University Press of Florida

Research Reports in Belizean Archaeology Volume 4

MISSING IMAGE

Material Information

Title:
Research Reports in Belizean Archaeology Volume 4
Physical Description:
Book
Language:
en-US
Creator:
Awe, Jamie - Director, Institute of Archaeology, Belmopan, Belize

Subjects

Subjects / Keywords:
Archaeology, Belize, Maya, Anthropology
Archaeology, Maya (People), Anthropology
Humanities

Notes

Abstract:
Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2007 Belize Archaeology Symposium. Divided into two sections, this volume features papers on the Terminal Classic period as well as general research reports from the 2006-2007 field season.
General Note:
Expositive
General Note:
Community College, Higher Education
General Note:
Adobe pdf reader
General Note:
Kara Schwartz kara
General Note:
Narrative text
General Note:
kara@upf.com
General Note:
http://florida.theorangegrove.org/og/file/55d8011d-42ca-4e52-a940-80be2f5aaf7b/1/RRBA%20Vol%204%20cc.pdf

Record Information

Source Institution:
University of Florida
Rights Management:
This work is licensed under a modified Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/3.0/. You are free to electronically copy, distribute, and transmit this work if you attribute authorship. You must attribute the work in the manner …
Resource Identifier:
oclc - 434569959
System ID:
AA00011735:00001


This item is only available as the following downloads:


Full Text

PAGE 2

Research Reports in Belizean Archaeology Volume 4 Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2007 Belize Archaeology Symposium Edited by John Morris, Sherilyne Jones, Jaime Awe and Chri stophe Helmke Institute of Archaeology Nati onal Institute of Culture and History Belmopan, Belize 2007

PAGE 3

Editorial Bo ard of the Institute of Archaeology, NICH John Morris, Sherilyne Jones, Jaime Awe and Christophe Helmke The Institute of Archaeology, Belmopan, Belize Jaime Awe, Director John Morris, Associate Director, Research and Education Brian Woodye, Associate Director, Parks Management George Thompson, Associate Director Planning & Policy Management Sherilyne Jones, Research and Education Officer Cover design: Christophe Helmke Frontispiece: Terminal Classic Sahcaba Molded -carved vase from Pooks Hill (Photograph by Christophe Helmke). Back cover: Late Classic Pa nel 2 from Xunantunich date d paleographically to AD 780-820 (Photograph by Christophe Helmke). Layout and Graphic Design: Sherily ne Jones, Christophe Helmke. ISBN 976-95165-3-8 Copyright 2007 Institute of Archaeology, National Institute of Culture and History, Belize. All rights reserved. Printed by Print Belize Limited. ii

PAGE 4

J. Morris et al. iii

PAGE 5

ACKNOWLEDGEMENTS We wish to express our sincerest thanks to every individual who contributed to the success of our fourth symposium, and to the subsequent publication of th e scientific contributions that are contained in the fourth volume of the Research Reports in Belizean Archaeology. A special thanks to Print Belize and the staff for their efforts to have the Symposium Volu me printed on time despite receiving the documents on very short notice. We extend a special thank you to all our sponsors. To the staff and management of our sister Institution, within NICH, Institute of Creative Arts (ICA), we thank you for your assistance and for graciously hosting the 2006 symposium. We are grateful to the Protected Areas Conservation Trust, Belize Tourism Board, Esso Standard Oil S.A. Ltd, Print Belize Ltd., Old Belize Cultural & Historical Center, Belize Electric Company Limited (BECOL), Galen University, Fort Street Tourism Village, Belize Communication and Security Ltd., and various other in stitutions that provided financial and logistical assistance. We are especially grateful to the Honorable Franci s Fonseca, Attorney General, Minister of Labor, Education and Culture whose dedication to educati on and continuous support c onstantly encourages us. Thanks to the Honorable Mark Espat, the previous Minister of Culture for his tremendous support of the Institute of Archaeology. A special th anks Mr. Yasser Musa, President of the National Institute of Culture and History for his constant encouragement to impr ove the professional capacity of our institution and for ensuring that the volume gets published. Thanks also to Mr. Victor Espat, Administrator, and the Board of Directors of NICH who provided considerable funding to ensure the success of our symposium and the subsequent publication of this volume. We extend a special thank to the staff and management of the Bliss Center of Creative Arts and the Institute of Creative Arts (ICA), thank you for your assistance and for graciously hosting the 2006 symposium. We are grateful to the Institute of Social and Cultural Research (ISCR), our sister institution in NICH, for providing our ISBN number. Meetings of this scope are never possible without professional participation. We therefore thank all of our colleagues who took time from their busy schedules to attend and present papers in our symposium. The various themes of their papers serve to reflect th e diversity of Belizean Archaeology, and provide a wealth of scientific information to the people of Beliz e. Thanks to our guest sp eaker, Dr. Nancy Adamson of Galen University in Belize. Several editors and anonymous revi ewers gave of their time and ma de constructive criticisms that enhanced our end product. George Thompson assistance in doing line editing and reviewing of all the bibliographic references is always invaluable and appreciated. Finally, it should be noted that our successes are a di rect result of the tremendous effort expended by the entire staff of the Institute of Archaeology. Special thanks must be given to Enrique Jack Itza, Melissa Badillo, Jorge Stakey Can, Darcy Billy Correa, Cl audia Elena, Nick Foster, David Griffith, Rafael Guerra, Delsia Marsden, Joyce Tun-Quewell, Wayn e Moore, George Thompson Program Chair and Brian Woodye, Executive Chairperson of the Symposium. We are grateful for your patience, expertise and invaluable help. The fourth volume of the Research Reports in Belizean Archaeology is a testimony to your dedication and your ability to address the challenges of our institution. John Morris, Sherilyne Jones and Jaime Awe Belmopan, Belize, June 2007 iv

PAGE 6

J. Morris et al. v

PAGE 7

CONTENTS page SECTION ONE: THE TERMINAL CL ASSIC PERIOD 1 1. Terminal Classic Maya of Be lize: Key Issues and Questions John M. Morris, Gyles Iannone and Jaime J. Awe 3 2. This is the End: Archaeological Transit ions and the Terminal Classic period at Caracol, Belize Arlen F. Chase and Diane Z. Chase 13 3. Fighting the Inevitable: Th e Terminal Classic Maya of the Upper Roaring Creek Valley Jaime J. Awe and Christophe G.B. Helmke 28 4. The Ancient Maya Drought Cult: La te Classic Cave Use in Belize Holley Moyes 43 5. On Drought and the Collapse of the Ancient Maya Gyles Iannone 55 6. Late and Terminal Classic Occupation of Pusilha, Toledo District, Belize Site Planning, Burial Patterns and Cosmology Geoffrey Braswell 67 7. Commoner Ritual and Socio-Political L ife in a Late Classic Neighborhood: Archaeological Investigations at the Northeast Group, Chan Site, Belize Chelsea Blackmore 79 8. Unexpected Discoveries in Chau Hiix Se ttlement: Suggestions for Terminal Classic Residence in North Central Belize A. Sean Goldsmith 89 9. The Curse of the Ware: Usi ng Ceramic Systems in Belize James J. Aimers 101 vi

PAGE 8

J. Morris et al. vii page 10. Fresh Volcanic Glass Shards in the Pottery Sherds of the Maya Lowlands Anabel Ford and Frank Spera 111 11. The Classic to Postclassic Transition in the Sibun Valley, Belize: Defining Terminal Classic Ceramic Assemblage Eleanor Harrison Buck and Patricia McAnany 119 12. The Terminal Classic in the Three Rivers Region Lauren Sullivan, Brett Houk and Fred Valdez, Jr., 135 13. Terminal Classic Events: Colha, Belize and the Central Maya Lowlands Fred Valdez, Jr. and Palma Buttles 147 14. Factors of Stress and Processes leading to the En d of the Classic Period at Blue Creek Thomas H. Guderjan and Coleen Hanratty 155 SECTION TWO: GENERAL RESEARCH REPORTS 167 15. The Middle Formative Ideological Foundations of Kinship: The Case from Cahal Pech James Garber, Jennifer Cochran and Jaime J. Awe 169 16. From House to Holy: Formative Development of Civic Ceremonial Architecture in the Maya Lowlands Terry Powis and David Cheetham 177 17. Rockshelter Excavation in th e Caves Branch River Valley Gabriel Wrobel, James Tyler and Jessica Hardy 187 18. Fifteen years of Settlement Pattern Surveys in the Three River Region of Northwestern Belize: What have We Learned Hubert R. Robichaux 197 19. Hinterland Settlement in Northwestern Belize: Excavations at the Medicinal Trails David M. Hyde and Shelley L. Fischbeck 213 20. Investigations of Akab Muclil: A Late Hi nterland Settlement in Northwestern Belize Antonio E. Padilla, Molly Morgan and Jon C. Lohse 221

PAGE 9

1 TERMINAL CLASSIC MAYA OF BELIZE: KEY ISSUES AND QUESTIONS John Morris, Gyles Iannone and Jaime Awe Recent archaeological investigations in Belize provide new, and significant, insights to th e Terminal Classic (A.D. 8001050) and subsequent collapse of ancient Maya civilization, but many key questions remain unanswered. Excavations over the past two decades at the Maya centers of Caraco l, Minanha, Xunantunich, Cahal Pech, Pusilha, La Milpa, Colha, Blue Creek, Chau Hiix, and other subterranean sites, both informs general phenomena surrounding the collapse and provides specific and varied examples. Research at th ese Belizean Maya sites document an increase of population and over-settlement of the region in the three centuries just prior to 1050 A.D. In the succeeding two hundred years before and after the Terminal Classic period the lowland Maya at these sites experienced a severe decline in many facets of their civilization including but not limited to popula tion numbers, sophisticated technology, architectural constructions, erection of stelae, disruptions in long distance trade networks, and fewer ar tistic traditions. Evidence for this decline in Late Terminal Classic Maya architecture, ceramic and technol ogical achievements, as well as evidence of increased warfare and migration, is reviewed and issues requiring further research are highlighted Introduction Investigations by archaeologists in the past two decades have revealed that many of the hallmarks once attributed to the collapse of the Classic Maya (A.D. 300-900) have their antecedents in the period known as the Terminal Classic (A.D. 800-1050) and perhaps even much further in antiquity than previously assumed. Cleavages and societal fractures in the preceding Late Classic Period (A.D. 550-750) (see Demarest 2004; Martin and Grube 2000; Sharer 1994) may also shed light on the demise of the ancient Maya. In particular, there are strong indications that ancient Maya society was a highly ranked and stratified one during the Late Classic period (Chase and Chase1993; Hammond 1986), but that many of the socio-political and economic institutions were (became) unwieldy. Maya scholars however, very rarely contemplate on the fragility of the base on which ancient Maya society was constructed. Clearly, the creation and emergence of a complex civilization is an extremely important process and should be thoroughly examined. Maya culture history research does not focus on the fundamental structures that enabled Maya society, which is unfortunate because we may be able to tease out the salient characteristics from within that underpin the ancient Maya collapse. The Terminal Classic period is a very short time span, roughly three times less than either the preceding Classic, or the succeeding Postclassic period. Despite its short length, the Terminal Classic period also remains one of the most enigmatic and controversial eras in the history of the an cient Maya. It marks an important period of political, economic, and social transformation throughout Mesoamerica that is mostly in reference to dramatic collapses. During this period, Maya centers in the Peten core area were declining, whereas polities to the north in the Yucatan region were on the rise. A recent book on Maya archaeology entitled The Terminal Classic edited by Demarest, Rice and Rice (2004) show how the ancient Maya collapsed but it spends far too much ink on warfare and social discontent and other paradigms such as envi ronmental determinism that are strongly emphasized at the expense of Research Reports in Belizean Archaeology Vol. 4, 2007, pp.3-12. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 10

Morris et al. ot her competing analyses and overviews. Are these accurate portrayals of what we know about the Terminal Classic Maya? The answer is a partial yes, but with the important caveat that with more data our thinking of how, why, and when the Maya collapsed is improving. Furthermore, investigations in Belize over the last few years are helping to correct this long-standing imbalance in reporting and information on the Terminal Classic and subsequent collapse of the ancient Maya. We must point out that while there have not been as many detailed, large-scale, studies of Terminal Classic Maya sites, as we have seen for Classic centers, several sites in Belize (Pusilha, Xunantunich, Caracol, Colha, Lamanai and La Milpa) have generated useful data that can be used to analyze the Terminal Classic Maya. The site of Xunantunich (Belize), in particular, has produced crucial details on the Terminal Classic Maya (Leventhal 1997). The construction and occupation of Xunant unich beyond A.D. 1000 is indicative of a vibrant culture desperately trying to hang on to its glorious past. Lecount (1999, 2005) detailed analysis and refinement of ceramic changes in the Late and Terminal Classic provides scholars with a useful tool to compare and contrast with their own findings. New research in caves by the BVAR (Awe et al 2003, 2005.) has also yielded important information on the issue of environmental degradation (read drought) that occurred in the Belize Valley and its implications for the Maya collapse. Despite these advances there remain many more questions than answers about the Terminal Classic Maya cultural development and demise. Terminal Classic Maya of Belize: Mayanists often query when were the latest Terminal Classic settlements recorded in Belize? Did the Maya collapse simultaneously at most lowland Maya sites? Was this demise punctuated or gradual? Research at various sites has been consistent on responding to these questions. Complex societies, like the ancient Maya, were not homogenous. Thus, change can (and did) occur differentially in various places and at different times and we cannot assume that these changes are uniformly mirrored across the rest of that society. Let us examine two areas of Belize. At Pusilha located in Southern Belize, Braswell (Braswell et al 2005 and this volume) ha s recovered Terminal Classic pottery from surface and floor contexts. He notes that Late Classic Pusilha was a Tepeu-sphere site sharing much with southern Peten, but little with the Belize Valley. However at the end of the 8th and early 9th centuries at Pusilha, Belize Red from the Belize Valley was imported which demonstrates exchange relations with new regions during this time period. Fine Orange was also imported or manufactured locally, and carved drinking vessels of the Brandy Snifter form also suggest ties with the northwestern Maya lowlands. But the crude and un-standardized Postclassic ceramic assemblage represents a sharp break from Classic traditions. Braswell argues that Pusilha may have been founded at the beginning of the Late Classic period by factions who sought southern Belize as a haven against political troubles in the Peten. Braswell notes that at the end of the Terminal Classic violence permeated the society as his investigations reveal that several individuals were left dead on the largest platform at Pusilha. In the Sibun Valley which is closer to the center of the country the situation differs. In a sense our questions takes on added import because ethnic considerations impinge on our analysis. Who were the people and settlers of the Terminal Classic in this area? McAnany et al. (2004) based on analyses of some of the late pottery and architecture from Sibun and several other sites in the region argue that the Belize Sibun settlers were Maya but heavily 4

PAGE 11

Terminal Classic Maya of Belize i nfluenced by Chichen Itza. McAnany (2004) proposes that by the end of Terminal Classic these people living in the Sibun River Valley had become totally amalgamated and fully incorporated in the culture of northern Yucatan. Evidence indicates that during the Terminal Classic period a rich assemblage of foreign traits, namely circular architecture appear to overlie the local traditions which were Peten affiliated such as stela and altar dedication. McAnany and Harrison Buck recent excavations in the Sibun Valley at Pechtun Ha, Oshon, and Obispo exposed examples of Chichen Itza style of circular architecture; these repr esent the southernmost examples found in the eastern lowlands. Excavations also recovered Terminal Classic ceramic types associated with Tepeu/Spanish Lookout (Roaring Creek Red dishes, Belize Red dishes and plates. Analyses of the ceramics also demonstrate that Garbutt Creek Red bowls were replaced by a variety of large basin and bowl forms with thick bolstered rims, some containing northern-style polychrome and incised designs. The form and rim treatment is argued to be reminiscen t of florescent medium Puuc and Chichen slate ware basins. Clearly, the circular structures in the Sibun Valley shares similarities with other examples from Maya lowlands, especially the Terminal Classic round structures at Uxmal (Kowalski et al. 1996), Nohmul (D. Chase and A. Chase 1982; Hammond 1985) as well as the early phases of the Caracol at Chichen Itza (Pollock 1936; Rupert 1935 [fig 3] ). McAnany points out that overarching similarity found in nearly all Terminal Classic circular architecture in the lowlands is the presence of a low plinth or step construction that surrounds the perimeter of the building The 8th century Maya has been characterized as a period of decentralization in which subordinate lords accrued greater powers and figured more prominently in political affairs (e.g. Culbert 1991:325; Fash 1991; Fash and Stuart 1991:172; Schele 1991a;78, 1991b; Stuart 1993:332, 336, 349). Powerful centers such as Caracol and Naranjo were clearly affected by this changing sociopolitical landscape. Caracol erected only one inscribed monument between 680-798 A.D. (Martin and Grube 2000:95), and Naranjo produced only one monument (Stela 20-746 A.D.) between 726-780 A.D. (Martin and Grube 2000:78-79). Minanha royal court came into prominence during a period of regional balkanization but eventually we observed the demise of this royal court at the same time when there is a return to prominence of Caracol and Naranjo; epigraphic records tells us that when Caracol and Naranjo reentered the arena of regional politics at the end of the 8th century, they moved quickly to restore their old teritiorres. This involved both centers undertaking military campaigns and/or initiating diplomatic missions within their borderlands, at places such as Bital and Ucanal (A. Chase 2004:330; Martin and Grube 2000). As Iannone documents, the filling in of elite residences prior to actual site abandonment occurred as part of the Terminal Classic collapse sequence and this has been documented at a number of other centers, including Xunantunich (Lecount et al 2002:44), at La Milpa (Hammond1999a: 13, 1999b and Lamanai (Graham 2003). These filling events suggest that one of the first consequences of the infamous Maya collapse was often the effacemen t of both the physical and social ruling houses of many lowland centers. Interestingly, Stela 11 at Copan, the latest monument at the site, may refer to just such an obliteration of a founders house (Martin and Grube 2000:212; Stuart 1993:344-346), an event which has also been interpreted by David Stuart (1993:346) as a possible metaphorical reference to the end of Copans ruling line. Thes e events suggest that the collapse itself may had had a punctuated quality that has yet to be incorporated into our 5

PAGE 12

Morris et al. m odel building The Collapse of Ancient Maya Civilization As Norm an Yoffee (2005:132) points out, the subject of collapse has been a focus in the writings of historians and philosophers for at least two millenniums [sic] (see also Tainter 1988; Yoffe and Cowgill 1988). Although a late addition to this realm of inquiry, laypeople and academics alike have still spent the past two centuries attempting to solve the mystery that is the ancient Maya collapse. Nevertheless, what forced the southern lowland Maya to desert their cities, and abandon many of their cultural practices, continues to be one of the most hotly debated archaeological topics, both inside and outside academia. This wideranging interest in the collapse cannot be simply attributed to our innate attraction to a mystery. Nor is it entirely rooted in the ro mantic draw of lost cities in the jungle. There is, in addition to these two stimuli, a very strong realization that the dramatic demise of the once successful southern lowlands city-states carries with it lessons that must be heeded by contemporary society (Diamond 2005:525; Yoffee 2005:132). For this reason, Arthur Demarest (2004:7) contends that research into the collapse is one of the very areas where Maya archaeology can make a general contribution to social science. Yoffee (2005:132) concludes that, Collapse studies in archaeology are importantif for no other reason than to show that the suppositions and common knowledge beliefs that many people have had about collapse are extremely questionable. So, the question remains, what have we learned about the colla pse of ancient states given our longstanding pr eoccupation with the topic? In the case of the fall of the Maya city-states during the Late Classic to Terminal Classic Transition, there is general consensus on a number of issues. For one, there is near universal agreement with respect to the material correlates of the collapse. These include: a decline in the erection of monuments, along with their associated dynastic texts and iconography of rulership; the demise of polychrome ceramic production; the abandonment of many forms of inscriptional art; the termination of elite building programs, in some case abruptlyin mid construction; a cessation in the construction of elaborate tombs; a drastic alteration in the form of governance; and, significant migrations and population losses (Culbert 1973, 1988; Chase and Chase 2004:15; Demarest 2001:106; Demarest et al. 2004a:569; Iannone 2005; Lucero 2002; Lowe 1985; Robichaux 2002:341; Webster 2002). Most Mayanists would also agree that the collapse was not an abrupt political fragmentation that impacted all of the citystates in the southern lowlands simultaneously. Rather, it was a protracted process that played itself out over three centuries, beginning around 750 A.D., and ending approximately 1050 A.D. (Rice et al. 2004:8). For some city-states the collapse was both rapid, and final. For others, and for support populations in general, the decline was often more gradual. Decades of sitespecific and regional analyses have made it quite clear that the Late Classic to Terminal Classic transition was also highly variable in characterthere is no one-size-fits-all collapse sequence for the southern Maya lowlands (Demarest 2001:106); Iannone 2005; Rice et al. 2004:10). Nevertheless, it is also believed that various lo cal collapse sequences can be combined to develop a broader, macroregional understanding of the decline (e.g., Iannone 2005; Rice et al. 3004:11). Finally, although a few dissenters remain (e.g., gill 2000), most scholars now view the collapse as a widespread sociopolitical and socioeconomic transformation that was most detrimental to the elite sector of ancient Maya society. In other words, what really collapsed in the southern lowlands was the 6

