, No. 5. Harvard University. Grayson, D. K. 1978 Minimum Numbers and Sample Size in Vertebrate Faunal Analysis. American Antiquity 43(1):53-65. 1984 Quantitative zooarchaeology: topics in the analysis of archaeological faunas. Academic Press, Orlando. Grayson, D. K. and C. J. Frey 2004 Measuring Skeletal Part Representati on in Archaeological Faunas. Journal of Taphonomy 2(1):27-42. Hayden, B. 1995 Pathways to power: principles for cr eating socioeconomic inequalities. In Foundations of Social Inequality, edited by T. D. Price and G. M. Feinman, pp. 15-85. Plenum Press, New York. 1996 Feasting in prehistoric and traditional societies. Food and the Status Quest. Berghahn Books, Oxford. Heyden, D. 1975 An Interpretation of the Cave underneath the Pyramid of the Sun in Teotihuacan, Mexico. American Antiquity 40(2):131-147. 1981 Caves, Gods, and Myths: World-Vi ew and Planning in Teotihuacan. In Mesoamerican Sites and World-Views, ed ited by E. P. Benson, pp. 1-39. Dumbarton Oaks, Washington D.C. 2005 Rites of Passage and Other Ceremonies in Caves. In the Maw of the Earth Monster, edited by J. E. Brady and K. M. Prufer, pp. 21-34. University of Texas Press, Austin. Ingold, T. 2000 The Perception Of The Environment: E ssays in livelihood, dwelling and skill. Routledge, London. 118

PAGE 119

Ishihara, R. 2007 Bridging the Chasm Between Religion and Politics: Archaeological Investigations of the Grietas at the Late Classic Maya Site of Aguateca, Peten, Guatemala, University of California. 2008 Rising clouds, blowing winds: Late Classic Maya rain ri tuals in the Main Chasm, Aguateca, Guatemala. World Archaeology 40(2):169-189. Ishihara, R. and C. S. Griffith 2004 Construction of Sacred Space in St ela Cave, Cayo District, Belize. Paper presented at the 69th Annual Meeting of the Society for American Archaeology, Montreal, Canada. Kavountzis, E. G. 2009 Evaluating cave use through spatial analys is of animal remains from Maya caves in Guatemala and Belize. Paper presented at the 74th Annual Meeting of the Society for American Archaeology, Atlanta. Kyriakidis, E. 2007a Archaeologies of Ritual. In The Arch aeology of Ritual, edited by E. Kyriakidis, pp. 289-308. Cotsen Institute of Archaeology, Un iversity of Califor nia, Los Angeles. 2007b The Archaeology of Ritual. Cotsen Adva nced Seminar 3. Cotsen Institute of Archaeology, University of California, Los Angeles. 2007c Finding Ritual: Calibrating the Evidence. In The Archaeology of Ritual, edited by E. Kyriakidis, pp. 9-22. Cotsen Institute of Archaeology, University of California, Los Angeles. 2007d In Search of Ritual. In The Archaeology of Ritual, edited by E. Kyriakidis, pp. 18. Cotsen Institute of Archaeology, Univ ersity of California, Los Angeles. Latour, B. 1993 We Have Never Been Modern. Harv ard University Press, Cambridge. MacLeod, B. and D. E. Puleston 1978 Pathways into Darkness: The Search for the Road to Xibalba. In Tercera Mesa Redonda de Palenque, edited by M. G. Roberston and D. C. Jeffers, pp. 71-77. vol. 4. Hearld Peters, Monterey, CA. Marcus, J. 2007 Rethinking Ritual. In The Archaeology of Ritual, edited by E. Kyriakidis, pp. 4376. Costen Institute of Archaeology, Univ ersity of California, Los Angeles. Mennell, S. 1996 All Manners of Food: Eating and Taste in England and France from the Middle Ages to the Present. Universi ty of Illinois Press, Urbana. Mercer, H. C. 1975 The Hill-Caves of Yucatan. Univer sity of Oklahoma Press, Norman. 119

PAGE 120

Mills, B. J. and T. J. Ferguson 2008 Animate Objects: Shell Trumpets and Ritu al in the Greater Southwest. Journal of Archaeological Method a nd Theory 15(4):338-361. Odum, E. P. 1971 Fundamentals of Ecology. W.B. Saunders Co., Philadelphia. Palka, J. 2002 Left/Right Symbolism and the Body in Ancient Maya Iconography and Culture. Latin American Antiquity 13(4):419-443. Pauketat, T. R. 2001 Practice and history in archaeology. Anthropological Theory 1(1):73-98. Pendergast, D. M. 1969 The Prehistory of Actun Balam, British Honduras. Royal Ontario Museum Occasional Paper 16. The Royal Ontario Mseum, Ontario. 1971 Excavations at Eduardo Quiroz Cave, British Honduras (Belize). Royal Ontario Museum Art and Archaeology Occasional Paper 21. The Royal Ontario Museum, Ontario. 1974 Excavations at Actun Polbiche, Be lize. Archaeological Monographs 1. Peterson, P. A. 2006 Ancient Maya Ritual Cave Use in the Sibun River Vally, Belize, Boston University. Pohl, M. 1976 Ethnozoology of the Maya: An Analysis of Fauna from Five Sites in Peten, Guatemala, Harvard University. 1981 Ritual Continuity and Transformati on in Mesoamerica: Reconstructing the Ancient Maya Cuch Ritual. Am erican Antiquity 48(3):513-529. 1983 Maya Ritual Faunas: Veterbrate Rema ins from Burials, Caches, Caves, and Cenotes in the Maya Lowlands In Civilization in the Ancient Americas, edited by R. Leventhal and A. Kolata, pp. 55-103. Univer sity of New Mexico Press, Albuquerque. Pohl, M. and J. Pohl 1983 Ancient Maya Cave Rituals. Archaeology 36:28-32. Prufer, K. M. and J. E. Brady (editors) 2005 Stone Houses and Earth Lords: Maya Re ligion In the Cave Context. University Press of Colorado, Boulder. Redfield, R. and A. Villa Rojas 1934 Chan Kom: A Maya Village, University of Chicago Press. Reitz, Elizabeth. and E. Wing 1999 Zooarchaeology. Cambridge University Press, Cambridge. 120