PAGE 13

Terminal Classic Maya of Belize Classic period style kingship institution and all of its constituent com ponents (Demarest 2001:106; Iannone 2005, 2006; Rice et al. 2004:9). Notwithstanding the aforementioned points of consensus, a number of contentious issues remain when it comes to the study of the ancient Maya collapse. On a theoretical level, there continues to be considerable debate as to how to characterize the collapse itself. After decades of referring to the demise of the southern lowland city-states as a collapse, it has recently become vogue to downplay the significance of the changes that occurred during the shift from the Late Classic to the Terminal Classic period. As such, while there are still authors who continue to refer to the collapse era as a demographic disaster as profound as any other in human history (Haug et al. 2003:1733; see also Gill 2000:313), others are reluctant to follow suit, opting for less loaded, less apocalyptic termssuch as cultural transitions, transformation, or realignment (see also Cowgill 1988; Demarest 2001:105106; Rice et al. 2004:6, 8; Sabloff 992:108). One strand of this terminological debate can be traced to the results of individual archaeological projects: those working in regions where the Late Classic to terminal Classic transition was violent and abrupt in charactersuch as the Petexbatun Region (e.g. Demarest 2004:101), are more at ease with the term collapse than those whose data support more subtle transformations (e.g, D. Chase and A. Chase 2004:26; Rice and Rice 2004). Here it is profitable to return to the seminal work of Joseph Tainter (1988:193), who argues that a Collapse is fundamentally a sudden, pronounced loss of an established level of sociopolitical complexity. As noted earlier in this paper, there is near universal acceptance that the key feature of the Late Classic to Terminal Classic Transitions was the demise of the institution of kingship in the southern lowlands. In other words, the upper tier of the ancient Maya sociopolitical hierarchy was truncated (see also Rice et al. 2004:9). This pronounced loss of an established level of so ciopolitical complexity was even felt at centers such as Lamanai, which has often been held up as one of the exceptions to the rule with respect to the collapse (Pendergast 1986). In our view, there is enough data to support the idea that, in terms of their grater degree of sociopolitical clout, the elite that ruled 8th century Lamanai were very different than their 9th and 10th century counterparts. For this reason, we remain quite comfortable with the term collapse, as long as it is used in the manner that Tainter suggests. Finally, of greatest significance is the fact that Mayanists continue to vociferously debate the reason behind the fall of the southern lowland city-states. As Tainter (1988:194) once pointed out: Collapse theorists have taken to heart the Maoist dictum to let a hundred schools of thought contend. This statement definitely rings true for Maya archaeology. Over the years, explanations for the ancient Maya collapse have run the gamut from the external to the internal, from the environmental to the social; from the unicausal to the multicausal (see Rice et al. 2004). This diversity in explanatory positions is partially the result of the infamous archaeological disease know as mysiteitus the most pronounced symptom of which is the delusion that ones microcosm is everyones macrocosm. The greater variety of opinions on the collapse are thus, at least partially, the result of site-specific and/or regional databases and developmental sequences for the Late to Terminal Classic transition being projected as proxies for the southern lowlands in their entirety (Demarest 2004a: 546-547). The diversity of interpretations for the collapse is also a direct result of the complexity of the problem at hand. As 7

PAGE 14

Morris et al. indicated previously, the study of individual city-states suggests that the process of collapse was not the sam e for every site or region. Not only are the local time frames highly variable, but the circumstances, characteristics, and contributing factors also appear to have been equally diverse (see various papers in Demarest et al. 2004b; Iannone 2005). For this reason, it is understandable that a myriad of theories would emerge to explain the collapse. Taken at face value, the variable data set does imply that a complex, multicausal model might best account for the collapse. In fact, most Mayanists appear to support this notion, as is attested by the vast majority of the papers contained in the recently published volume on the Terminal Classic edited by Arthur Demarest, Prudence Rice, and Don Rice (2004b). Importantly, recent crosscultural treatments, such as those by Jared Diamond (2005) and Ronald Wright (2004), also support complex multicausal models for the collapse of early states. Nevertheless, unicausal models still abound. In fact, one of the more popular models in recent years one which has been offered as the eloquently simple panacea for all our collapse problemsis the drought hypothesis (e.g. Gill 2000) One archaeological data set that presents challenges for the drought hypothesis is the infilling and/or termination of elite residential courtyards prior to total site abandonment. Excavations at Belizean centers such as Dos Hombres, Blue Creek, and possibly Punta de Cacao, have exposed large Terminal Classic deposits of ceramic sherds and other items on the floors and stairs associated with elite courtyards (Adams et al. 2004:337). These deposits, some of which, block the entrances to the courtyards themselves, are thought to represent acts associated with termination of the elite inhabitants themselves, as opposed to the actual architectural features that they are associated with (Adams 2004:338). Even more dramatic is the total infilling of royal residential courtyards, and their associated rooms and residential buildings, during the early part of the Terminal Classic period. This practice has been recorded at La Milpa (Hammond 199a, 199b; Hammond and Thom as 1999), Lamanai (Graham 2004), and Minanha (Iannone 2005, 2006). At the latter center the royal residential courtyard and its associated buildings were swept clean prior to laying down a ca. 20cm thick layer of fine sediments over all of the architectural components. Large limestone slabs were then carefully laid upon all floor and stair surfaces, and sturdy construction pens were constructed to hold the ca. 5 meters of rubble that was required to bury all of the features a ssociated with the late Classic royal residential compound (except the upper third of the largest pyramidal structure). Significantly, there are no signs of destruction associated with this event; the vaulted buildings were, for the most part, buried intact by laboriously carrying dry-stone fill through the narrow doorways and meticulously stacking it to the ceilings. Following this massive investment of organized labor, a rather impoverished Terminal Classic residential group was erected over the buried elite architecture. If droughts were the cause of the decline, and if things were as bad as the proponents of this model say, how come the Minanha Maya did such a careful, and labor intensive job of burying their Late Classic Royal residence? And why did they do this when a sizable population continued to reside at the center? The only reasonable answer is that, although the Minanha community survived into the Terminal Classic, its royal court and by association the Minanha city-state itself ceased to exist. In other words, the institution of kingship met its demise prior to any form of overall site abandonment. This data conflicts with Gills model of drought induced famine. For example, Gill 8

PAGE 15

Terminal Classic Maya of Belize (2000:374) argues that, Fam ine is generally a class affliction. It hits the poor hardest and firstthen disintegration begins with the lowest levels of society, including the peasants, and it proceeds from the bottom up. Elsewhere, he surmises that, In the case of the Maya, due to the length and severity of the droughts, the flow of food and water was shut off, the people died from hunger and thirst, and the social system collapsed from the bottom up(Gill 2003:371). The Minanha data, along with that from the other centers mentioned above, suggests the exact opposite. The first segment of society to meet their demise was the elite, and particularly the royalty. That the surviving communities were still relatively strong is implied by the amount of labor and organization that went into the infilling and termination events previously discussed. To quote Gill (2000:345), the construction of major formal buildings, or the lack thereof, can be an indicator of the vitality of a city and, by impli cation, of the population with a certain degree of vitality. Starving, thirsting wretches on the doorstep of death did not carry out these building acts. In summary, the Terminal Classic droughts may have been difficult to overcome because of their severity. Nevertheless, the actual archaeological record provides no strong evidence for a pan-lowland famine. Equally important is the fact that the culture history clearly shows that some regions, such as the Petexbatum, may have collapsed prior to the onset of droughts, for various economic and political reasons. The initial collapse may have eventually put other areas in jeopardy, as migrants from the distressed region put demographic pressures on the city-states of the east. Indications are that these migrations began prior to the onset of any droughts, although aridity may have played a role in later demographic shifts. Conclusions A heightened interest by scholars in the Terminal Classic Maya has emerged (Demarest 2004; Rice and Rice 2003). What evidence is there for increasing social and political stress during the Terminal Classic period? We can point to the signs of extensive warfare and the construction of public ( versus domestic) architecture for defensive purposes appearing by ca. 700 A.D. in the Belize Valley (Brown and Garber 2005) thus signaling that Maya society was undergoing cleavages and other social stress. The rise of new city-states in the early 8th century again prior to any proposed droughtsmay have also put increased strain on the established tributary and exchange networks. Eventually, populations did begin to declinealthough our understanding of the actual demographic losses continues to be hampered by the fact that our assessments of the Terminal Classic population levels are likely underestimations, given the fact that Late Classic ceramics continued to be used well into the Terminal Classic (D. Chase and A. Chase 2004:18-19; Demarest et al. 2004a: 550; Rice and Rice 2004:128; Sabloff 1973: 113-116). What is clear is that the elite, and particularly the royal segments of society, fell first. The collapse was th erefore initially a top down process, a factor that once again contradicts the expectations of a collapse model based on drought and famine. Drought may have contributed to the decline of the institution of kingship, and eventually the whole population, but it was clearly a secondary cause. Other recent excavations and analyses of Terminal Classic materials from Pusilha, Caracol, Sibun Valley sites, Belize River Valley Xunantunich, and the Three Rivers Region have resulted in new perspectives on the demise of ancient Maya civilization, and what is clearly one of the most fascinating stages of Mesoamerican cultural development. In this regard, it is important that researchers not lose sight of what ought to be the major research goalsto understand the trajectory of 9

PAGE 16

Morris et al. how and why politically and econom ically complex societies wax and wane; the how, why and when of ancient Maya civilization fall. References Cited Adams, Richard E. W., Hubert R. Robichaux, Fred Valdez Jr., Brett A. Houk, and Ruth Mathews 2004 Transformations, Periodicity, and Urban Development in the Three Rivers Region. In The Terminal Classic in the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 324-341. University of Colorado Press, Boulder Awe, Jaime J., Cameron Griffith, and Sherry Gibbs 2005 Cave Stelae and Megalithic Monuments in Western Belize. In In the Maw of the Earth Monster: Mesoamerican Ritual Cave Use, edited by James E. Brady and Keith M. Prufer, Pp. University of New Mexico Press, Albuquerque. Brady, James E. 1994 The Role of Caves in Ancient Maya Cosmovision. Paper presented at the Earthly Matters II Roundtable. Dumbarton Oaks, Washington, DC. Brady, James E. and Andrea Stone 1986 Naj Tunich: Entrance to the Maya Underworld. Archaeology 39(6): 18-25. Chase, Diane Z., and Arlen F. Chase 2004 Hermeneutics, Transitions, and Transformations in Classic to Postclassic Maya Society. In The Terminal Classic in the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 12-27. University of Colorado Press, Boulder. Cowgill, George L. 1988 Onward and Upward with Collapse. In The Collapse of Ancient States and Civilizations, edited by Norman Yoffee and George L. Cowgill, pp. 244-276. University of Arizona Press, Tucson. Culbert, T. Patrick (editor) 1973 The Classic Maya Collapse University of New Mexico Press, Albuquerque. 1988 The Collapse of Classic Maya Civilization. In The Collapse of Ancient States and Civilizations, edited by Norman Yoffee and George L. Cowgill, pp. 69-101. University of Arizona Press, Tucson. Danien, Elin C. and Robert J. Sharer (editors) 1992 New Theories on the Ancient Maya University of Pennsylvania, University Museum, Monograph 77. Philadelphia, PA. Demarest, Arthur 2004 Ancient Maya: The Rise and Fall of a Rainforest Civilization. Cambridge University Press, Cambridge (UK). Demarest, Arthur A. 2001 Climate Change and the Classic Maya Collapse: The Return of Catastrophism. Latin American Antiquity 12:105-107 Demarest, Arthur A., Prudence M. Rice, and Don S. Rice 2004a The Terminal Classic in the Maya Lowlands: Assessing Collapses, Terminations, and Transformations. In The Terminal Classic in the Maya Lowlands, edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 545-572. University of Colorado Press, Boulder. Demarest, Arthur A., Prudence M. Rice, and Don S. Rice (editors) 2004b The Terminal Classic in the Maya Lowlands University of Colorado Press, Boulder. Diamond, Jared 2005 Collapse: How Societies Choose to Fail or Succeed Viking, Toronto. Garber, James F., M. Kathr yn Brown, W. David Driver, David M. Glassman, Christopher J. Hartman, F. Kent Reilly III, and Lauren A. Sullivan 2004 Archaeological Investigations at Blackman Eddy. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 48-69. University Press of Florida, Gainesville. Gill, Richardson B. 2000 The Great Maya Droughts. University of New Mexico Press, Albuquerque. 10

PAGE 17

Terminal Classic Maya of Belize Gifford, James C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley. Harvard University, Peabody Museum of Archaeology and Ethnology, Memoir 18. Cambridge, MA. Graham, Elizabeth A. 2004 Lamanai Reloaded: Alive and Well in the Early Postclassic. In Research Reports in Belizean Archaeology, Volume 1 pp. 223-241. Guderjan, Thomas H. 2004 Public Architecture, Ritual, and Temporal Dynamics at the Maya Center of Blue Creek, Belize. Ancient Mesoamerica 15: 235-250. Hammond, Norman 1999a A Pillar of State Majestic, Though in Ruin: The Royal Acropolis of La Milpa. Context 14(1): 11-15. 1999b A Maya Throne Room at La Milpa. PreColumbian Art Research Institute Newsletter 28: 12-13. Hammond, Norman, and Ben Thomas 1999 Another Maya Throne Room at La Milpa. Context 14(1): 15-16. Haug, Gerald H., Detlef Gunt her, Larry C. Peterson, Daniel M. Sigman, Konrad A. Hughen, and Beat Aueschilmann 2003 Climate and the Collapse of Maya Civilization. Science 299: 1731-1735. Healy, Paul F. 1990 Excavations at Pacbitun, Belize: Preliminary Report on the 1986 and 1987 Investigations. Journal of Field Archaeology 17(3): 247-262. 1992 The Ancient Maya Ballcourt at Pacbitun, Belize. Ancient Mesoamerica 3: 229-239. Healy, Paul F., David Cheetham, Terry G. Powis, and Healy, Paul F., Bobbi Hohmann, and Terry G. Powis 2004 The Ancient Maya center of Pacbitun. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 207-227. University Press of Florida, Gainesville. Iannone, Gyles 2005 The Rise and Fall of an Ancient Maya Petty Royal Court. Latin American Antiquity 16:26-44. 2006 Investigations in the Buried Royal Residential Courtyard at Minanha, Belize. In Research Reports in Belizean Archaeology, Volume 3, edited by John Morris, Sherilyne Jones, Jaime Awe, and Christophe Helmke, pp. 149-160. Institute of Archaeology, National Institute of Culture and History, Belmopan, Belize. Lowe, John W.G. 1985 The Dynamics of Apocalypse: A Systems Simulation of the Late Classic Maya Collapse. University of New Mexico Press, Albuquerque. Lucero, Lisa 2002 The Collapse of the Classic Maya: A Case Study for the Role of Water Control. American Anthropologist 104: 814-826. Martin, Simon and Nikolai Grube 2000 Chronicle of the Maya Kings and Queens: Deciphering the Dynasties of the Ancient Maya. Thames and Hudson, London (UK). McAnany, Patricia A. 1995 Living with the Ancestors: Kinship and Kingship in Ancient Maya Society. University of Texas Press, Austin. 2004 KAxob: Ritual, Work and Family in an Ancient Maya Village. University of California-Los Angeles (UCLA), Cotsen Institute of Archaeology, Monumenta Archaeologica 22. Los Angeles. McKillop, Heather 2004 The Ancient Maya: New Perspectives. ABC Cleo, San Diego. Pendergast, David M. 1986 Stability Through Change: Lamanai, Belize, from the Ninth to the Seventeenth Century. In Late Lowland Maya Civilization: Classic to Postclassic, edited by Jeremy A. Sabloff and E. Wyllys Andrews V., pp. 223249. University of New Mexico Press, Albuquerque. Rice, Prudence M., and Don S. Rice 2004 Late Classic to Postclassic 11

PAGE 18

Morris et al. 12 Transformations in the Peten Lakes Region, Guatemala. In The Terminal Classic in the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 125-139. University of Colorado Press, Boulder. Rice, Prudence M., Arthur A. Demarest, and Don S. Rice 2004 The Terminal Classic and the Classic Maya Collapse in Perspective. In The Terminal Classic and the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 1-11. University of Colorado Press, Boulder. Robichaux, Hubert R. 2002 On the Compatibility of Epigraphic, Geographic, and Archaeological Data, with a Drought-Based Explanation for the Classic Maya Collapse. Ancient Mesoamerica 13:341-345. Sabloff, Jeremy A. 1973 Continuity and Disruption during the Terminal Late Classic Times at Seibal: Ceramics and Other Evidence. In The Classic Maya Collapse edited by T. Patrick Culbert, pp. 107-131. University of New Mexico Press, Albuquerque. 1992 Interpreting the Collapse of Classic Maya Civilization. A Case Study of Changing Archaeological Prespectives. In Metaarchaeology: Reflections by Archaeologists and Philosophers edited by Lester Embree, pp. 99-119. Kluwer Academic Publishers, Boston. Schele, Linda and David A. Freidel 1990 A Forest of Kings. William Morrow, New York, NY. Sharer, Robert J. 1994 The Ancient Maya, 5th edition. Stanford University Press, Stanford, CA. Tainter, Joseph A. 1988 The Collapse of Complex Societies Cambridge University Press, New York. Webster, David 2002 The Fall of the Ancient Maya Thames and Hudson, New York. Wright, Ronald 2004 A Short History of Progress Anansi, Toronto. Yoffee, Norman 2005 Myths of the Archaic State Cambridge University Press, New York. Yoffee, Norman, and George L. Cowgill (editors) 1988 The Collapse of Ancient States and Civilizations University of Arizona Press, Tucson.

PAGE 19

2 THIS IS THE END: ARCHAEOLOGICAL TRANSITIONS AND THE TERMINAL CLASSIC PERIOD AT CARACOL, BELIZE Arlen F. Chase and Diane Z. Chase Ongoing excavations at Caracol, Belize have yielded a significant amount of data pertinent to the Terminal Classic Period. These data are especially relevant to interpretations of the Southern lowland Maya collapse. Caracol has produced meaningful temporal and spatial data that help to expose conceptual problems in the identification, interpretation, and dating of Terminal Classic contexts. Of particular interest have been the on-floor ceramics in many of Caracols central buildings that have been system atically collected, assembled, and illustrated over more than two decades. These materials not only provide information relative to sub-assemblages of vessels that were physically used together, but also allow latest building (and, in some cases, room) function to be inferred. These ceramic sub-assemblages also contain trade items that pe rmit insight into the social and economic relationships that were maintained by Caracols latest inhabitants. Ceramics that was used in the sites extensive settlement system can be compared with ceramics used at the site ep icenter, which reveals the use of two distinct ceramic subcomplexes, by Caracols in habitants during the Terminal Classic era. When combined with ot her data classes, the archaeology from Caracol permits a detailed understanding of what transpired at the site before its final abandonment. Introduction There are a series of times in Maya prehistory when there appears to be rapid change, providing challenges to archaeological interpre tation. When such transitions are accompan ied by alterations in material culture that affect only one segment of any given society, ar chaeological analysis may be hampered, particularly if sampling design and preconceived models do not readily lead to the identification of contemporary variation. This is especially true for the transition between the Preclassic and Classic Periods in the Maya area sometimes called the Protoclassic and is also the case for the transition between the Classic and Postclassic eras usually called the Terminal Classic Period. At Caracol, Belize, the Terminal Classic Period elite material culture changed, but the material culture used by the bulk of the population remained relatively constant. The recognition of this fact has clear repercussions for reconstructions of the Terminal Classic Period and the Classic Maya collapse. The Terminal Classic The Terminal Classic era, dating from approximately A.D. 780 through A.D. 900, is key to interpreta tions of the ancient Maya because it precedes or is conjoined with the abandonment of many of the larger cities. The Terminal Classic has proven particularly enigmatic, especially as hieroglyphic texts do no t generally exist to guide the archaeologist and material culture was in the proc ess of changing as were traditional interm ent patterns and other aspects of the archaeological record (e.g. D. Chase and A. Chase 2006). While broadly recognized as being the latest temporal era before the Classic Maya collapse, traditionally the Terminal Classic was identified only in terms of recognized ceramic type fossils. As originally defined for Uaxactun, Guatemala (and subsequently elsewhere), Terminal Classic ceramics were the detritus left over when recognized Late Classic types were Research Reports in Belizean Archaeology Vol. 4, 2007, pp.13-27. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 20

Arlen F. Chase & Diane Z. Chase rem oved from analyzed samples (Smith 1955). Unfortunately, while various scholars have noted problems in the identification of Terminal Classic and Early Postclassic materials (e.g. Pendergast 1986, 1990), a full contextual revision of Terminal Classic ceramics and other artifactual materials has not yet taken place. Type fossils continue to be used to define the transition between the Late Classic and Postclassic Periods. Several factors led to this overdependence on specific ceramic subsets to recognize the Terminal Classic time period (see also A. Chase and D. Chase 2004a: 344). Grounded in an older model of Maya sites as vacant ceremonial centers, there was a widespread, but mistaken, belief that palace buildings were not lived in, being only sporadically used. If they were not true residences, there would be no contemporary garbage associated with building floors. Thus, there was no intentional searching for in situ remains, it was a foregone conclusion that none existed. The end result was a selffulfilling prophecy in which previously identified type-fossil ceramic types took on a meaning of their own and became the only way to recognize the Terminal Classic Period. Because many of the last remains at any site were never sealed in construction, archaeologists have had methodological problems interpreting these data. Material found during the excavation of buildings often appeared to be extremely broken and fragmentary; the assumption was that these artifacts were remnants of later reoccupation. The material in the humus and overlying plazas and latest constructions was viewed as being unusable analytically (e.g. Adams 1971). The type fossil approach to the Terminal Classic, in conjunction with the cessation of dated monument erection, led to views of rapid depopulation at sites like Tikal, Guatemala; in fact, the type fossils were ascribed to subsequent squatters living in collapsing buildings (Culbert 1973). Yet, the archaeological reality is likely something very different, if the conclusions derived from Caracol can be extended to other sites. The earlier assumptions that Classic Maya buildings did not have use-related material associated with their floors have been laid to rest by archaeological work at Caracol (A. Chase and D. Chase 2004a, 2005; D. Chase and A. Chase 2000) and at other sites like Aguateca (Innomata 1997; Innomata and Triadan 2000) and, even further such as Xochicalco (Webb and Hirth 2003) that emphatically demonstrate that de facto refuse may indeed be found. However, these various projects also have shown that any consideration of the Terminal Classic requires a critical and careful use of data that interdigitates laboratory analysis with field methodology. Areal clearing of building rooms and platforms at Caracol has revealed that many are associated with abandonment materials. Some of the materials are pottery and artifacts left where they were in the process of being used. Others constitute sheet refuse (Schiffer 1987), items thrown or placed in a certain area for later trash disposal. Refuse not only includes pottery but also burnt animal bone and the remains of craft production activ ities. However, in all of these instances, the debris is composed of reconstructable vessels and broken artifacts that were clearl y used at that locus. In a few rare cases, Terminal Classic trash was compacted into units of fill that were meant to form a building block for a larger, but never completed, construction. These kinds of contexts shift the focus of archaeological study to a close collaboration between the analyst and the excavator (e.g., Berggren and Hodder 2003). Rather than simply counting and typing sherds, the analyst can reconstruct whole 14

PAGE 21

The Terminal Classic Period at Caracol vessels, deduce m inimum vessel counts, and hopefully see what the entire assemblage or sub-assemblage looked like in specific contexts. Ceramic vessel forms can be combined with other remains, such as faunal materials, to more fully delineate ancient activities. This approach better permits an interpretation of specific contexts. The archaeological focus becomes the functional study and analysis of deposits as well as the delineation of assemblages, subassemblages, and subcomplexes and the identification of single-point-in-time abandonment materials; this allows far better intra-site dating and inter-site cross dating. Thus, not only does a consideration of abandonment materials lead to interpretations of functi on, but they also lead to refined dating of materials not usually possible in traditional an alyses in other than burial or cache contexts. When archaeological materials are considered in this holistic way, different paradigms and interpretations may evolve. This can be seen in the research that has been undertaken at El Ceren, El Salvador, by Payson Sheets (1998) and his colleagues (Sheets et al. 1990) where ancient community structure is being postulated and commonly held assumptions about the dating and distribution of archaeological materials have been challenged. Similarly, there are other excavated contexts in the Southern lowlands that have served as Rosetta stones for the Terminal Classic Period. In particular, we wish to talk about contexts that we have excavated at Nohmul, Belize in the late 1970s and, more recently, at Caracol. Contexts at both of these sites were and are important to resolving lingering questions of cross dating in the Maya lowlands and elsewhere in Mesoamerica. They also have an impact on interpretations of the Classic Maya collapse. Nohmul, Belize In order to investig ate the transition between the Postclassic and Classic Periods, Structure 20 at Nohmul, Belize was selected for excavation by analyzing mapped settlement. The construction seemed oddly out of place in the eastern plaza of the Nohmul epicenter, as it blocked access to two other range structures. Its low square form also seemed out of place in the repertoire of Classic Period architecture. With these facts in mind, we excavated Nohmul Structure 20 in 1978 with the aid of Norman Hammonds Corozal Project, which was excavating at Cuello at the time. In 1979 we returned to excavate an anomalously placed Nohmul Structure 9 (D. Chase and A. Chase 1982; D. Chase 1982a, 1982b). The two structures excavated at Nohmul both proved to be single-phase constructions built dir ectly on the latest plaza floor in the eastern plaza at that site. Besides being stratig raphically late, both buildings were also of unusual form. Subsequently defined as a patio-quad, Structure 20 proved to be almost identical in size to others only known from the site of Chichen Itza; Structure 9 proved to be a round platform and building, almost identical in size to an earlier version of Chichen Itzas famous Caracol (not to be confused with the site of that name in central Belize). Thus, ar chitecturally, both buildings appeared out of place in northern Belize and had some ob vious bearing on the occupation at Nohmul just prior to its abandonment. Even more surprising than the architecture were the ceramic materials. Within the fill of Structure 9 and abutting the rear of Structure 20 were the same kinds of pottery. The materials associated with Structure 20, however, included both in situ de facto debris and sheet refuse, indicating coeval use of diverse ceramic types that 15

PAGE 22

Arlen F. Chase & Diane Z. Chase could be largely reconstructed. Based on cross-f its between small sherds in the interior patio area of Structure 20 and the more complete shattered vessels in the alley behind the building, we were able to show that the materials behind the building had once been used within the edifice. The Structure 20 vessels consisted of a series of finewares and cooking ve ssels that included local and imported slatewares, local blackwares and redwares, red and cream tricklewares, as well as a grater bowl, a drum, and a San Jose V serving platter (Figure 1). Functionally, the building minimally had been used to feed and presumably entertain a group of high status individuals. Temporally, this excavated pottery proved to be a ceramic stone for aligning Terminal Classic ceramic sequences between the Northern and Southern lowlands (D. Chase and A. Chase 1982). At the time of our Nohmul excavations, Chichen Itza was widely viewed as a Postclassic site (some books still incorrectly place Chichen Itza in this time horizon; e.g. M. Coe 2005). Archaeologists now recognize that the bulk of Chichen Itzas architecture and occupation date to the Terminal Classic Period (Cobos 1999, 2004; Ringle et al. 1998). Much of the older temporal argument that placed Chichen Itza in the Postclassic Period derived from Tozzers (1957) ethnohistoric interpretation of Chichen Itza and the ceramic alignment of the Sotuta, Hocaba, and Cehpech ceramic complexes (Brainerd 1958; Smith 1971; A. Chase and D. Chase 1985). What the data from Nohmul showed we re problems in the then accepted sequential ceramic dating of Northern lowland pottery by demonstrating that several diagnostic types found in the Cehpech and the Sotuta ceramic complexes had to be coeval with other materials designated as being Terminal Classic in the Southern lowlands (San Jose V and local northern Belize wares). Thus, a single functional context (Structure 20) from an outlier site (Nohmul, Belize), probably only peripherally related to more main-stream interactions, provided a ceramic Rosetta stone for cross-dating and re-evaluating commonly held temporal suppositions and paradigms regarding the Terminal Classic Period. Santa Rita Corozal, Belize Our Nohmul expe rience with userelated materials replayed itself at Santa Rita Corozal between 1979 and 1985, but only in Postclassic contexts. At Santa Rita Corozal we were able to find abundant Late Postclassic abandonment materials in association with architectural remains and to make functional interpretations of subassemblages and ritual patterns (D. Chase 1982a, 1985a, 1985b; D. Chase and A. Chase 1988, 2000). However, in situ earlier Terminal Classic Period deposits remained elusive. While abunda nt and having many of the same local types as were found at Nohmul, most of the Terminal Classic ceramics at Santa Rita Corozal occurred in building fill. Nowhere was sheet refuse of a Terminal Classic date located and the stratigraphic sequence indicated a disjunction between the Terminal Classic and Postclassic Periods. Caracol, Belize Given our success with recovering abandonment materials at both Nohmul and Santa Rita Corozal, th e recovery of these kinds of deposits was one of our initial goals when we started excavations at Caracol in 1985. By our second season at Caracol, we had already identified in situ on floor remains associated with both Structures A3 and B19. By our four th field season in 1988, we were excavating abandonment 16

PAGE 23

The Terminal Classic Period at Caracol 17 Figure 1. Terminal Classic pottery associated with Nohmul Structures 9 and 20: a Kik Red; b Taak Orange-Red; c Ohel Red; d Campbells Red; e Sansomal Black-on-Cream; f. Chacil Red-on-Black; g Tzibana Gouged-Incised; h Chambel Striated; i Red Neck Mother Striated; j Chembeku Modeled; k Xixilic Incised; l Metzabok Slate; m. Savinal Cream; n Usukum Gouged-Incised; o Achiote Black; p Kik Red; o Buyuk Striated. materials associated with palace buildings on Caana. The recognition that these materials were associated with palaces meant that an inordinate amount of time was spent on articulation between the fieldwork and the laboratory work so that both could inform each other in a feedback relationship. And, if palaces were to be cleared, there was also a need for stabilization. With this in mind and in consort with the then Belize Department of Archaeology, we approached the United States Agency for International Development in 1988 for funds for tourism development. When these were awarded, we were able to begin to excavate palaces and range structures in earnest and, in the process, recovered abundant in situ artifactual materials that could be used for functional and temporal information. Later, with the assistance of the Tourism Development Project directed by Dr. Jaime Awe, we were able to continue the areal clearing of Caracols palaces. Even more recently, we have recovered abandonment materials in a non-palace context that serves as a second Rosetta stone for cross-dating Terminal Classic ceramics with wider portions of Belize and Mesoamerica. However, the recognition of Caracols Terminal Classic Period material expression was a long-term negotiation with the archaeological record. In spite of extensive excavation in more than 107 residential groups at the site, the recovery of easily recognizable Terminal Classic remains proved somewhat problematic. When refuse was found in the outlying settlement, it was usually comprised of utilitarian vessels that could not be tightly dated as to type (e.g. A. Chase and D. Chase 2004a: 354). Burial units that contained ceramics in the settlement area tended to contain ceramic vessels that could be labeled as generically Late Classic. Only in a few

PAGE 24

Arlen F. Chase & Diane Z. Chase cases could Terminal Classic contexts be easily identified and, even in these contexts, some of the vessels could just as easily have been labeled Late Classic. Nevertheless, the widespread occurrence of Terminal Classic pottery in isolated contexts in the settlement area indicated that Caracol was occupied during this time period and that population numbers were likely significant; minimally 25% of residential groups, and probably at least double this figure, may be assigned a Terminal Classic Period dating (A. Chase and D. Chase 2005:85). However, the bulk of the ceramics that were used in these clearly late residential groups are difficult to sort as to time; many are more comfortable with simply a Late Classic designation. This is not the case in the site epicenter. In contrast with the more temporally amorphous materials found in Caracols outlying settlement, the large numbers of ceramics that have been recovered from the floors of Caracols epicentral buildings are quite consistent among the different palaces and contain pieces that would be at home in Terminal Classic assemblages elsewhere in the Southern Maya lowlands. However, isolated and recognizable finewares and domestic Terminal Classic pottery sufficiently occurs in the core settlement to demonstrate temporal overlap with the palace remains. The Caracol research has shown that minimally two distinct sub-assemblages were in use during the Terminal Classic Period at the site. One sub-assemblage, found in Caracols residential groups, continued to use local utilitarian and fineware ceramics with only the occasional inclusion of pieces easily defined as Terminal Classic from the second subassemblage. The second sub-assemblage, associated with Caracols epicentral palaces, used distinctive fineware ceramics that in some cases are tradewar es and in other cases are local imitations of forms and decoration found elsewhere in the Maya lowlands; utilitarian wares in the epicenter are a combination of local and non-local styles. Because the one ceramic sub-assemblage is almost always associated with palaces and the other occurs more frequently in non-elite contexts, we (A. Chase and D. Chase 2004a, 2005) have referred to these different subassemblages as class-linked ceramics. The epicentral Terminal Classic palace ceramics are quite plentiful and each recovered sub-assemblage contains materials that correlate it to other epicentral sub-assemblages. In contrast, however, the sites epicentral burial ceramics that may be dated to the Terminal Classic (Figure 2a-c,e) mirror ceramics found in Late Classic contexts from throughout the site. While the late burials can be sorted out stratigraphically, ceramically, they are difficult to clearly di stinguish from more generic Late Classic Period interments. Epicentral caching practices demonstrate that polychrome ceramics, presumably imported to Caracol, also continued to be used in the Terminal Classic Period (Figure 2d). Interestingly, among the first Terminal Classic associations uncovered was the pairing of incensarios in Caracols temples, a pattern al so found in the later Postclassic Period at Santa Rita Corozal (D. Chase 1988). Paired Terminal Classic incensarios have been found in association with three Caracol temples Structures A3, A6, and B19. A double pair of incensarios also occurs with Structure A31. The incensario forms are both local and exotic. The pairings and their locations are highly symbolic. However, non-paired incensarios are associated with two palace rooms on the 18

PAGE 25

The Terminal Classic Period at Caracol Figure 2. Terminal Classic pottery from the Caracol ep icenter associated with burials and caches: a S.D. C18U-2 in Structure B5 (Joyac Cream Polychrome; Zacatel Cream Polych rome; eroded Infierno Black); b S.D. C C172C-3 in Structure I20 (Zacatel Cr eam Polychrome [2]; Batcab Red Polychrome [2]); c S.D. C117C-4 from Structure B34 (Joyac Cream polychrome; possibly Carro Modeled; Zacatel Cream Polychrome); d S.D. C4B-1 from Structure B19 (burned Tinaja Red [2]; Dant a Orange-Polychrome [3]) ; S.D. C75B-2 from Structure B66 (Zacatel Polychrome[2]). summit of Caana, along with serving wares, storage vessels, and a free-standing burner (or portable stove); the presence of incensarios in this location suggests that Caracols highest elite carried out both private and public ritual with these items. Most palace contexts include a combination of serving and storage vessels. A sealed room suite on the side of Structure B19 contained 1 large storage jar, 13 blackware tripod plates, and 4 large cups or vases (Figure 3). Specialized storing and serving vessels are associated with the Canna mid-range range structure. Serving and storing vessels, including a mocaljete, were also recovered across the B Plaza from Caana in Structure B4 and B6. Structure A39 in the Central Acropolis similarly contained serving and storage vessels, as well as a burner. Vessels associated with the C Group palace included serving vessels and 2 lid-like incensarios (see Awe 1985 for Caledonia comparisons). Besides incensarios, the Structure A6 subassemblage contains a wide range of serving vessels ranging from platters to plates to drinking cups as well as minimally three cooking pots. The vessels associated with the Barrio floors and fill blocks also contain a host of serving vessels, a large number of which were non-local, as well as cooking vessels, miniature vessels, and the lower half of a gigantic drum (Fi gure 4). Other locales within the epicentral limits have produced similar pieces from this same ceramic subassemblage that focuses on both local and 19

PAGE 26

Arlen F. Chase & Diane Z. Chase Figure 3. Terminal Classic pottery on room floor beneath fill in sealed room immediately east of Structure B19 : ag and i-r Infierno Black; h Valentin Unslipped Figure 4. Terminal Classic pottery associ ated with the Barrio complex: a Torro Gouged-Incised; b, m, u. Cameron Incized; c Azucar Impressed; d-e: Pabellon Modeled-Carved; f-g Sahcaba Modeled-Carved; h, n-q, t, x, kk-mm ww, yy Tinaja Red; i-k, Ceiba Unslipped; l. San Julio Modeled; s. Platon Punctated; r, v Martins Incised; w, y, cc, dd, ff-hh, jj. Valentin Unslipped; z Apop Modeled; aa-bb : Sombrero Appliqud; ff Encanto Striated; ii., nn-ss. Pantano Impressed. 20

PAGE 27

The Terminal Classic Period at Caracol Figure 5. Censers associated with th e front of Structure A31: a related to Lamanai Orange-Redware; b probably Miseria Appliqud; c possibly Kilikan Composite; d undesignated type. exotic pottery. During the 2006 field season, excavation of Structure A31, a nonpalace epicentral construction west of the A Group ballcourt, yielded a sub-assemblage of 21 vessels; while the majority of these vessels could be classified as either serving or utilitarian wares, four exotic incensarios were also recovered (Figure 5). These censers, clearly in a Terminal Classic abandonment context, are of relevance to issues of cross dating both within and outside of the Maya area. Although important, ceramics do not constitute the only data class that informs us about the Terminal Classic at Caracol. Other artifact classes and non-portable remains also provide important clues. In situ remains on Caracols latest residential floors include imported seafood, shell, jadeite, and obsidian. Small line-of-stone buildings of Terminal Classic date are found in many epicentral plazas at Caracol and crude substructures of similar date ring the epicenter and were used as manufacturing loci for stone tools and bone implements. In at least one case, the low epicentral remains are associated with a reset altar. It is suspected that some of Caracols other monuments, such as Stela 3, were also moved about and reset in the Terminal Classic, perhaps also accounting for a sheetcopper frog that was found in a Caracol substela cache by Satterthwaite. However, the latest date on any of Caracols monuments is A.D. 859 on Stela 10 in the A. Plaza. Based on radiocarbon dates from epicentral palace floors (A. Chase and D. Chase 2004a), occupation of Caracols epicentral palaces continued at least 40 years beyond this date, indicative of a disconnect between monument erection and the latest palace occupation. Excavations at Caracol also have revealed vibrancy to Terminal Classic construction efforts. Not only was Caana remodeled and elevated after A.D 800, but the rebuilding of Structure B20 was most likely undertaken after the cessation of monument erection. Similarly, two Late Classic tombs at the base of Structure B19 were ritually desecrated in the Terminal Classic Period, possibly the end result of internal political squabbles (D. Chase and A. Chase 2003). Other late Terminal Classic modifications are in evidence in the Barrio palace buildings, dated by late material deposited as fill. Interestingly, a number of unfinished constructions or incomplete re21

PAGE 28

Arlen F. Chase & Diane Z. Chase modeling efforts have also been recovered (Figure 6). This can be inferred from various archaeological da ta. In Barrio, the Central Acropolis, and the South Acropolis, stone robbing, as we ll as potentially contemporary rebuilding, of latest structures was in evidence. Excavation immediately south of the Caracol epicenter revealed a huge fill pile containing Terminal Classic artifacts and garbage; it is suspected that this mound of earth was destined to become the supporting platform for a late acropolis. Unused piles of cut stones, undoubtedly destined for use on remodeled constructions were recovered from in front of Structure A7 and from within the inner courtyard of the Northwest Acropolis. All of these data suggest that the end of Caracol occurred rather unexpectedly. Figure 6. Map of epicentral Caracol showing locations of unfinished constructions or of incomplete re-modeling efforts at the site at the time of abandonment Clues to the end of Caracols epicentral occupation also may be found in several deposits on Caracols latest plazas and building floors that have yielded unburied bodies and human bone. In particular, the complete body of a 6-year-old child was found in an inner doorway of a palace on the Caana summit (D. Chase and A. Chase 1998, 2000) a nd two bodies were found at the western base of Structure B28, partly beneath a stela fragment and in association with an additional 17 human mandibles. Many excavations in the Caracol epicenter have yielded isolated human bone. While a common explanation would be to ascribe the occurrence of this human material to cannibalism by the sites latest inhabitants (e.g. D. Chase and A. Chase 1982), not all the material was partial or disarticulated. Other associated artifactual remains include implements of war (D. Chase and A. Chase 2002). It is, therefore, also possible that there was a slaughter of some of the last epicentral inhabitants of the site, which would explain the abandonment material, the sheet refuse, and the scattered human remains indicative of the differential surv ival of human bone in a tropical environment. Conclusion Gaining an understanding of the Maya Terminal Classic has proven to be a long and winding road characterized by numerous potholes and detours (D. Chase and A. Chase 2004). Originally defined in terms of dates on stone monuments and type-fossil ceramic markers, the archaeological data now emphasize the great diversity in the archaeological record that exists for the Terminal Classic Period. Some sites, like Dos Pilas (Demarest 2004), collapsed before the advent of the Terminal Classic and the cessation of monument erection. Other sites, like Caracol, successfully lasted at least two generations past the latest monuments. The final remains at Caracol are indicative of a prospering elite with access to fresh fish from the sea (Teeter and Chase 2004) and exotic materials from elsewhere in Mesoamerica. That Caracol had widespread exterior contacts dur ing the Terminal 22