PAGE 121

Saunders, N. J. 1994 Tezcatlipoca: jaguar metaphors and the Aztec mirror of nature. In Signifying Animals: Human Meaning in the Natural World, edited by R. Willis, pp. 159-177. Routledge, London. Schiffer, M. B. 1972 Archaeological context and systemic context. American Antiquity 37:156-165. Simpson, E. H. 1949 Measurement of Diversity. Nature 163:688. Stahl, A. B. 2008 Dogs, Pythons, Pots, and Beads: The Dynamics of Shrines and Sacrificial Practices of Banda, Ghana, 1400-1900 CE. In Memory Work: Archaeologies of Material Practices, edited by B. J. M ills and W. H. Walker, pp. 159-186. School for Advanced Research Press, Santa Fe, New Mexico. Stone, A. 1995 Images from the Underworld: Naj Tuni ch and the Tradition of Maya Cave Painting. University of Texas Press, Austin. Taube, K. 1997 A God Named Zip. Archaeology 50(3):39. 2003 Ancient and Contemporary Maya Concep tions about Forest and Field. In The Lowland Maya Area: Three Millennia at th e Human-Wildland Interface, edited by A. Gomez-Pompa, M. F. Allen, S. L. Fedick and J. J. Jimenez-Osornio, pp. 461-492. Food Products Press, New York. Thompson, E. H. 1897 Cave of Loltun, Yucatan: Report of Explorations by the Museum, 1888-89 and 1890-91. Memoirs of the Peabody Museum of American Archaeology and Ethnology, Vol. 1, No. 2. Harvard University. Thompson, J. E. 1970 Maya History and Religion. University of Oklahoma Press, Norman. 1975 Introduction. In The Hill-Caves of Yucatan: A Search for Evidence of Man's Antiquity in the Caverns of Central Ameri ca, edited by H. C. Mercer, pp. vii-xliv. University of Oklahoma Press, Norman. Tozzer, A. 1941 Landa's Relacion de las Cosas de Yucata n. Harvard University Press, Cambridge. Tozzer, A. and G. M. Allen 1910 Animal Figures in the Maya Codices. Papers of the Peabody Museum Vol. 4 No.3. Harvard University, Cambridge. 121

PAGE 122

122 Tylor, E. B. 1958 Primitive culture. Harper and Row, New York. Vail, G. 1997 The Deer-Trapping Almanacs in the Madrid Codex. In Papers on the Madrid Codex, edited by V. R. Bricker and G. Vail. Middle American Research Institute, Tulane University, New Orleans. Vogt, E. Z. 1976 Tortilla for the Gods: A Symbolic Analysis of Zinacanteco Rituals. Harvard University Press, Cambridge. Von Nagy, C. 1997 Some Comments on the Madrid Deer Hunting Almanacs. In Papers on the Madrid Codex, edited by V. R. Bricker and G. Va il, pp. 27-71. Middle American Research Institute, Tulane University, New Orleans. Walker, W. H. 1995 Ceremonial Trash? In Expanding Arch aeology, edited by J. M. Skibo, W. H. Walker and A. E. Nielson, pp. 67-79. University of Arizona, Tucson. 1998 Where are the Witches of Prehistory? Journal of Archaeological Method and Theory 5(3):245-308. 1999 Ritual, life histories, a nd the afterlives of people and things. Journal of the Southwest 41:383-405. 2008 Practice and Nonhuman Social Actors: Th e Afterlife Histories of Witches and Dogs in the American Southwest. In Me mory Work: Archaeologies of Material Practices, edited by B. J. Mills and W. H. Walker, pp. 137-158. School for Advanced Research Press, Santa Fe, New Mexico. Wing, E. and S. J. Scudder 1991 The exploitation of animals. In Cuello : An Early Maya Community, edited by N. Hammond, pp. 84-97. Cambridge University Press, Cambridge. Wrobel, G. 2008 Temporal Changes in the Mortuary Ritu al Use of the Caves Branch Rockshelte, Belize. Foundation for the Advancement of Mesoamerican Studies, Inc. Grantee Report. Zedeno, M. N. 2008 Bundled worlds: The roles and interacti ons of complex societies from the North American plains. Journal of Archaeol ogical Method and Theory 15(4):362-378.

PAGE 123

BIOGRAPHICAL SKETCH Elyse Anderson is a graduate student at the University of Florida studying Maya zooarchaeology under the direction of Dr. K itty Emery. She has conducted archaeological research in Guatemala, Peru, and France. A Wa shington state native, she graduated with her Bachelor of Arts from the Un iversity of Wa shington in 2005. 123