PAGE 29

The Terminal Classic Period at Caracol Classic is q uite clear in the archaeological record. Fine Orange ceramics (e.g. Smith 1958), all traded into Caracol, occur throughout the site epicenter, as do copies of model-carved ceramics suggesting a fusion of disparate styles on the elite level. The two smaller censers associated with Caracol Structure A31 (one called a frying pan incensario and the other a Mixtec incensario) are both pa rts of a widespread Mesoamerican ritual pa ttern that Ringle and his colleagues (1998) have tied to a Quetzalcoatl cult. These censer types are known from Chichen Itza (Brainerd 1958), Tayasal (A. Chase 1983:1097), Zacualpa (Wauchope 1948), Zaculeu (Woodbury and Trik 1953), Chinkultic (Ball 1980), Oaxaca (Caso et al. 1967), Teotihuacan (Linne 1934:111), Cholula (Acosta 1975; Joyce 1914), Teotenango (Vargas 1975), and Tula (Diehl 1983; Cobean 1990:488). Although the dating for these materials is nowhere precisely established, for the most part these kinds of censers appear to be dated to approximately A.D. 900, which also matches the inferred Caracol dating. Importantly, the Caracol Mixtec incensario is composed of Fine Orange paste, a feature that serves to emphasize its inclusion as Terminal Classic pottery. Two of the censers recovered in association with Stru cture A31 during 2006 (Figure 5) emphasize how much we have yet to learn. One large globular vessel has no known stylistic analogi es; the second large urn is a tradeware from the Lamanai area, 125 kilometer north of Caracol. While tentatively dated to the Middle Postclassic at Lamanai (Graham 1987), the piece occurs in a Terminal Classic context at Caracol. Interestingly, an almost identical censer comes from Actun Yaxteel, where a Terminal Classic date was suggested based on related material (Awe, personal communication, 2006; Awe and Helke 2000). This dating confusion goes to the heart of the Terminal Classic problem in dealing with rapid transitions in the archaeological record. For Caracol, the Terminal Classic represented a behavioral shift from the Late Classic Period (D. Chase and A. Chase 2006). Whereas trade items were widely distributed throughout the population during the Late Classic (A. Chase and D. Chase 2004b; D. Chase and A. Chase 2004), this was not the case in the Terminal Classic when the sites elite essentially hived themselves off from the general population and emphasized their status through the ostentatious use of different, and usually foreign, goods (A. Chase and D. Chase 2004a). Final abandonment materials at Caracol suggest a marked separation between the sites elite and other population segments. Archaeological data also suggest that the final abandonm ent of Caracol may have been the result of aggression in which some of the epicentral inhabitants were killed and left unburied on plaza and building floors. Whatever caused the ultimate end of Caracol at approximately A.D. 900 (based on radiocarbon dates), th e latter half of the century leading up to this abandonment was a time of great extern al contact throughout Mesoamerica. Archaeological remains indicate that Caracol was clearly a player within this broader arena. The panMesoamerican activities participated in by the sites latest epicentral inhabitants sharply contrasts with the internal disintegration of long-standing social patt erns that in the past had stressed a shared identity between the elite and the rest of Caracols population. In contrast to the Late Classic Period, the latest Caracol elite was not concerned with displaying even symbolic egalitarianism (D. Chase and A. Chase 2006). Elsewhere, we have suggested that the last elite at Caracol were incorporated into a broader political 23

PAGE 30

Arlen F. Chase & Diane Z. Chase hegem ony (A. Chase and D. Chase 2004a, 2005). So, then, what does this entire picture mean? To us, these data suggest that the Maya collapse was related to both local and wider Mesoamerican phenomena. The latest elite at Caracol was outward looking, participating in far-flung trade and actively using disparate styles an d items in new ritual patterns. At the same time, the non-elite of this era maintained a more traditional and less exotic lifestyle, making them harder to identify archaeologically without substantial effort. As archaeologists, we have problems dealing with transitional periods like the Terminal Classic. Only now, after 22 field seasons, are we beginning to understand the complexities involved in dating and modeling the Classic Maya collapse at Caracol through the long-term and timeconsuming excavation and analysis of carefully defined archaeological contexts. Reference Cited Adams, Richard E.W. 1971 The Ceramics of Altar de Sacrificios Papers of the Peabody Museum of Archaeology and Ethnology Vol 63(1), Harvard University, Cambridge. Acosta, Jorge R. 1975 La Ceramica de Cholula. in R. Pina Chan, Ed., Los Pueblos y Senorios Teocraticos Vol. 1, pp. 123-134. Instituto Nacional de Antropologa e Historia, Mexico City, D.F. Awe, Jaime J. 1985 Archaeological Investigations at Caledonia, Cayo District, Belize M.A. Thesis. Trent University, Peterborough. Awe, Jaime J. and Christophe G.B. Helmke 2000 Fashionably Late: A Postclassic Censer from the Roaring Creek Valley, Belize. in C.S. Griffith, R. Ishihara, and J.J. Awe, Eds., The Western Belize Regional Cave Project: A Report of the 1999 Field Season pp. 187-198. Department of Anthropology Occasional Paper 3. University of New Hampshire, Durham. Ball, Joseph W. 1980 The Archaeological Ceramics of Chinkultic, Chiapas, Mexico Paper 43. New World Archaeological Foundation, Brigham Young University, Provo. Berggren, Asa and Ian Hodder 2003 Social Practice, Method, and Some Problems of Field Archaeology. American Antiquity 68(3):421-434. Brainerd, George W. 1958 The Archaeological Ceramics of the Yucatan Anthropological Records 19. University of California, Berkeley. Caso, Alfonso, Ignacio Bernal, and Jorge Acosta 1967 La Ceramica de Monte Alban Memorias del INAH 13, Instituto Nacional de Antropologa e Historia, Mexico City, D.F. Chase, Arlen F. 1983 A Contextual Consideration of the Tayasal-Paxcaman Zone, El Peten Guatemala. Ph.D. Dissertation, Department of Anthropology, University of Pennsylvania. University Microfilms International No. 8406652. Chase, Arlen F. and Diane Z. Chase 1985 Postclassic Temporal and Spatial Frames for the Lowland Maya: A Background. in A. Chase and P. Rice, Eds., The Lowland Maya Postclassic pp. 9 22. University of Texas Press, Austin. 2004a Terminal Classic Status-Linked Ceramics and the Maya "Collapse:" De Facto Refuse at Caracol, Belize. in A. Demarest, P. Rice, D. Rice, Eds., The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation pp. 342-366. University of Colorado Press, Boulder. 2004b Exploring Ancient Economic Relationships at Caracol, Belize. Research Reports in Belizean Archaeology 1:115-127. 2005 Contextualizing the Collapse: Hegem ony and Terminal Classic Ceramics from Caracol, Belize. in S. Lopez Varella 24

PAGE 31

The Terminal Classic Period at Caracol and A. Foias, Eds., Ge ographies of Power: Understanding the Nature of Terminal Classic Pottery in the Maya Lowlands, pp. 73-91. BAR Monograph S1447. Oxford, England. Chase, Diane Z. 1982a Spatial and Temporal Variability in Postclassic Northern Belize Ph.D. Dissertation, Department of Anthropology, University of Pennsylvania. University Microfilms International No. 8307296. 1982b The Ikilik Ceramic Complex at Nohmul, Northern Belize. Ceramica de Cultura Maya 12: 71-81. 1985a Ganned But Not Forgotten: Late Postclassic Archaeology and Ritual at Santa Rita Corozal, Belize. in A. Chase and P. Rice, Eds., The Lowland Maya Postclassic, pp.104-125. University of Texas Press. 1985b Between Earth and Sky: Idols, Images, and Postclassic Cosmology. in M.G. Robertson and V.M. Fields, Fifth Palenque Round Table, 1983, Vol VII pp. 223-233. Pre-Columbian Art Research Institute, San Francisco. 1988 Caches and Censer wares: Meaning from Maya Pottery. in L. Lackey and C. Kolb, Eds., A Pot for All Reasons: Ceramic Ecology Revisited pp. 81-104. Laboratory of Anthropology, Temple University, Philadelphia. Chase, Diane Z. and Arlen F. Chase 1982 Yucatec Influence in Terminal Classic Northern Belize. American Antiquity 47: 596-614. 1988 A Postclassic Perspective: Excavations at the Maya Site of Santa Rita Corozal, Belize. Monograph 4. Pre-Columbian Art Research Institute, San Francisco. 1998 The Architectural Context of Caches, Burials, and Other Ritual Activities for the Classic Period Maya (as Reflected at Caracol, Belize). in Stephen D. Houston, Ed., Function and Meaning in Classic Maya Architecture pp. 299-332. Dumbarton Oaks, Washington, D.C. 2000 Inferences about Abandonment: Maya Household Archaeology and Caracol, Belize. Mayab 13:67-77. 2002 Classic Maya Warfare and Settlement Archaeology at Caracol, Belize. Estudios de Cultura Maya 22:33-51. 2003 Secular, Sagrado, y Revisitado: La Profanacion, alteracion, y reconsagracion de los Antiguos Entierros Mayas, in A. Ciudad Ruiz, M.H. Ruz Sosa, and M.J.Iglesias Ponce de Leon, Eds., Antropologa de la Eternidad: La Muerte en la Cultura Maya pp. 255-277, Publicacin 7,Sociedad de los Estudios Mayas, Madrid. 2004 Archaeological Perspectives on Classic Maya Social Organization from Caracol, Belize. Ancient Mesoamerica 15:111-119. 2006 Framing the Maya Collapse: Continuity, Discontinuity, Method, and Practice in the Classic to Postclassic Southern Maya Lowlands. in G. Schwartz and J. Nichols, Eds., After Collapse: The Regeneration of Complex Societies, pp. 168-187. University of Arizona Press, Tucson. C obean, Robert H. 1990 La Ceramica de Tula, Hidalgo Instituto Nacional de Antropologa e Historia, Mexico, D.F. Cobos, Rafael 1999 Fuentes Historicas y Arqueologia: Convergencias y Divergencias en la Reconstruccion del Period Clasico Terminal de Chichen Itza. Mayab 12:58-70. 2004 Chichen Itza: Settlement and Hegemony during the Terminal Classic Period. in A. Demarest, P. Rice, D. Rice, Eds., The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation pp. 517-544. University of Colorado Press, Boulder. Coe, Michael D. 2005 The Maya Seventh Edition. Thames & Hudson Inc., New York. Culbert, T. Patrick 1973 The Maya Downfall at Tikal. in T.P. Culbert, Ed., The Classic Maya Collapse 25

PAGE 32

Arlen F. Chase & Diane Z. Chase pp. 63-92. University of New Mexico Press, Albuquerque. Demarest, Arthur 2004 After the Maelstrom: Collapse of the Classic Maya Kingdoms and the Terminal Classic in Western Peten. in A. Demarest, P. Rice, D. Rice, Eds., The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation pp. 102-124. University of Colorado Press, Boulder. Diehl, Robert A. 1983 Tula Thames & Hudson Inc., New York. Graham, Elizabeth 1987 Terminal Classic to Early Historic Period Vessel Forms from Belize. in P.M. Rice and R.J. Sharer, Eds., Maya Ceramics, Vol. 1, pp. 73-98. International Series 345. BAR, Oxford. Inomata, Takeshi 1997 The Last Day of a Fortified Maya Center: Archaeological Investigations at Aguateca, Guatemala. Ancient Mesoamerica 8:337-351. Inomata, Takeshi and Triadan, Daniela 2000 Craft Production by Classic Maya Elites in Domestic Settings: Data from Rapidly Abandoned Structur es at Aguateca, Guatemala. Mayab 13:57-66. Joyce, Thomas A. 1914 Mexican Archaeology London. Linne, Sejourne 1934 Archaeological Researches at Teotihuacan, Mexico Ethnographic Publication 1. Museum of Sweden, Stockholm. Pendergast, David 1986 Stability Through Change: Lamanai, Belize, from the Ninth to the Seventeenth Century. in J.A. Sabloff, and E.W. Andrews V, Eds., Late Lowland Maya Civilization: Classic to Postclassic pp. 223-249, University of New Mexico Press, Albuquerque. 1990 Up From the Dust: The Central lowlands Postclassic as Seen from Lananai and Marco Gonzales. in R. Clancy and P.D. Harrison, Vision and Revision in Maya Studies pp. 169-177. University of New Mexico Press, Albuquerque. Ringle, William M., Tomas Gallareta Negron, and George J. Bey III 1998 The Return of Quetzalcoatl: Evidence for the Spread of a World Religion during the Epiclassic Period. Ancient Mesoamerica 9(2):183-232. Schiffer, Michael B. 1987 Formation Processes of the Archaeological Record University of New Mexico, Albuquerque. Sheets, Payson D. 1998 Place and Time in Activity Area Analysis: A Study of Elevated Contexts Used for Artifact Curation at the Ceren Site, El Salvador. Revista Espanola de Antropologia Americana 28:63-98. Sheets, Payson D., Harriet F. Beaubien, Marilyn Beaudry, Andrea Gerstle, Brian McKee, C. Dan Miller, Hartmut Spetzler, and David B. Tucker 1990 Household Archaeology at Ceren, El Salvador. Ancient Mesoamerica 1:81-90. Smith, Robert E. 1955 Ceramic Sequence at Uaxactun, Guatemala Publication 20, Middle American Research Institute, Tulane University, New Orleans. 1958 The Place of Fine Orange Pottery in Mesoamerican Archaeology. American Antiquity 24:151-160. 1971 The Pottery of Mayapan Papers of the Peabody Museum of Archaeology and Ethnology 66. Harvard University, Cambridge. Teeter, Wendy and Arlen F. Chase 2004 Adding Flesh to Bones: Using Zooarchaeology Research to Answer BigPicture Questions. Archaeofauna 13:155172. Tozzer, Alfred M. 1957 Chichen Itza and Its Cenote of Sacrifice: A Comparative Study of Contemporaneous Maya and Toltec Memoirs of the Peabody Museum of 26

PAGE 33

The Terminal Classic Period at Caracol 27 Archaeology and Ethnology 11-12. Harvard University, Cambridge. Vargas P., Ernesto 1975 La Ceramica. in R. Pina Chan, Ed., Teotenango: El Antiguo Lugar de la Muralla pp. 189-265. Direccion de Turismo, Gobierno del Estado de Mxico, Mxico. D.F. Wauchope, Robert 1948 Excavations at Zacualpa, Guatemala Middle American Research Institute Publication 14. Tulane University, New Orleans. Webb, Ronald W. and Kenneth G. Hirth 2003 Xochicalco, Mexico: The Abandonment of Households at an Epiclassic Urban Center. in T. Inomata and R.W. Webb, Eds., The Archaeology of Settlement Abandonment in Middle America pp. 29-42. University of Utah Press, Salt Lake City. Woodbury, Richard B. and Aubrey S. Trik 1953 The Ruins of Zaculeu, Guatemala (2 volumes). United Fruit Company, Richmond.

PAGE 34

3 FIGHTING THE INEVITABLE: THE TERMINAL CLASSIC MAYA OF THE UPPER ROARING CREEK VALLEY Jaime J. Awe and Christophe G.B. Helmke Archaeological investigations in the upper Roaring Creek Valley have revealed a complex Maya landscape of settlements and ritual cave sites. This paper reports on data recorded at these settlements, which, also provides excellent evidence for local continuities and discontinuities in Classic Maya culture and the uneven changes and transformations of the Terminal Classic period. This is par ticularly true of sites such as Cahal Uitz Na and Chaac Mol Ha, and Pooks Hill. Several caves and rockshelters have also been investigated in the valley and they reflect a peak in local ritual cave utilization between 800 and 900 AD. The Terminal Classic period brought with it great challenges to the traditional Classic period style of life for the Maya and many communities were unable to cope, which is evident by the large-scale d ecline and abandonment of many formerly successful city-states. Depopulation in this area was likely prompted by this broader set of social and political stresses and partly brought on by environmental degradation. Introduction The Terminal Classic period, which unfolded during the ninth century AD, was a time of challenges and transformations in Maya civilization. It witnessed the decline and eventual abandonment of many great cities like La Milpa an d Pusilha. It brought with it political changes that led to the dissolution of the power previously enjoyed by paramount hereditary rulers, and saw a rise in community-based political systems. The Terminal Classic also brought changes in external relationships, and to the economic engine that fueled Classic period Maya civilization. More importantly, these changes challenged the Mayas ability to successfully sustain their previous way of life. The Belize River Valley region of the Maya Lowlands provides a fertile environment for the study of the uneven changes and transformations of the Terminal Classic period. For example, while sites such as Cahal Pech (Awe 1992, 2006) and Buenavista (Ball and Taschek 2004; Taschek and Ball 2004) were being abandoned, others such as Xunantunich (AD 820 849) (Lecount et al. 2002; Leventhal and Ashmore 2004; Yaeger 2000; Helmke et al. 2006) and Caracol (AD 793 859) (Chase, Grube and Chase 1991; Martin and Grube 2000; Chase and Chase 2004) witnessed a veritable revival in regal activity during the Terminal Classic as marked by the dedication of carved monuments. Elsewhere, ceramic and settlement data indicate that Baking Pot (Audet and Awe 2004; Audet 2007) and Pacbitun (Healy et al. 2004; Healy et al. 2007) continued to thri ve, even though regal activities in the epicenters appear diminished. Data recorded at several other settlements along the Belize River system also provide excellent evidence for local continuities and discontinuities in Classic period Maya culture. This is particularly true of sites along the Roaring Creek, a major tributary of the Belize River (Fig. 1). Setting of the Upper Roaring Creek Valley and Previous Investigations From its base at the Hidden Valley (or Thousand Foot) falls, the Roaring Creek (Fig. 2) flows northward through a narrow alluvial valley delineated to the east and west by steep karstic pr ecipices carved from Research Reports in Belizean Archaeology Vol. 4, 2007, pp.28-42. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 36

Figure 1. Map of the Belize Valley showing the location of major sites and the Roaring Creek valley. Figure 2. Perspective view of the Upper Roaring Creek valley. The settlement survey area completed as of 2003 is shown in grey and the floodplain devoid of housemounds in yellow. All ancient Maya architectural remains are depicted to scale in red. Topography is based on a Digital Elevation Model derived from Synthetic Aperture Radar data (obtained under academic license from Global TerrainTM). Map, data processing and perspective view by AndrewBevan. 29

PAGE 37

Jaime Awe & Christophe G. B. Helmke otherwise low-lying and rolling lim estone hills. The valley eventually widens near the location of a minor centre known as Chaac Mool Ha (Fig. 2), at a point where the stream begins to meander toward its confluence with the Belize River. For research purposes we chose to divide the Roaring Creek valley in to two areas. The Upper Valley extends southward from Chaac Mool Ha to the source of the creek at the base of the Mountain Pine Ridge. The Lower Valley extends northward from Chaac Mool Ha to the confluence of the Roaring Creek and the Belize River. The history of archaeological research in the upper Roaring Creek spans a little more than a decade (1993 to present). Most of this research has been conducted under the auspices of the Western Belize Regional Cave Project, a branch of the Belize Valley Archaeological Reconnaissance Project under the direction of the senior author. The primary focus of the BVAR investigations in the upper Roaring Creek addresses the role of caves in ancient Maya society and the interrelationship between surface and subterranean sites in antiquity (see Awe 1998, Helmke et al. 2004). Summaries of previous research in the Roaring Creek are provided elsewhere (Awe et al. 1998: 225; Helmke 2000a: 293). Prehistoric Settlements in the Upper Roaring Creek It has been over a decade since BVAR archaeologists began their study of the upper Roaring Creek valley. As part of this regional study we have recorded more than a dozen cave sites, a major center, a minor center, and several square kilometers of settlement and plazuela groups. The largest surface site known for the upper valley is the major center Cahal Uitz Na that covers approximately 3 hectares (Fig. 3). valley is the major center Cahal Uitz Na that covers approximately 3 hectares (Fig. 3). This site is comparable in size to Cahal Pech (2.8 ha), Minanha (3.6 ha), and Altun Ha (4.1 ha). The monumental epicenter of Cahal Uitz Na comprises several pyramidal temples, palatial and administrative range structures, a ball court, and a series of 10 limestone and slate monuments (see Awe and Helmke 1998; Conlon and Ehret 1999). A 250-meter long causeway connects the central precinct to the en trance of a cave we designated as Actun Nakbe. Seven of the monuments recorded at Cahal Uitz Na are made from slate and the other three are from Figure 3 Map of Cahal Uitz Na limestone. Two of the limestone monuments flank the axial outset stair of one of the principal pyramidal temple structures, suggesting that these served as panels. One slate monument and one of the 30

PAGE 38

Terminal Classic Maya of the Upper Roaring Creek Valley 31 panels are decorated but not inscribed. The limestone panel (Fig. 4a) has what appears to be a biconically drilled hole at its center, and the slate monument (Fig. 4b) has a crude petroglyphic face carved on it. The face on the latter monument is similar to several simple faces that we discovered carved on flowstone at the entrance to Actun Uayazba Kab, a cave approximately 500 meters due west of Cahal Uitz Na. This congruence in rock art between the two sites helps to anchor the petrographic corpus of Actun Uayazba Kab to some point in the Late Classic (see Helmke and Awe 1998; Helmke et al. 2003). In all probability, Cahal Uitz Na served as the local regal centre for the polity centered on the upper Roaring Creek valley, but despite this role, it may have been subservient to one of the larger centers to the west in the greater Belize River valley region (see Helmke and Awe in press). Survey to the north and east of Cahal Uitz Na recorded the minor centre of Chaac Mool Ha and several other formal plazuela groups. The fairly regular spacing of the plazuelas at radii of c. 500 m suggests that they may have exerted some level of influence over landhold ings and the smaller settlement units in the valley (see Helmke et al. 2004). Chaac Mool Ha is located within the alluvial valley bottom at a distance of c. 3.6 km to the north of Cahal Uitz Na, a distance that is consistent with other smaller satellite sites within the Belize Valley (Driver & Garber 2 004). The site is comprised of two groups linked by a 240 m long causeway and access to the north group appears to have been gained via a ballcourt that served as the formal entrance in keeping with patterns observed at several other sites, but most notably at Baking Pot (see Aimers 1997; Ferguson 1999b). The overall size of the site as well as comparable architectural features and configurations make Chaac Figure 4. Monuments of Cahal Uitz Na. a) Limestone monument, with biconical perforation at the middle. Set as the eastern panel to the axial outset stair of Structure 5. The monument measures 1.4 m at its widest. b) Slate monument with crudely carved petroglyphic face. The monument measures 2.09 m at its longest. Drawings by Christophe Helmke. Mool Ha intimately comparable to the site of X-ual-canil situated within the eastern periphery of Cahal Pech (see Iannone 2003). Among the larger plazuela groups is the site known as Pooks Hill, located 1 km to the northwest of Chaac Mool Ha. Pooks Hill (Helmke 2006a, b), in contrast to Chaac Mool Ha, is located on the flank of a hill that overlooks the valley and creek to the east. This shrine-focused settlement cluster also has relatively la rge architecture that includes a rounded eastern funerary shrine, three range structures and a sweatbath (Fig. 5) (Helmke 2006a, b; Helmke and Awe 2005).

PAGE 39

Jaime Awe & Christophe G. B. Helmke Figure 5. Plan of the Pooks Hill plazuela group showing the locations of the Eastern shrine as well as the sweatbath structure. Plan and survey by Christophe Helmke. Based on our detailed excavations of the rounded eastern shrine and the sweatbath we were able to establish that they are among the latest buildin gs erected at Pooks Hill, and likely constructed in the second half of the tenth century. Both structures are also reminiscent of Terminal Classic architecture in the western Peten and northern Yucatan Peninsula; a date supported by the cultural remains recovered by excavations at both loci by Helmke (2006a, b, Helmke and Awe 2005). Results of our settlement survey recorded a density of just over a hundred mounds per square kilometer in the upper Roaring Creek valley (Helmke et al. 2004). Pottery collected from the surface of most house-mounds, and those from controlled excavations at Cahal Uitz Na, Pooks Hill and mounds near the entrance to Actun Tunichil Mucnal (Song et al. 2000) predominantly dated to the latter half of the Late Classic period. Using accepted models of Maya demography, we estimate that between 4500 and 7500 people may have been living in the upper Roaring Creek around AD 750-850 (see Helmke et al. 2004). Subterranean Sites in the Upper Roaring Creek Valley As we indicated above, more than a dozen subterranean sites have been discovered in the upper Roaring Creek. These sites are represented by both caves and rockshelters (Awe et al. 1998). During the last decade we conducted intensive research in four caves (Actun Tunichil Mucnal, Actun Yaxteel Ahau, Laberinto de las Tarantulas and Ac tun Nak Beh), and two rockshelters (Actun Uayazba Kab, and Uayak Na). Investigations in Actun Tunichil Mucnal (Awe et al. 2005; Moyes 2005; Gibbs 2000) identified four areas with evidence for prehistoric Maya utilization. It was also found that the cave was used for ritual purposes from Early Classic to Terminal Classic times (c. AD 200 950). Evidence for Early Classic usage was limited to the two principal entrance areas (the Eastern Entrance and the Sinkhole Entrance). This contrasts significantly with the interior of the cave where four loci, designated as the Upper Entrance Chamber (Griffith 1998), the Sinkhole Tunnels (Helmke 1999), the Stelae Chamber (Awe et 32

PAGE 40

Terminal Classic Maya of the Upper Roaring Creek Valley al. 2005) and the Main Cha mber (Moyes and Awe 1998; Moyes 2005), contained extensive evidence for Late to Terminal Classic use. Adjoining the Eastern Entrance is a series of small chambers that were collectively named the Upper Entrance Chamber. The material culture and deposits from this area had been affected by recent looting, but our inves tigations revealed the presence of a small subsidiary chamber of difficult access that had escaped plunder. This small chamber (known as the Hideaway) was found to contain 13 Late to Terminal Classic ceramic vessels, including 4 bowls, the fragmented remains of 5 dishes, 3 ollas (ritually killed or fragmented) and the ring base of another vessel (see Griffith 1998). Excavations revealed a ceramic sequence spanning back to at least the Early Classic as well as an abundance of Terminal Classic ceramic sherds within the uppermost matrices suggesting that the small masonry platform built within the chamber is coeval to the late usage of this area. The second area, known as the Sinkhole Tunnels, is a maze of small passages that pass through breakdown and an extinct phreatic system. Despite the diminutive size of the passages all exhibited evidence of human usage as attested by ceramic and human remains. Much like the Upper Entrance Chamber, the ceramic assemblage of the Sinkhole Tunnels spans the entirety of the Classic period, though Late and Terminal Classic types comprise c. 60 % of the total assemblage (see Helmke 1999). Set further within the cave, ca. 325m away from the Upper Entrance Chamber and ca. 195m from the Sinkhole Entrance is the third area, designated as the Stelae Chamber. The area acquired its name on account of the presence of two slate monuments that are supported in upright position by several broken speleothems at the center of a ledge overlooking the chamber. At the base of, and dispersed around, the monuments we recovered the fragmentary remains of at least two Roaring Creek Red dishes, two ollas, one Molded-carve d vase (Fig. 6), two obsidian blades, plus a small slate slab decorated with a crude representation of the central Mexican deity Tlaloc. It has been argued elsewhere (Awe et al. 2005) that the slate monuments were carved in the form of a stingray spine and an obsidian blood-letter, and that the placement of stela-like monuments in caves appears to be a Terminal Classic tradition in western Belize. Representations of Tlaloc, despite their limited Early Classic representations at central Peten sites, also seem to share a similar, late, temporal distribution in Belize as supported by the recent discovery of Terminal Classic Tlaloc masks adorning Structure B5 at Caracol (see Awe 2003; Ishihara et al. 2006). Finally the presence of a Molded-carved vase in the Stelae Chamber suggests that most if not all of the Late Classic ceramic vessels actually date to the late facet of the Sp anish Lookout complex and as such date to the Terminal Classic, the same date we attribute to the erection of the slate monuments. The furthest area within the cave to exhibit cultural activity is the Main Chamber, the entrance to which is located c. 450 m from the Eastern Entrance and c. 120 m from the Stelae Chamber. Contained within the Main Chamber are the skeletal remains of 14 individuals, more than 200 ceramic vessels, three or more metates, several groundstone adzes, faunal remains, a stone rasp and a few ceramic ocarinas. Children represent half of the skeletons; the other seven include both adult males and females. All evidence indicates that these individuals were sacrificed in association with agricultural fertility rituals (Awe et al. 2005). The ceramics found within the Main Chamber fall typologically within the two 33

PAGE 41

Jaime Awe & Christophe G. B. Helmke facets of th e Spanish Lookout complex as defined by Gifford ( 1976) and complicates conclusive dating of the materials (see also Chase & Chase 2004). To help clarify temporal assignation two radiocarbon dates (AA 57315, and AA57316) were submitted to and processed by the Arizona AMS Lab in Tucson, which produced dates of 1202 + 32 or 710 950 AD, and 1191 + 31 or 720 960 AD in the 2 Sigma range. Both samples derived from charcoal collected in the Main Chamber and help to confirm the Terminal Classic usage of this locus in Actun Tunichil Mucnal. Figure 6. Artifacts of the Stelae Chamber, Actun Tunichil Mucnal. a) & d) Roaring Creek Red with thumb impressions. b) Possible Mountain Pine Red vessel. c) Cayo Unslipped olla neck. e) Ahkutu Molded-carved vase. f) Slate artifact. Located approximately 500 meters south of Actun Tunichil Mucnal, Actun Uayazba Kab is actually a large rockshelter with a narrow tunnel that winds its way into a series of small and dark recessed chambers. Investigat ions at this site recorded a complex corpus of prehistoric Maya cave art that includes pictographs, petroglyphs, partly sculpted faces as well as modified speleothems (Helmke and Awe 1998; Helmke et al. 2003). While much of these art forms have a long antiquity in Mesoamerica, their temporal distribution in the Maya area clusters significantly during the Terminal Classic period and finds support with the crossdating of the slate monument from Cahal Uitz Na mentioned above. In 1997 and 1998 we initiated several excavation units within the two entrance chambers of the site (Griffith 1998; Ferguson and Gibbs 1999). These investigations uncovered numerous burials (Gibbs 2000) in simple cists. This pattern of interment is in marked contrast to the pattern noted in Actun Tunichil Mucnal and Yaxteel Ahau, and likely is indicative of important differences in the treatment of the dead within rockshelters and cave sites. The ceramic artifacts reco vered in association with the burials, in excavation units and from ledges high above the cave entrance spanned from the end of the Early Classic to the Terminal Classi c period and not only coincides with the stylistic temporal placement of cave art in Actun Uayazba Kab, but strengthens this dating. Investigations in Actun Yaxteel Ahau, Laberinto de las Tarantulas and Actun Nak Beh reflect relatively similar patterns of cave use to Actun Tunichil Mucnal. The major difference between them is that while Yaxteel Ahau and Tunichil Mucnal both contain numerous skelet ons that were likely the victims of human sacrifice, similar remains and evidence are practically absent at the other two cave sites. The fact that the two are large river cave s may further help to tie these two cave sites together in terms of emic classifications of cave sites and the 34

PAGE 42

Terminal Classic Maya of the Upper Roaring Creek Valley types of activities that were deem ed suitable for different sites. One specific similarity that Laberinto de las Ta rantulas shares with Actun Tunichil Mucnal is that the former also contains a slate monument within the dark zone of the site (Awe et al. 2005). Beside this there is great uniformity in their cultural assemblages, they all reflect initial usage at the start of the Early Classic, and considerable intensification of ritual activity during the Terminal Classic period. Prehistoric Ceramics from the Upper Roaring Creek Valley During the last decade we have had an opportunity to cond uct analysis of the prehispanic ceramics from both the subterranean and surface sites that we investigated in the upper Roaring Creek valley. At Cahal Uitz Na, Fergusons (1999a) excavations in the ballcourt as well as Ehret and Conlons (1999) testing of the plaza platforms noted that initial settlement of this site likely began around the end of the Late Preclassic period. The same is true for the earliest utilization of the Actun Uayazba Kab rockshelter. While a few sherds of the Late Prec lassic have also been found at Pooks Hill, it is more likely that this site has its foundation in the ensuing Early Classic (see Helmke 2000a, 2006b). Having said this, however, it is important to note that only few diagnostic types have actually been identified for the Late Preclassic period and in no instance have they been isolated stratigraphically. Wherever we recovered Late Preclassic pottery, it was always mixed with Early Classic materials. Among the diagnostic Preclassic types identified were specimens from the Sierra, Hill Bank, Stumped Creek, and Chan Pond ceramic groups (see Gifford 1976 for ceramic group descriptions). Unlike Late Preclassic types, Early Classic ceramics appear to have a much wider distribution in th e valley. Indeed, just about every major surface and subterranean site investigated has produced evidence for either permanent settlement or limited utilization during this time frame. The most identifiable Early Cl assic diagnostic types recorded in the upper Roaring Creek valley include basal-flanged dishes of the Dos Arroyos, Minanha, and Balanza ceramic groups, although the latte r only occurs in relatively low frequency. Specimens of the brown Pucte group represent other gloss ware pottery. Unslipped pottery, predominantly in jar forms, is particularly present in vessels of the Succotz ceramic group. The 8th century Late Classic period witnessed continued growth in the Roaring Creek valley. Settlements like Cahal Uitz Na increased in size, and cave sites reflect limited but consistent utilization for ritual purposes. During this time polychrome pottery of the Saturday Creek and Saxche groups, and monochrome dishes and bowls of the Mountain Pine, Sotero, and Teakettle Groups represent the ceramic materials of choice in burials, caches and offerings. Despite the earlier Classic period developments in the valley, they pale in comparison with the dynamism and pervasiveness of the Terminal Classic. At the onset of the ninth century, the upper Roaring Creek Maya communities continued to expand. Considerable new architecture was added to sites such as Pooks Hill, Cahal Uitz Na and like ly Chaac Mool Ha. It is about this time too that most of the monuments at Cahal Uitz Na were erected, and that the sweatbath was added to the main plaza at Pooks Hill, while the eastern shrine of the site was remodeled to a circular ground plan. The Terminal Classic ceramic assemblage for the Roaring Creek valley is also considerably more complex and diverse than in preceding phases. They include a broad range of types that fit perfectly within 35

PAGE 43

Jaime Awe & Christophe G. B. Helmke 36 the local Belize Valley Spanish Lookout and New Town complexes, plus several other types of foreign origin or inspiration. Local Terminal Classic ceramics represented in the assemblage include unslipped jars of the TuTu Camp and Cayo ceramic groups, and several red monochromes of the Garbutt Creek and Vaca Falls ceramic group. Some large basins of the Roaring Creek Red type occur with both ring and pedestal base and some are decorated in the tradition of Daylight Orange: Dark night Variety of the New Town complex. The consistent cooccurrence of Roaring Creek and Daylight Orange ceramics at all sites in the Roaring Creek indicates that these are wholly contemporaneous. This conclusion naturally runs counter to their original classification (Gifford 1976) in which Daylight Orange specimens were assigned to the New Town complex. A variety of Roaring Creek Red has also been identified in which the interiors of large basins are slipped brown. We have established these as the Kanan Variety. Ceramics from the Vaca Falls group dominate the Late to Terminal Classic ceramic assemblages from both surface and cave sites and include rare forms that anticipate those of the later Buk-style ceramics from Lamanai and nearby Chau Hiix. The specimens in question recovered from Pooks Hill exhibit pre-slip incised and punctuated designs above the shoulder of the vessels and in all cases these display prominent pedestal bases. Duck Run incised pyriform vases with pedestal bases have also been noted in burials and fragments of others have been foun d within caves. In addition, a heretofore unclassified type of ceramics is represented by calcite-tempered, large, sharply incurv ing bowls that are slipped variegated br own and/or black on the interior, while the exterior is left unslipped and embellished by four squared trefoils along the rim, produced by excising the background. These have tentatively been grouped under a new type that we designate as Ik Excised. Of indisputable foreign origin are fragments of at least one or more vessels of Tohil Plumbate that were discovered in Actun Tunichil Mucnal. Other likely imports are various vessels of the Cabrito Cream Polychrome type that were discovered in Tarantul a Cave, in Structure ATM-M-1 at the entrance to Actun Tunichil Mucnal, and within the Upper Entrance Chamber of Actun Tunichil Mucnal proper (Helmke 2000b; Song et al. 2000), as well as a large pedestal vessel akin to Buk-style censers from Lamanai (Pendergast 1981; Graham 1987; Graham and Pendergast 1989). Discovered in Actun Yaxteel Ahau, the Buk-style vessel is particularly interesting in terms of its regional distribution and dating (see Awe and Helmke 2000) (Fig. 7). Both Pendergast and Graham (Pendergast 1981; Graham 1987; Graham and Pendergast 1989) have dated these censers to the Middle Postclassic period and have noted th at their distribution is centered predominantly in northern Belize and the offshore islands. The Yaxteel Ahau specimen is the first reported from a cave site where most associated artifacts date to the Terminal Classic period rather than the Postclassic. This pattern also seems to fit other Buk-style censers that were recently discovered in association with Structure A31 at Caracol (Chase and Chase 2006; Arlen Chase personal communication 2006). The deposit in which these Caracol specimens were discovered has been assigned a Terminal Classic date suggesting that the introduction of Buk-style censers in fact reaches back much earlier to the Terminal Classic (see also Graham 1987). Other vessels of either foreign origin and/or inspiration are a number of Moldedcarved, red-slipped vases. Found at several caves (Actun Tunichil Mucnal and Actun

PAGE 44

Terminal Classic Maya of the Upper Roaring Creek Valley Figure 7. The Buk-style censer recovered from Actun Yaxteel Ahau. a) Photo of the vessel (photo by Megan Bassendale). b) Drawing of the vessel (drawing by Christophe Helmke). Nak Beh) and surface sites (Pooks Hill), these vessels are a hallmark of the Terminal Classic period (c. 830 950 AD) across the entire Maya lowlands, and although used mainly in feasting activities, they are also found in other ritual contexts including caves and burials. The vessels exhibit rich and detailed decorative panels depicting both the real and the supernatural worlds. While no Fine Orange moldedcarved vases have been found in the Roaring Creek, the specimens discovered can be assigned to two major types. The first is the Sahcaba type (Smith and Gifford 1966), which may represent imports to the area. The second type is considerably more prevalent, to which we have attributed the designation of Ahkutu Molded-carved, in keeping with the vessel type assigned glyphically to these vases by the ancient Maya (see Helmke 2001). It is notable that the fragmentary molded-carved vases reported from Footprint and Chanona Caves (Graham et al. 1980) are of this same exact type, indicating that there is a regional pattern to the depositio n of these vases in caves. The favored themes of Ahkutu vases include the presentation to a ruler of a captive taken in raids and the conjuring of deified ancestors as part of vision quest ceremonies (Helmke 2000c). Glyphic texts just below the rim of the vessels all make reference to the names and titles of a single ancient owner, part of whose name was Olom (see Helmke 2003; 2006a). Conclusion Archaeological investigations in the upper Roaring Creek Valley have revealed a complex Maya landscape of settlements and ritual cave sites. Presently, the earliest securely dated material from the valley 37

PAGE 45

Jaime Awe & Christophe G. B. Helmke com es from the end of the Late Formative period (c. AD 100 300). This is followed by gradual but steady growth throughout the Classic period with greatest population density occurring sometime in the ninth century AD. Investigations of subterranean sites in the valley reflect a similar peak in local ritual cave utilization between 800 and 900 AD. This increase in ritual activity was accompanied by the first attestation of human sacrifices for the area while the seizure of captives is celebrated in the iconographic programs depicted on contemporary molded-carved vases. While intensifying cave usage could possibly reflect little more than a symptom of a larger local population, we beli eve that increased ritual activity at such places were more likely prompted by a broader set of social and political stresses emerging at this time and partly brought on by environmental degradation. The Terminal Classic period brought with it great challenges to the traditional Classic period style of life for the Maya. That many communities were unable to cope is evident by the large-scale decline and abandonment of many formerly successful city-states. We believe that the Roaring Creek valley Maya responded, in part, to the trials and tribulations of the Terminal Classic period with increased ritual activity in their sacred cave sites. That they managed to survive a little longer than many of their neighbors is evident in their continued development during the Terminal Classic period. Eventually, however, they were simply fighting the inevitable for sometime during Early Postclassic times, they too abandoned their homeland. References Cited Aimers, James J. 1997 Preliminary Investigations of Architecture in Plaza 2 of Group I at Baking Pot. Belize Valley Archaeological Reconnaissance Project: Progress Report of the 1996 Field Season edited by Jaime J. Awe and James M. Conlon, pp. 21-45. Department of Anthropology, Trent University, Peterborough. Audet, Carolyn M. 2007 Political Organization in the Belize Valley: Excavations at Baking Pot, Cahal Pech and Xunantunich. Unpublished PhD dissertation, Department of Anthropology, Vanderbilt University, Nashville. Audet, Carolyn M. and Jaime J. Awe 2004 Whats Cooking at Baking Pot: A Report of the 2001 to 2003 Seasons. Research Reports in Belizean Archaeology Vol. 1: 49-59. Awe, Jaime J. 1992 Dawn in the Land Between the Rivers: Formative Occupation at Cahal Pech, Belize and its Implication for Preclassic Developments in the Maya Lowlands. Unpublished PhD dissertation, Institute of Archaeology, University of London, London. 1998 The Western Belize Regional Cave Project: Objec tives, Context, and Problem Orientation. The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 1-22. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. 2003 The Monumental Masks of Structure B5 at Caracol. NICH Culture Vol. 1 (1): 5-7. 2006 Maya Cities and Sacred Caves: A Guide to the Maya Sites of Belize Cubola Productions, Benque Viejo del Carmen. Awe, Jaime J., Cameron Griffith and Sherry Gibbs 2005 Cave Stelae and Megalithic Monuments in Western Belize. In the Maw of the Earth Monster, edited by James E. Brady and Keith M. Prufer, pp. 223-248. University of Texas Press, Austin. Awe, Jaime J. and Christophe G.B. Helmke 1998 Preliminary Report on the Reconnaissance of Cahal Uitz Na, Roaring Creek Valley, Cayo District, Belize. The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by 38

PAGE 46

Terminal Classic Maya of the Upper Roaring Creek Valley Jaime J. Awe, pp. 205-221. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. 2000 Fashionably Late: A Postclassic Censer from the Roaring Creek Valley, Belize. The Western Belize Regional Cave Project: A Report of the 1999 Season edited by Cameron S. Griffith, Reiko Ishihara and Jaime J. Awe, pp. 187-198. Department of Anthropology, Occasional Paper No. 3. University of New Hampshire, Durham. Awe, Jaime J., Christophe G.B. Helmke and Cameron S. Griffith 1998 Archaeological Reconnaissance in the Roaring Creek Valley: Caves, Rockshelters, and Settlements. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season edited by Jaime J. Awe, pp. 223-244. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. Ball, Joseph W. and Jennifer T. Taschek 2004 Buenavista del Cayo: A Short Outline of Occupational and Culture History at an Upper Belize Valley Regal-Ritual Center. The Ancient Maya of th e Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 149-167. University Press of Florida, Gainesville. Chase, Arlen F. and Diane Z. Chase 2004 Terminal Classic Status-linked Ceramics and the Maya Collapse: DeFacto Refuse at Caracol, Belize. The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 342-366. University Press of Colorado, Boulder. 2006 Trying to Understand the End: Continued Investigation of Small Structures in and near Caracols Epicenter: 2006 Field Report of the Caracol Archaeological Project Department of Anthropology, University of Central Florida, Orlando. Chase, Arlen F., Nikolai Grube and Diane Z. Chase 1991 Three Terminal Classic Monuments from Caracol, Belize, Research Reports on Ancient Maya Writing Number 36, Center for Maya Research, Washington, DC. Conlon, James M. and Jennifer J. Ehret 1999 Survey at the Major Center of Cahal Uitz Na, Cayo District, Belize. The Western Belize Regional Cave Project: A Report of the 1998 Field Season, edited by Jaime J. Awe, pp. 33-44. Department of Anthropology, Occasional Paper No. 2. University of New Hampshire, Durham. Driver, W. David and James F. Garber 2004 The Emergence of Minor Centers in the Zones between Seats of Power. The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 287-304. University Press of Florida, Gainesville. Ehret, Jennifer J. and James M. Conlon 1999 Plaza Excavations at Cahal Uitz Na, Cayo District, Belize: A Preliminary Report. The Western Belize Regional Cave Project: A Report of the 1998 Field Season edited by Jaime J. Awe, pp. 53-68. Department of Anthropology, Occasional Paper No. 2, University of New Hampshire, Durham. Ferguson, Josalyn M. 1999a Salvage Excavations of the Ballcourt at Cahal Uitz Na. In The Western Belize Regional Cave Project: A Report of the 1998 Field Season edited by Jaime J. Awe, pp. 45-52. Department of Anthropology, Occasional Paper No. 2. University of New Hampshire, Durham. 1 999b The Ballcourts at Baking Pot, Belize: Analysis of the Ballgame at a Maya Civic Centre. Unpublished MA thesis, Department of Anthropology, Trent University, Peterborough. Ferguson, Josalyn M. and Sherry Gibbs 1999 Report on the 1998 Excavations at Actun Uayazba Kab, Roaring Creek Valley, Belize. The Western Belize Regional Cave Project: A Report of the 1998 Field Season edited by Jaime J. Awe, pp. 112-145. Department of Anthropology, Occasional Paper No. 2, University of New Hampshire, Durham. 39

PAGE 47

Jaime Awe & Christophe G. B. Helmke Gibbs, Sherry A. 1998 Human Skeletal Remains from Actun Tunichil Mucnal and Actun Uayazba Kab. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 71-95. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. 2000 An Interpretation of the Significance of Human Remains from the Caves of the Southern Maya Lowlands. Unpublished MA thesis, Department of Anthropology, Trent University, Peterborough. Gifford, James C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley. Memoirs of the Peabody Museum of Archaeology and Ethnology Vol. 18, Harvard University, Cambridge. Graham, Elizabeth A. 1987 Terminal Classic to Early Historic Period Vessel Forms from Belize. Maya Ceramics: Papers from the 1985 Maya Ceramic Conference, Vol. 1 edited by Prudence M. Rice and Robert J. Sharer, pp. 73-98. BAR International Series 345 (i), Oxford. Graham, Elizabeth A., Logan McNatt, and Mark A. Gutchen 1980 Excavations in Footprint Cave, Caves Branch, Belize. Journal of Field Archaeology Vol. 7 (2): 153-172. Graham, Elizabeth A. an d David M. Pendergast 1989 Excavations at the Marco Gonzalez Site, Ambergris Cay, Belize, 1986. Journal of Field Archaeology, Vol. 16 (1): 1-16. Griffith, Cameron S. 1998 Excavations and Salvage Operations in Actun Tunichil Muknal and Actun Uayazba Kab, Roaring Creek Valley, Belize. The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 23-36. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. Healy, Paul F., Bobbi Hohmann and Terry G. Powis 2004 The Ancient Maya Center of Pacbitun. The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 207-227. University Press of Florida, Gainesville. Healy, Paul F., Christophe G.B. Helmke, Jaime J. Awe and Kay S. Sunahara 2007 Survey, Settlement, and Population History at the Ancient Maya Site of Pacbitun, Belize. Journal of Field Archaeology Vol. 32 (1): 17-39. Helmke, Christophe G.B. 1999 Exploration and Investigations of the Sinkhole Tunnels, Actun Tunichil Muknal, Belize. The Western Belize Regional Cave Project: A Report of the 1998 Field Season edited by Jaime J. Awe, pp. 146-165. Department of Anthropology, Occasional Paper No. 2, University of New Hampshire, Durham. 2000a Pooks Hill 1, Op erations 1 through 3: Salvage Excavations of Structure 4A, Roaring Creek Valley, Cayo District, Belize. The Western Belize Regional Cave Project: A Report of the 1999 Field Season edited by Cameron S. Griffith, Reiko Ishihara and Jaime J. Awe, pp. 287-330. Department of Anthropology, Occasional Paper No. 3, University of New Hampshire, Durham. 2 000b An Analysis of the Ceramic Remains from Structure ATM-M1, Roaring Creek Valley, Belize. The Western Belize Regional Cave Project: A Report of the 1999 Season edited by Cameron S. Griffith, Reiko Ishihara and Jaime J. Awe, pp. 239261. Department of Anthropology, Occasional Paper No. 3. University of New Hampshire, Durham. 2000c Terminal Classic Molded-carved Ceramics from Structures 193, 198 and 199, Baking Pot, Cayo District, Belize. The Western Belize Regional Cave Project: A Report of the 1999 Field Season, edited by Cameron S. Griffith, Reiko Ishihara & Jaime J. Awe, pp. 15-42. Department of Anthropology, Occasional Paper No. 3, University of New Hampshire, Durham. 2001 The Last Supper: Competitive Feasting and the Terminal Classic Moulded-carved Ceramic Tradition in the Central Maya 40

PAGE 48

Terminal Classic Maya of the Upper Roaring Creek Valley Lowlands. Unpublished MA Dissertation. Institute of Archaeology, University College London, London. 2003 The 2002 Season Investigations at Pooks Hill, Belize. Papers from the Institute of Archaeology Vol. 14: 119-128. 2006a A Summary Report of the 1999-2002 Seasons of Archaeological Investigations at Pooks Hill, Cayo District, Belize. Research Reports in Belizean Archaeology Vol. 3: 173-191. 2006b A Report of the 2005 Season of Archaeological Investigations at Pooks Hill, Cayo District, Belize. The Belize Valley Archaeological Reconnaissance Project: A Report of the 2005 Field Season edited by Christophe G.B. Helmke and Jaime J. Awe, pp. 39-92. Institute of Archaeology, National Institute of Culture and History, Belmopan. Helmke, Christophe G.B. and Jaime J. Awe 1998 Preliminary Analysis of the Pictographs, Petroglyphs and Sculptures of Actun Uayazba Kab, Cayo District, Belize. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 145-203. Department of Anthropology, Occasional Paper No. 1. University of NewHampshire, Durham. 2005 The Ancient Maya Sweatbath at Pooks Hill. Belize Today Vol. 5 (1): 2426. in press Organizacin te rritorial de los antiguos mayas de Belice Central: confluencia de datos arqueolgicos y epigrficos. Mayab No. 19. Helmke, Christophe G.B., Jaime J. Awe and Cameron S. Griffith 2003 El Arte Rupestre de Belice. Arte rupestre de Mexico oriental y Centro America, edited by Martin Kunne and Matthias Strecker, pp. 97-117. Gebr. Mann Verlag, Berlin. Helmke, Christophe, Andrew Bevan and Jaime Awe 2004 Everyday Maya: Life along the Roaring Creek Valley. Current World Archaeology Nr. 4: 10-16. Helmke, Christophe G.B., Pierre Robert Colas and Jaime J. Awe 1998 Comments on the Typology, Epigraphy and Iconography of the Actun Tunichil Muknal Vase and Belize Valley ModeledCarved Vessels. The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 97-144. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. Helmke, Christophe, Nikolai Grube and Jaime Awe 2006 A Comprehensive Review of the Carved Monuments and Hieroglyphic Inscriptions of Xunantunich, Belize. The Belize Valley Archaeological Reconnaissance Project: A Report of the 2005 Field Season edited by Christophe G.B. Helmke & Jaime J. Awe, pp. 143-186. Institute of Archaeology, National Institute of Culture and History, Belmopan. Iannone, Gyles 2003 Rural Complexity in the Cahal Pech Microregion: Analysis and Implications. Perspectives on Ancient Maya Rural Complexity edited by Gyles Iannone and Samuel V. Connell, pp. 13-26. Cotsen Institute of Archaeology, University of California, Los Angeles. Ishihara, Reiko, Karl A. Taube and Jaime J. Awe 2006 The Water Lily Serpent Stucco Masks at Caracol, Belize. Research Reports in Belizean Archaeology Vol. 3: 213-223. LeC ount L., Jason Yaeger, Richard M. Leventhal, and Wendy Ashmore 2002 Dating the Rise and Fall of Xunantunich, Belize: A Late and Terminal Classic Lowland Maya Secondary Center. Ancient Mesoamerica Vol. 13 (1): 41-63. Leventhal, Richard M. and Wendy Ashmore 2004 Xunantunich in a Belize Valley Context. The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 168-179. University Press of Florida, Gainesville. Martin, Simon and Nikolai Grube 41

PAGE 49

Jaime Awe & Christophe G. B. Helmke 42 2000 Chronicle of the Maya Kings and Queens: Deciphering the Dynasties of the Ancient Maya Thames & Hudson, London. Moyes, Holley 2005 Cluster Concentrations, Boundary Markers, and Ritual Pathways: A GIS Analysis of Artifact Cluster Patterns Actun Tunichil Muknal, Belize. In the Maw of the Earth Monster, edited by James E. Brady and Keith M. Prufer, pp. 269-300. University of Texas Press, Austin. Moyes, Holley and Jaime J. Awe 1998 Spatial Analysis of Artifacts in the Main Chamber of Actun Tunichil Muknal, Belize: Preliminary Results. The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 37-69. Department of Anthropology, Occasional Paper No. 1. University of New Hampshire, Durham. Pendergast, David M. 1981 Lamanai 1980: Summary of Excavation Results, 1974-1980. Journal of Field Archaeology, Vol. 8 (1): 29-53. Smith, Robert E. and James C. Gifford 1966 Maya Ceramic Varieties, Types and Wares at Uaxactun: Supplement to Ceramic sequence at Uaxactun, Guatemala. Middle American Research Institute Publication Vol. 28 (4): 125-174. Song, Rhan-Ju, Peter A. Zubrzycki and Christophe G.B. Helmke 2000 A Report on the 1999 Excavations of Structure ATM-M1 in the Cahal Uitz Na Periphery, Roaring Creek Valley, Belize. The Western Belize Regional Cave Project: A Report of the 1999 Season, edited by Cameron S. Griffith, Reiko Ishihara and Jaime J. Awe, pp. 209-238. Department of Anthropology, Occasional Paper No. 3. University of New Hampshire, Durham. Taschek, Jennifer T. and Joseph W. Ball 2004 Buenavista del Cayo, Cahal Pech, and Xunantunich: Three Centers, Three Histories, One Central Place The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research, edited by James F. Garber, pp. 191-206. University Press of Florida, Gainesville. Yaeger, Jason R. 2000 Changing Patterns of Social Organization: The Late and Terminal Classic Communities at San Lorenzo, Cayo District, Belize. Unpublished PhD dissertation, Department of Anthropology, University of Pennsylvania, Philadelphia.

PAGE 50

Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 43-54 Copyright 2007 by the Institute of Archaeology, NICH, Belize 4 ANCIENT MAYA DROUGHT CULT: LATE CLASSIC CAVE USE IN BELIZE Holley Moyes The ancient Maya of Belize used caves since the Early Middle Preclassic period. Although they were always ritual venues, work at Chechem Ha Cave in western Belize illustrate s that ritual practices in caves changed over time in both form and intensity. A distinct change in practice occurred between the Early and Late Classic periods. Beginning in the Late Classic, ritual practice entailed the placement of large partial or whole vessels, mostly jars, in caves. Preferred activity areas were in remote areas of the sites or in spaces with restricted access that were small and difficult to reach. This change in practice correlates with a dry period that begins approximately A.D. 700 and lasts over 500 years. This paper presents evidence that activities noted in ancient Maya cave sites dating to the beginning of this period represent a Late Classic drought cult. Introduction Since the 1970's there have been a steadily increasing number of archaeological, iconogra phic, and epigraphic studies regarding ancient Mesoamerican caves (Brady and Prufer 2005). The most important collective finding of these studies is the establishment of caves as sacred space and their exclusive us e as ritual venues by Pre-Columbian people. The second major finding is that caves are integrally connected with water control and agricultural fertility. This can be demonstrated in deep antiquity as early as the Olmec civilization (1200200BC). For instance, the water/fertility theme is illustrated in the El Rey monument from Chalcatzingo. Although it has been variously interpreted, most scholars agree that the image represents a man who is a king or ancestor sit ting on a cloud scroll within a cave. Mist or smoke emanates from the entrance. Clouds rain on the scene and corn and other vegetation is depicted on the surface on top of the cave (Angulo 1987:133-158; Grove 1984: 110-111; Reilly 1994:78-79). The water/fertility theme is pervasive throughout Mesoamerica and is found much later in the Preclassic Maya murals from the San Bartolo that date to the first century B.C. (Saturno et al. 2005). The mural on the north wall illustrates a creation event in which maize tamales and gourds of water are being handed out of the entrance of a cave, probably the cave of origin. The association suggests that both the first maize and primordial water originated in caves. The ancient Maya Maize God is depicted at the mouth of the cave accepting the offerings. The 17th century Popol Vuh story of the Maya creation tells us that the Maize God is intimately connected with caves, as he is an underworld denizen. The archaeological record demonstrates that in the Classic period many of the cave rites were water-related. This is not surprising when we consider that a number of deities thought to reside in caves were associated with agricultural success. For instance, the Maya Rain God Chac is depicted sitting in his cave/house on a Classic period vase (Coe 1978:78, no.11). A reified example of this is found at the cave of La Pailita in Guatemala where a life size sculpture of Chac sits on his throne in the cave's interior (Graham 1997). Activity areas in caves also suggest that water was an important feature of cave rites. In his survey of 48 caves in the Yalahau area of Quintana Roo, Dominique

PAGE 51

Holley Moyes 44 Rissolo (2001; 2005) noted that many of the caves in his survey c ontained interior water features such as intermittent pools. Both rock art and architec tural modifications tended to be associated with these features and art from the cave of Pak Che'n contained rain-god motifs. In the formal spatial analysis conducted by the Western Belize Regional Cave Project in the Main Chamber of Actun Tunichil Muknal by myself and Jaime Awe, we found that 51% of the artifact assemblage was placed in intermittent pools (Moyes 2001; Moyes and Awe 1998; 2000). So, given that cave rites were strongly associated with agricultural success entailing the control of rainfall, we might expect cave ritual to be affected by climatic factors. The use of rituals as a type of technology to anticipate and minimize agricultural risk has been demonstrated among the modern Maya by David Freidel and Justine Shaw (2000). Based on the 43 ethnographic and ethnohistoric cases studied, they reported that where agriculture was risky, primarily due to water availability, ritual increased. What this implies is that ethnogr aphically it has been documented that water availability is one of the primary concerns of modern Maya agriculturists and that ritual investment is somewhat linked to agricultural risk based on environmental factors. Archaeological investigations at Chechem Ha Cave coupled with recent paleo-environmental da ta suggest that a prolonged dry period during the Late Classic period provoked a ritual response that can be detected in cave usage. It is demonstrated by changes in ritual practice between the Early and Late Classic periods that produced distinctly different signatures in the archaeological record. These ritual transformations are demonstrated by variation in the use of space, differences in the condition and proven ience of artifacts, and changes in ritual use-intensity. What I refer to as use-intensity is closely related to what is known in anthropological contexts as "ritual density." This examines why some societies or historical periods have more ritual than othe rs (Bell 1997:173). In the archaeological record, use-intensity can be studied by identifying a material signature that correlates with ritual activity. The Cave Chechem Ha has been under investigation by the Western Belize Regional Cave Project (WBRCP) directed by Dr. Jaime Awe since 1998. It was discovered in 1989 and opened as a tourist venue in 1995. The ca ve is located in western Belize on the we stern bank of the Macal River near the Guatemalan border (Awe et al. 2005). The surface sites closest to the cave are Tipu located 5.6km to the north, Las Ruinas 6km to the northwest, and Minanh located 6.5km to the southwest. Chechem Ha is a complex cave system that contains over 300m of tunnels. The system is 198m in length and consists of two primary conduits, Tunnel 1 and Tunnel 2 (Figure 1). There are four elevated side passages and eleven shelves located from 37m above the Tunnel 1 floor. Artifacts are found throughout the entire caves system in niches and alcoves alo ng the tunnel floors, in the elevated passages, and on all eleven shelves. Based on radiocarbon dates, the site was used from the Early Middle Preclassic (1100-820 B.C.) possibly as early as 1300 B.C. to the Late Classic period (Moyes 2002, 2004). These are the earliest radiocarbon dates for Maya ritual cave use in the lowlands and are contemporaneous with the earliest settlers in the Belize Valley (Garber et al. 2004). The latest date for usage based on AMS dating of charcoal is A.D. 960. This is important because the cave's use spanned the 2,000-year

PAGE 52

Late Classic Cave Use in Belize 45 Figure 1 Map of Chechem Ha tunnel system. development and collapse of the Classic period socio/political system, providing a broad temporal perspective on ritual cave use within a single site. The entrance of the cave was blocked with medium to large sized boulders no later than A.D. 960. This correlates roughly with the Classic Maya Collapse and agrees with termination events from the two nearest surface sites Las Ruinas de Arenal (Taschek and Ball 1999) and Minanh (Iannone 2001) occurring approximately A.D. 850. Methods Changes in ritual practice over time in the cave were evaluated using detailed mapping, implementing a rigorous dating program, and conducting test pitting as well as broad horizontal ex cavations. Two proxy measures were used to evaluate the site's use-intensity--the number of ceramic entities and charcoal flecks present on the excavated levels of Chamber 2. Chamber 2, located 134 m from the cave entrance deep within the dark zone (see Figure 1), exhibited the heaviest utilization and deepest subsurface deposits. The chamber could only be accessed using an artificial light source. Both archaeological and iconographic evidence demonstrates that the Maya used wood torches to light their way in caves (Morehart 2 005; Morehart et al. 2005). This is born out in archaeological studies. Charcoal flecks from torches as well as unburned wood fragments (Prufer 2002:614; Gann 1925:111) are found in virtually every utilized cave dark zone in Mesoamerica. A large partially burned torch fragment was located at Chechem Ha on Ledge 10 and charcoal flecks were present throughout the s ite. All evidence from the archaeological record indicates that torches were the sole source of fuel used in caves throughout the ancient Maya era. In his study of plant remains at Chechem Ha, Morehart (2002) reported that all charcoal flecks collected from surface deposits were Pinus species. Experime nts conducted with ocote pine torches collected from the forest around the cave illustrated that they produced a steady rain of charcoal flecks that could be used as a proxy to evaluate

PAGE 53

Holley Moyes 46 fluctuations in th e amount of activity occurring in the cave over time. The use-intensity measure must be distinguished from estimates of frequency of use because, depending on the signature that is used, it is impo ssible to distinguish whether the charcoal in the deposits is a result of more or fewer discrete rituals, the participation of more or fewer individuals in a fixed number of rituals, the result of rituals of a longer or shorter duration, or the effects of a change in ritual practice. Two methods are helpful in resolving this issue. One method is to use multiple proxies and the other is to employ an indirect signature that is less likely to be linked to the ritual itself. In this case, charcoal flecks dropping from torches serve as a good indirect proxy measure because torches were a functional necessity for entering the dark areas of the cave. Numbers of charcoal flecks could then be compared with the numbers of ceramics from particular temporal periods. Two field seasons were spent mapping the modern surface finds of the cave (Moyes, 2002, 2004, 2005, 2006). A test-pitting program conducted in the third season demonstrated that the cave had deep subsurface deposits containing wood charcoal that enabled us to date the deposits. A total of 44 AMS dates for the site were processed at the University of Arizona Accelerator Mass Spectrometry (AMS) Laboratory. Dates were calibrated using Oxcal 3 and are reported at the 2 sigma probability. Areas with no subsurface deposits were dated using ceramic chronologies based on James Gifford's (1976) type-variety-mode system (Jaime Awe personal communication 1999; James Aimers personal communication 2003; Joseph Ball personal communication 1999; Joseph Ball and Jennifer Taschek personal communication 2005, Ishihara 2000; Kay Sunahara personal communication 2001). These data were supplemented with AMS dates derived from charcoal present on the surface. There were a total of 1901 ceramic sherds, whole, or partia l vessels that after in situ reconstruction represented 563 vessels (these were re-tabulated from preliminary counts in 2006), 470 of which were typed for chronology. In the fourth season a broad 2m x 8m horizontal excavation was conducted in the chamber. This area was crucial in understanding early cave use because it is necessary to traverse this area to enter the deeper parts of the system. The excavation consisted of 18 natural and cultural layers that were excavated to bedrock (Moyes 2004; Moyes 2006). Using Photomapping, an in-field GIS technology developed by Mark Aldenderfer and Nathan Craig (Aldenderfer and Craig 2002; Craig 2000; Craig et al. 2006), the numerous charcoal flecks on each excavated surface in Chamber 2 was counted. Results The ceramic chronology from both surface and subsurface c ontexts determined that 4% were from the Jenny Creek complex of the Middle Preclassic period, 5% from the Barton Creek complex of the Late Preclassic Period, only a single sher d dated to the later part of the Late Preclassic Mount Hope Complex, 39% to the combined Terminal Preclassic/Early Classic or Floral Park/Hermitage period, 2% to the early Late Classic Tiger Run complex, and 51% to the Late Classic Spanish Lookout complex. Based on these figures one might at first glance expect that the cave underwent it's most intensive usage in Late Classic period but this is not a safe assumption. Because ceramics are portable objects brought into the cave as offerings or as an element of a ritual they are a dire ct proxy and are likely to be affected by changes in ritual practices. Proxy measures based on charcoal revealed that there was very little use-

PAGE 54

Late Classic Cave Use in Belize 47 intensity in the Late Classic period. AMS dates from the Chamber 2 excavations determined that Levels 1-13 dated to the Maya era. Level 1 was the modern surface. Levels 2-6 dated to th e Early Classic period, Level 7 to the Late Preclassic, and Levels 813 to the Early Middle Preclassic. The number of charcoal fl ecks per level ranged from 265 to 8,244. A correction to the raw data was made because the excavated surface areas were not of identical size on each level. As the excavations progressed the cave walls curved inward and spatial areas narrowed toward the bottom. Therefore the numbers of flecks were divided by the area of the excavated level to produce a use-intensity index. Figure 2 illustrates the variation between levels. Beginning in the Early Middle Preclassic period there is heavy usage. This falls off after 820 B.C. There is little usage in Level 7. This level dated to 350-40 B.C. but is bracketed by much earlier and later dates. It suggests that the ch amber received lowintensity usage for a long period of time. Level 6 is the time of heaviest use-intensity and dates from the Terminal Preclassic to the beginning of the Early Classic period 210-420 A.D. Over the Early Classic useintensity wanes and by the Late Classic period there is very little use-intensity of the chamber. When we view the two data sets side by side we find both agreement and discrepancies between the ceramic chronology and use-intensity (Figure 3). In the Middle Preclassic period charcoal counts are high but ceramic counts are at their lowest belying the amount of site usage at this early period. Both sets illustrate low amounts of both charcoal and ceramics in the Late Preclassic period, which suggests sparse usage of the site. During the Early Classic period the ceramic counts and charcoal indices increase considerably suggesting an intensification of use. In the Late Classic period it is just the opposite. The ceramic data suggests that there is a greater intensity of use than indicated by the amount of charcoal found on the surface. This presents a quandary because we know that the chamber was indeed used in the Late Classic period. There are caches of Late Classic ceramics on the Chamber 2 surface and on Ledge 9 located on the south side of the chamber. Because there is so little charcoal on the surface and so many ceramic entities within the site, these data suggest that ceramic vessels may have taken on a greater importance in rituals conducted during this time. Figure 2 Graph of use-intensity index shows numbers of charcoal flecks by excavated level divided by surface area of excavated space. Chart moves from youngest to oldest deposits Between the Early and Late Classic periods, the use of space changed within the cave on a global scale. The maps in Figure 4 illustrate the spatial variation of diagnostic ceramics for these two periods. The Early Classic distributions occurred in three major areas: the passage le ading to Chamber 2, Ledge 10 located in Chamber 2, and in the deepest passages of Tunnel 1. In the Late Classic period Ledge 10 fell out of use but the other 10 ledges were used for the first time in the cave's history. In addition to the ledges, artifacts were also placed in the elevated passages, crawl spaces, and other spatially restricted spaces and even those

PAGE 55

Holley Moyes 48 areas that were difficult to access. While some of these spaces were used in earlier periods they clearly became the activity areas of choice in the Late Classic. Figure 3 Chart shows the percentages of ceramic and charcoal data sets for each major temporal period. There is less than 1% of the total number of charcoal flecks on the surface of Chamber 2 which is curios considering that over 50% of the ceramics in the cave dated to the Late Classic period Finally, the condition of the ceramic assemblage changed somewhat between the Early and Late Classic periods. In the Early Classic whole or partially intact vessels were rare while in the Late Classic there were 51 fully intact vessels and numerous others that were partially intact. Most of these are jar forms. Figure 5 illustrates the typical condition of the assemblages. Based on five dates collected from the ledges and from Elevated Passage 3 it appears that this change in practice occurred sometime after A.D. 680 (Table 1). To summarize we see that during the Late Classic period over 51% of the caves entire ceramic assemblage was imported into the site over a relatively short time span between A.D.680 and A.D.960. During this time the deposition of ceramic vessels, mostly jars, was a major focus of cave rites. Activity areas shifted from the tunnel floor to ledges and other difficult to access, remote and hard to reach areas. Although there were a large number of objects imported into the site in the Late Classic period, the use-intensity data suggest that relatively small groups did this. Additionally the number of participants in these rites was limited due to the choice of activity areas such as small high ledges or other restricted spaces that could accommodate only a few people. Not only this, but participants could not have spent long periods of time in the cave because long rites would be expected to produce greater amounts of charcoal rain from torches. These findings suggest that in the Late Classic period groups that entered the cave conducted rituals of short duration in very restricted areas. Figure 4 Maps of cave illustrating differences in the distributions of ceramics be tween the Early (top) and Late (bottom) Classic periods. In the Late Classic ceramics are more widely di stributed throughout the site. Ledges and high level passages become the preferred activity areas at this time.

PAGE 56

Late Classic Cave Use in Belize 49 Figure 5 Photo illustrating differences between typical Early and Late Classic cave ceramic assemblages. a) Early Classic assemblage is highly fragmented whereas b) Late Classic assemblage contains more whole or partial vessels. AZ Lab # Period Area Radiocarbon Calibrated Date Alternative Age 2 Sigma Probabilities AA57293 Late Classic Ledge 6 1187 AD 720-960 AA57288 Late Classic Ledge 4 1210 AD 690-900 AD760-900(85.4%) AA59754 Late Classic Ledge 6 1224 AD 680-900 AA59753 Late Classic EP3 1239 AD 680-890 AA57291 Late Classic Ledge 7 1244 AD 680-890 Table 1 Late Classic AMS dates calibrated using OxCal 3. A Regional Phenomenon The placement of intact and partially intact Spanish Lookout-style jars (Late Classic) in remote or difficult to access areas is not limited to Chechem Ha but is a widespread phenomenon in Belizean cave sites during this period. Some caves sites were only used in the Late Classic period and all known Maya cave sites in Belize show evidence of Late Classic usage. Large Late Classic whole or partially intact vessels are reported by the Western Belize Regional Cave Project in the Entrance Chambers and Main Chamber at Actun Tunichil Muknal (Griffith 1998; Moye s 2001), Barton Creek Cave (Mirro and Mi rro 2001), Flour Camp Cave (Ishihara 2001), Laberinto de las Tarantulas (Helmke et al. 1999), and Yaxteel Ahau (Mirro and Halperin 2000; Owen and Gibbs 1999). These are often located in remote or hard to reach areas. In his survey of 48 caves in southern Belize, Prufer (2002) also found numerous caves containing partial or whole Late Classic vessels placed in difficult-to-access or remote areas. The pattern was also noted at Edwardo Quiroz Ca ve in the Chiquibul region of southern Cayo District (Pendergast 1971) and at Rio Frio Cave in the Mountain Pine Ridge (Pendergast 1970). In the Caves Branch area Graham and her colleagues (1980) illustrated a num ber of Late Classic jars and bowls from the high ledge at Footprint Cave and numerous intact Late Classic jars and dishes were found in Alcoves I and II at Actun Polbiche in the Sibun Hills (Pendergas t 1974). Patricia McAnany and her colleagues (2003) also

PAGE 57

Holley Moyes 50 working in the Sibun area reported large Late to Terminal Classic vessels found in inaccessible areas at Pottery Cave. A Late Classic Drought Cult Given the relationship of caves to agriculture and particularly water-related rites we would expect that ritual practices might be integrally linked to climatic conditions, especially ra infall. One of the best methods for deriving local paleoclimate reconstructions is the analysis of speleothems from caves. Speleothems, which include both stalactites and stalagmites, are formed by calcite deposition that accumulated as bands or rings (Ford and Williams 1989; Hill and Forti 1997). In areas with annual wet and dry seasons the bands suggest yearly events much like tree rings. Bands can be dated using radiocarbon or Uranium-series (U-series) dating. Dates are determined for specific bands by bracketing absolute dates (Schwarcz and Rink 2001). Because this is a yearly cycle, there is good potential for the recovery of fine-scaled temporal climatic data. James Webster (2000) conducted a paleo-climate study of relative rainfall availability using a speleothem from the Vaca Plateau in western Belize located approximately 5km from Chechem Ha Cave. Webster evaluated the thickness and frequency of bands, their color, luminescence, and isotopic ratios ( 18O). His data demonstrated that from A.D. 700 to A.D. 1225 there was a long dry period with spikes in dryness one standard deviation below average occurring at approximately A.D. 809, 928, 1126, and 1206. The spikes in dryness correlate with the Classic Maya Collapse occurring in the eighth century. The timing of the dry cycle coincides well with changes in ritual practice at Chechem Ha cave beginning about A.D. 680. Given the deep antiquity of Maya beliefs that associate caves with gods that control water; in a time of environmental crisis caves would have been the logical ritual venue to propitiate these deities. The ubiquitous presence of intact or partial jars in these assemblages is also suggestive because of the cosmological connotations associat4ed with jars. Ethnographic evidence reported by Tarn and Prechtel (1986:176) from Lake Atitlan in Guatemala point out that Maria Castellana, a female creatrix, is associated with the moon, which is thought to hold rainwater. The moon is envisioned as a large olla. During rainy season it turns sideways until the water spills out. Images in the Dresden codex suggest that this is an ancient belief. On page 74 is an illustration of the primordial flood that Karl Taube (1992:100, 1995: 69-71) associated with the original Maya creation event. An old woman, named in the codex as Chac Chel and identified by Taube as Goddess O, the Moon Goddess, hangs in the sky. In her hands there is an inverted olla pouring water. Because she is pictured in the cave paintings at Naj Tuni ch, Brady (1989:47-49) has long argued that th is female deity is associated with caves. In Late Classic cave rituals water jars likely relate to Maya mythology that equates rain and water deities with these vessels. Conclusion The evidence suggests that ritual practice in caves in the Late Classic period represents a drought cult that is pervasive on a regional scale. When we consider the relationship between caves, water control and agricultural fertility it is hardly surprising that caves would become the focus of ritual during a time of climatic stress. At Chechem Ha Cave the phenomena can be temporally circumscribed beginning after A.D. 680 and ending by A.D. 960, about the time that the local sites were abandoned. The correlation between the

PAGE 58

Late Classic Cave Use in Belize 51 changes in practice and the onset of a prolonged dry period can hardly be accidental. The depositi on of large jars in hard to reach areas also suggests that rituals to propitiate rain deiti es became more costly and elaborated at this time and that the offering itself takes on greater importance. This information is important in understanding how the dry period affected the lives of Maya people. The study not only defines the behaviors that produced the artifact record but implies a cognitive aspect as well. These data confirm that there was in fact a ritual response to the environmental stress. The dry period was not an abstract concept imperceptible to Maya people but rather was a perceived problem. The blocking and possible termination of the cave following this last period of usage and the area's abandonment suggests that there was in fact a perceived ritual failure. Acknowledgments I would like to thank Dr Jaime Awe, principal investigator for the Western Belize Regional Cave Project (WBRCP). Appreciation is extended to the Morales and Plytez families, crewmembers Jim Aimers, Anthony Beardall, Jim Conlon, Jennifer Ehret, Cameron Griffith, Christophe Helmke, Reiko Ishihara, Mike Mirro, Grant Polley, Connie Price, and Kay Sunahara. Mark Aldenderfer provided computer equipment for the project and Nathan Craig was instrumental in implementing the photo mapping technique. Thanks to Joseph Ball and Jennifer Taschek for identifying problematic ceramics. Various aspects of the project were funded by an NSF dissertation improvement grant to Dr. Ezra Zubrow, the Foundation for the Advancement of Mesoamerican Studies, the Cave and Karst Conservancy, the Cave Research Foundation, and the Mark Diamond fund at the University at Buffalo. References Cited Aldenderfer, Mark S. and Nathan Craig 2002 A GIS-based in-field data recording system for archaeological excavation. Paper presented at the 13th Annual Workshops in Archaeometry University at Buffalo. Angulo, Jorge V. 1987 The Chalcatzingo Reliefs: an Iconographic Analysis. In Ancient Chalcatzingo edited by David C. Grove, pp. 133-158, University of Texas Press, Austin. Awe, Jaime J., Sherry Gibbs, and Cameron Griffith 2005 Stelae and Megalithic Monuments in the Caves of Western Belize. In In the Maw of the Earth Monster: Mesoamerican Ritual Cave Use, edited by James E. Brady and Keith M. Prufer, pp. 223-248, University of Texas Press, Austin. Bell, Catherine 1997 Ritual: Perspectives and Dimensions. Oxford University Press, New York. Brady, James E. 1989 Investigation of Maya Ritual Cave Use with Special Reference to Naj Tunich, Peten, Guatemala. Unpublished Ph.D. Dissertation, Department of Anthropology, University of Califor nia, Los Angeles. Brady, James E. and Keith Prufer 2005 Maya Cave Archaeology: A New Look at Religion and Cosmology. In Stone Houses and Earth Lords: Maya Religion in the Cave Context, edited by Keith M. Prufer and James E. Brady, pp. 365-380, University Press of Colorado, Boulder. Coe, Michael 1978 Lords of the Underworld. Princeton University Press, Princeton. Craig, Nathan 2000 Real-Time GIS Construction and Digital Data Recording of the Jiskairumoko Excavation, Peru. The Society for American Archaeology Bulletin 18(1):24-28. Craig, Nathan, Mark Aldenderfer, and Holley Moyes 2006 Multivariate Visualization and Analysis of Photomapped Artifact Scatters, Journal of Archaeological Science 20:1-10. Ford, Derek C., and P. W. Williams 1989 Karst Geomorphology and Hydrology. Unwin, Hyman, Boston. (London)

PAGE 59

Holley Moyes 52 Freidel, David and Justine Shaw 2000 The Lowland Maya Civilization: Historical Consciousness and Environment. In The Way the Wind Blows: Climate, History, and Human Action, edited by Roderick J. McIntosh, Joseph A. Tainter, and Susan Keech McIntosh, pp. 271-300, Columbia University Press, New York. Gann, Thomas 1925 Mystery Cities: Exploration and Adventure in Lubaantun. London, Duckworth. Garber, James F., M. Kathryn Brown, Jaime J. Awe, and Christopher J. Hartman 2004 Middle Formative Prehistory of the Central Belize Valley: An Examination of Architecture, Material Culture, and Sociopolitical Change at Blackman Eddy. In The Ancient Maya of th e Belize Valley: Half a Century of Archaeological Research, edited by James F. Garber, pp. 25-48. University Press of Florida, Gainesville. Gifford, James C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley. Memoirs of the Peabody Museum of Archaeology and Ethnology 18, Harvard University, Cambridge, MA. Graham, Elizabeth, Logan McNatt, and Mark A. Gutchen 1980 Excavations in Footprint Cave, Belize. Journal of Field Archaeology (7):153-172. Graham, Ian 1997 Discovery of a Maya Ritual Cave in Peten, Guatemala. Symbols (Spring):28-31. Griffith, Cameron 1998 Excavations and Salvage Operations in Actun Tunichil and Actun Uayazba Kab, Roaring Creek Valley, Belize. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 37-70, Dept. of Anthropology, Occasional Paper No. 1, University of New Hampshire, Durham. Grove, David C.and Susan Gillespie 1984 Chalcatzingo's Portrait Figurines and the Cult of the Ruler. Archaeology 37 (4):2733. Helmke, Christophe G. B., Cameron S. Griffith, Michael J. Mirro 1999 The 1996 and 1998 Investigations at Laberinto De Las Tarantuals, Cayo District, Belize. In The Western Belize Regional Cave Project: A Report of the 1998 Field Season, edited by Jaime J. Awe, pp. 205221, Department of Anthropology, Occasional Paper No. 2, University of New Hampshire, Durham. Hill, Carol A., and Paolo Forti 1986 Cave Minerals of the World. National Speleological Society, Huntsville, Alabama. Iannone, Gyles 2001 Fall of the House of Minanh: A Case of Late Classic Political Disruption in West Central Belize. In Archaeological Investigations in the North Vaca Plateau, Belize: Progress Report of the Third (2001) Field Season, Edited by Gyles Iannone, Ryan Primrose, Adam Menzies, and Lisa McParland, pp.127-133, Peterborough, Dept. of Anthropology, Trent University. Ishihara, Reiko 2000 An Investigation of the Ancient Maya Ritual Cave Activity at Actun Chechem Ha, Cayo District, Belize. Unpublished Honors Thesis, University of Tsukuba. 2001 Report on a Reconnaissance Mission of Actun Uchentzub (Flourcamp Cave), Cayo District, Belize. In The Western Belize Regional Cave Project: A Report of the 2000 Field Season, edited by Cameron Griffith, Reiko Ishihara, and Jaime J. Awe, pp. 241-250, Department of Anthropology, Occasional Paper No. 4, University of New Hampshire, Durham. McAnany, Patricia A., Kimberly A. Berry, and Ben S. Thomas 2 003 Wetlands, Rivers, and Caves: Agricultural and Ritual Practice in Two Lowland Maya Landscapes. In Perspectives on Ancient Maya Rural Complexity, Monograph 49, edited by Gyles Iannone and Samuel V. Connell, pp. 71-82, The Cotsen Institute of Archaeology, University of California, Los Angeles. Mirro, Michael and Christina Halperin 2000 Archaeological Investigations on Ledge 1 of Actun Yaxteel Ahau, Roaring Creek

PAGE 60

Late Classic Cave Use in Belize 53 Valley, Cayo District, Belize. In The Western Belize Regional Cave Project: A Report of the 1999 Field Season, edited by Cameron S. Griffith, Reiko Ishihara, and Jaime J. Awe, pp. 263-280, Department of Anthropology, Occasional Paper No. 3, University of New Hampshire, Durham. Mirro, Michael and Vanessa Mirro 2001 A Report of the 2000 Field Season at the Barton Creek Cave, Cayo District, Belize. In The Western Belize Regional Cave Project: A Report of the 2000 Field Season, edited by Cameron S. Griffith, Reiko Ishihara, and Jaime J. Awe, pp. 97126, Department of Anthropology, Occasional Paper No. 4, University of New Hampshire, Durham. Morehart, Christopher T. 2002 Ancient Maya Ritual Cave Utilization: A Paleoethnobotanical Perspective. Master's Thesis, Department of Anthropology, Florida State University. 2005 Plants and Caves in Ancient Maya Society. In Stone Houses and Earth Lords: Maya Religion in the Cave Context, edited by Keith M. Prufer and James E. Brady, pp. 167-186, University Press of Colorado, Boulder. Morehart, Christopher T., David L. Lentz, and Keith M. Prufer 2005 Wood of the Gods: The Ritual Use of Pine (Pinus spp.) by the Ancient Lowland Maya. Latin American Antiquity 16(3): 255274). Moyes, Holley 2001 The Cave as a Cosmogram: The Use of GIS in an Intrasite Spatial Analysis of The Main Chamber of Actun Tunichil Muknal, A Maya Ceremonial Cave in Western Belize. M.A. Thesis, Florida Atlantic University. Boca Raton, FL. 2002 The Use of GIS in the Spatial Analysis of an Archaeological Cave Site. Journal of Cave and Karst Studies 64(1):9-16. 2004 Changes and Continuities in Ritual Practice at Chechem Ha Cave, Belize: Report on Excavations Conducted in the 2003 Field Season. Report submitted to the Foundation for the Advancement of Mesoamerican Studies, Inc., http://www.famsi.org/reports 2005 The Sweatbath in the Cave: A Modified Passage in Chechem Ha Cave, Belize. In Stone Houses and Earth Lords: Maya Religion in the Cave Context, edited by Keith M. Prufer and James E. Brady, pp. 187-212, University Press of Colorado, Boulder. 2006 The Sacred Landscape as a Political Resource: A Case Study of Ancient Maya Cave Use At Chechem Ha Cave, Belize, Central America. Unpublished Ph.D. Dissertation, Department of Anthropology, State University of Ne w York at Buffalo. Moyes, Holley and Jaime J. Awe 1998 Spatial Analysis of Artifacts in the Main Chamber of Actun Tunichil Muknal, Belize: Preliminary Results. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by Jaime J. Awe, pp. 23-36, Department of Anthropology, Occasional Paper No. 1, University of New Hampshire, Durham, NH. 2000 Spatial Analysis of an Ancient Cave Site. ArcUser 3(4):64-68 Owen, Vanessa A. and Sherry Gibbs 1999 Report of Investigations on Ledge 2 at Actun Yaxteel Ahau, Roaring Creek Valley, Belize. In The Western Belize Regional Cave Project: A Report of the 1998 Field Season edited by Jaime J. Awe, pp. 186204, Department of Anthropology, Occasional Paper No. 2, University of New Hampshire, Durham. Pendergast, David M. 1970 A. H. Anderson's Excavations at Rio Frio Cave E, British Honduras. Art and Archaeology Occasional Papers 20, Royal Ontario Museum, Toronto. 1971 Excavations at Eduardo Quiroz Cave, British Honduras (Belize). Royal Ontario Museum Art and Archaeology Occasional Paper 21. Royal Ontario Museum, Toronto. 1974 Excavations at Actun Polbiche, Belize. Archaeology Monograph 1. Royal Ontario Museum, Toronto.

PAGE 61

Holley Moyes 54 Prufer, Keith M. 2002 Communities, Caves, and Ritual Specialists: A Study of Sacred Space in the Maya Mountains of Southern Belize. Ph.D Dissertation, Southern Illinois University, Carbondale. Reilly III, F. Kent 1994 Visions to Another World: Art, Shamanism, and Political Power in Middle Formative Mesoamerica, Department of Art History, University of Texas, Austin. Rissolo, Dominique A. 2001 Ancient Maya Cave Use in the Yalahau Region, Northern Quintana Roo, Mexico. Ph. D. Dissertation, Department of Anthropology, University of California, Riverside. 2005 Beneath the Yalahau: Emerging Patterns of Ancient May Ritual Cave Use from Northern Quintana Roo, Mexico. In In the Maw of the Earth Monster: Mesoamerican Ritual Cave Use, edited by James E. Brady and Keith M. Prufer, pp. 342-372, University of Texas Press, Austin. Saturno, William A., Karl A. Taube, David Stuart, and Heather Hurst 2005 The Murals of San Bartolo, El Petn, Guatemala, Part 1: The North Wall. Ancient America No. 7. Schwarz, Henry P. and W. J. Rink 2001 Dating Methods for Sediments of Caves and Rockshelters with Examples from the Mediterranean Region. Geoarchaeology 16(4):355-371. Stone, Andrea J. 1995 Images from the Underworld: Naj Tunich and the Tradition of Maya Cave Painting. University of Texas Press Austin. Tarn, Nathaniel and Martin Prechtel 1986 Constant Inconstancy: The Feminine Principle in Atiteco Mythology. In Symbol and Meaning beyond the Closed Community: Essays In Mesoamerican Ideas edited by Gary Gossen, pp. 173-184, The Institute for Mesoamerican Studies, The University at Albany, State University of New York, Albany. Taschek, Jennifer T., and Joseph W. Ball 1999 Las Ruinas de Arenal: Preliminary Report on a Subregional Major Center in the Western Belize Valley (1991-1992 Excavations). Ancient Mesoamerica 10:215-235. Taube, Karl A. 1992 The Major Gods of Ancient Yucatan. Dumbarton Oaks, Washington, D.C. 1995 The Legendary Past: Aztec and Maya Myths. University of Texas Press, Austin.

PAGE 62

5 ON DROUGHT AND THE COLLAPSE OF THE ANCIENT MAYA Gyles Iannone This paper critically evaluates the drought hypothesis, which is currently the most popular explanation for what has long been termed the Maya collapse. It is argu ed that the drought hypothesis is an overly simplistic explanation for what was clearly a very complex event. Th e paper contends that, although a series of severe droughts beginning in the mid-8th century, and culminating in the early 10th century may have provided a succession of final blows to the institution of kingship at many southern lowland centers, the roots of the political and economic collapse can actually be traced back to the beginning of the 8th century, if not earlier. The paper concludes that the Maya collapse was primarily an internal, sociopolitical process. Data from various centers in Belize are used to support this conclusion. Introduction Over the past few decades, archaeologists along w ith a whole range of scholars from other disciplines have increasingly turned to climatic data in their efforts to enhance our understanding of the ancient Maya colla pse (ca. 750-1050 A.D.). Some of the earliest efforts to tie the Terminal Classic decline to climatic change were those of Dahlin (1983), Folan et al. (1983), Gunn and Adams (1981) and Messenger (1990). In the last ten years in particular, evidence for a series of devastating droughts has been building (see Robichaux 2002:341). Importantly, this data has emerged from a range of independent studies. The most significant of these are critically discussed he re. It is concluded that, although droughts likely played a key role in the collapse, it is apparent that many of the Maya city-states had already entered a period of power shari ng and decentralization prior to the onset of any debilitating climatic changes. Droughts are thus best viewed as components of a multi-causal explanation for the collapse. The Rio Candelaria Discharge Model In 1994 and 1995, one of the first convincing arguments for a droughttriggered collapse was presented by Joel Gunn and his colleagues (Gunn et al. 1995), who examined the annual discharge rate of the Rio Candelaria over a thirty-two year period. They then tied the fluctuations in the discharge rate to the double sea-breeze effect, which directly affects the intensity of rainy season precipi tation. Significantly, Gunn et al. were able to project their model back three thousand years, based on the notion that cooler temperatures in the northern hemisphere resulted in less precipitation in the Yucatn. One of their key conclusions was that there was an extended period of drought between 750 and 950 A.D. precisely the time of the infamous collapse. Yucatn Lake Cores Support for the proposed collapseera drought was provided in 1995, when David Hodell and his colleagues (Hodell et al. 1995) presented the first results from their analysis of sediment cores obtained from a series of lakes located in the central Yucatan Peninsula. The first core was derived from Lake Chichancanab, ...the largest closed-basin lake in Yucatn, Mexico (Hodell et al. 1995:391). The core was examined with th e specific goal of Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 55-65. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 63

Gyles Iannone Figure 1 Map of the Maya subarea showing the location of sites mentioned in the text documenting temporal fluctuations in sediment composition and oxygen isotope ratios. Associated C14 dates were employed to date the recognized climatic changes. These data confirmed that the period between 800 and 1000 A.D. was the driest period of the middle to late Holocene epoch. More importantly, the analysis suggested that the peak of the arid period occurred at 922 A.D. (Hodell et al. 1995:393) a date which fit squarely within the generally accepted collapse sequence. The following year, the same research team outlined similar findings based on the analysis of a sediment core from Lake Punta Laguna, located north of Lake Chichancanab (C urtis et al. 1996). In this study the authors were able to isolate three exceptionally arid events centered at... 862, 986, and 1051 A.D. (Curtis 1996:37). Once again, these dates articulated well with the established timeframe for the collapse. In a later article published in 2001, Hodell et al. (2001) revisited their Lake Chichancanab data set. Here they were able to document a drought cycle of approximately 208 years, each comprised of multiple 50-year oscillations. Interestingly, they also provided intriguing evidence to suggest that the drought cycle was linked to a previously documented solar forcing cycle of approximately 206 years; this led them to postulate that solar forcing may in fact be the root cause of global droughts (see also de Menocal 2001:668). Hodell and his colleagues were also able to fine-tune their drought chronology. Their new assessments suggested that the ke y dry period for those interested in the Maya collapse extended from 750 to 1025 A.D. (Hodell et al. 2001:1368). Two periods of peak aridity were recognized in this analysis, one centered on 800 A.D., the other 1020 A.D. 56

PAGE 64

Drought and Collapse of the Ancient Maya The Great Maya Droughts In his 1994 Ph.D. dissertation, and his 2000 book entitled The Great Maya Droughts Richardson Gill presented a model that was firmly grounded in the notion that drought, and the resulting famine, was solely responsible for the ancient Maya collapse (Gill 1994:456). Once again, Gill relied on the basic premise that colder temperatures in Europe and North America were indicative of less precipitation, and thus droughts, in the Maya subarea. Specifically, Gill postulated that meteorological data and tree-ring records both suggested that the period between 790 and 950 A.D. was punctuated by three severe droughts. Gill then articulated this climatic data with the dates for the last erected monuments at the largest southern lowland centers to create a model for three phases of drought and abandonment, which he suggested occurred between 760-810, 810-860, and 860-910 A.D. Gill (2000:320321) subsequently marshaled John Lowes (1985) seminal work on the collapse to propose a model of a pa tterned decline with a slight linear collapse from southwest to northeast. More specifically, Gill (2000:384) suggested that the collapse began first in the southwestern portion of the southern lowlands. This was followed by collapses in the southeastern region, and final the central region. In the most apocalyptic conclusion of all of the drought proponents, Gill (1994:476; 2000:387) proposed that most of the Maya of the southern lowlands perished from thirst and starvation during the Late Classic to Terminal Classic transition. The Cariaco Basin Finally, recent support for the drought hypothesis has come from the analysis of laminated sediments cores from the Cariaco Basin, located off the northern coast of Venezuela (Haug et al. 2003; Peterson and Haug 2005). Here, the researchers focused their attention on sediments that were deposited between 700 and 950 A.D., with the expressed purpose of examining the environmental characteristics of the Late Classic to Terminal Classic transition. In the end, they were able to document a century-long decline in precipitation that was punctuated by periods of severe drought lasting between three to nine years, on average. Of particular significance were multiyear drought periods, which the authors projected to have occurred around 760, 810, 860, and 910 A.D. This pattern of a 40 to 50 year peak drought cycle corresponded with the results of the Yucatn lake core study discus sed previously (Haug et al. 2003:1734). The date s also fit nicely with the three phases of abandonment proposed by Gill (Peterson and Haug 2005:327). Summary In summary, the diverse studies discussed above all point to the fact that the period between 800 and 1000 A.D. was likely the driest time span in the mid to late Holocene epoch. There is, however, considerable divergen ce of opinion with respect to the timing of the episodes of peak aridity or multiyear droughts which are postulated to have occurred during this dry period. In the end, the different data sets do suggest that drough t was likely a key component in the collapse sequence (see also Robichaux 2002:341). Nevertheless, as Hubert Robichaux ( 2002:341) cautions, one important question remains to be answered: Was the Terminal Classic drought so severe that the Maya succumbed directly because of its effects alone, as Gill has suggested? Or was drou ght one of a number of debilitating factors that combined in some way to fell the Maya? 57

PAGE 65

Gyles Iannone A Critical Assessment of the Drought Hypothesis Over the years, a variety of criticisms have been leveled against the drought hypothesis. Some of these are easily addressed. Others, however, are much more debilitating. Only the more pertinent criticisms are discussed here. The Philosophical Dilemma: Unicausal versus Multicausal Explanations In general terms, the unicausal basis of the drought hypothesis particularly as it has been proposed by Gill (2000) has led some to characterize it as a form of environmental determinism (see Gunn et al. 2002:79). Gill (2000:25) himself recognizes the unicausal qualities of the model, but defends these on philosophical grounds by arguing that: Sometimes, unicausal or monocausal explanations, in fact, fit the available data better than do multicausal ones (Gill 2000:25). Nevertheless, Gill does acknowledge that, when it comes to the ancient Maya collapse, complex, multicausal models have always been favored over unicausal engines (Gill 2000:363). In fact, both the complexity of the problem, and the diverse nature of the database itself, does imply that multicausal models likely have the greatest explanatory potential (see various papers in Demarest et al. 2004b; Iannone 2005). Along these lines, Peter de Menocal (2001:669) asks the ques tion: How did past cultures respond to the longer, multi-century scale climate changes that punctuated late Holocene climate? He concludes that, In all cases, the observed societal response reflects an interaction between human cultural elements (socioeconomic, political, and secular stresses) and persistent multicentury shifts in cl imate. The various nonarchaeologists who have been instrumental in generating the climatic data that provides the empirical foundation for the drought hypothesis concur. For example Haug et al. (2003:1734) surmise that: No one archaeological model is likely to capture completely a phenomenon as complex as the Maya decline. In their most recent article, Peterson and Haug (2005:328) conclude that: Although the match between Gills drought model and our findings is quite good, we accept that no single cause is likely to explain a ph enomenon as complex as the Maya decline. Environmental Issues Som e have been critical of the drought hypothesis because the wetter southern lowlands seem to have been affected more adversely than the drier northern zone. This cr itique is easily dispelled, however, when one considers that the southern lowland city-states were at a disadvantage compared to their northern lowland neighbors because, during arid periods, people living in the latter had access to groundwater, whereas the former were much more reliant on artificial and seasonal water collection facilities (Curtis et al. 1996:45-46; Gill 2000:380; Peterson and Haug 2005:327). As suc h, the southern lowland city-states were naturally more susceptible to the deleterious affects of drought (see also de Menocal 2001:670; Gill 2000:365-366; Haug et al. 2003:1734; Peterson and Haug 2005:322-324, 327). Others have disputed the efficacy of the drought theory becaus e its effects appear to have been patchy in nature some centers being hit hard, whereas neighboring centers emerge virtually unscathed, or they appear to suffer the effects of drought much later in the collapse sequence. This scenario is, however, characteristic of drought affected regions (Gill 2000:380). According to Gunn et al. (2002:80) ...there is a range of annual wetness and dryness. These differences occur not onl y on a yearly basis: at times they extend as droughts lasting 58

PAGE 66

Drought and Collapse of the Ancient Maya several years or even d ecades. Sometimes they strike part of the Yucatn Peninsula or highlands but not others. Thus, climatic differences occur through space and time at a number of scales (see also Haug et al. 2003:1734). This may explain why evidence for drought has been found in some parts of the Yucatn Peninsula, but not others. Finally, some scholars have been critical of the actual data sets employed by the proponents of the drought model. For example, Demarest (2001:107) stresses that, the explanatory potential of global climatic studies...are no substitute for carefully constructed regional archaeological and ecological research. Robichaux (2002:341) contends that Gills method of analysis is problematic because he tends to make his results applicable to the entire Maya lowlands. He also ca utions that, The Rio Candelaria model and the lake-sediment studies, by their nature have the greatest applicability to the specific areas that were investigated but possibly have applications to the broader lowland area (Robichaux 2002:341). Even some of the scholars who have contributed to the formulation of the drought model include caveats, such as the following: The paleoclimatic record from Lake Chichancanab alone cannot support the case for regional Late Classic drying...Although the sediment record from Chichancanab suggests a causal link between climate and the Late Classic collapse, the region inhabited by the Maya was geologically, ecologically and climatologically diverse. The response of climate to any single forcing factor may not have been the same throughout the Maya region and so the pattern of cultural change may not have been uniform (Hodell et al. 1995:393-394). Issues of Culture History The drought hypothesis meets one if its biggest challenges when it is considered alongside archaeologically generated culture history. One of the problems with the drought chronology is that the ranges for the dates of peak aridity are rarely worked into the analysis, and they are therefore somewhat misleading (Peterson and Haug 2005:327). We must keep this in mind when we attempt to mesh the culture history and drought data sets. Another key issue is that it appears as though the southern Maya city-states were already experiencing socioeconomic and sociopolitical problems before the first peak drought period occurred (Demarest 2004; de Menocal and Cook 2005:S92). Haug et al. (2003:1733) argue that a climatically favorable (relatively wet) period between 550 and 750 A.D. led to rapid growth in popula tion, a factor that would have made the so uthern lowland citystates particularly vulnerable to multiyear droughts (see also Gill 2000:360). John Lowe (1985:204-205) confirms that there was a rapid expansion in the number of Maya centers prior to 751 A.D., particularly in the central portion of the southern lowlands (see also Demarest 2004:119). After this the death rate of centers increased, even though the number of sites that erected monuments increased. Key here is that, even though more centers were erecting monuments, the number of monuments erected actua lly declined. Taken together, these data suggest that the citystates of the southern Maya lowlands had entered a period of political decentralization as early as 750 A.D. (Iannone 2005). Another data set ac tually suggests that the period of decentralization likely began before 750 A.D. The evidence for this can be found in the political and/or royal titles employed by the Maya. Beginning early in the 8th century, not only did the number of different titl es increase, but their frequency of use in in scriptions also rose dramatically (see Jackson 2005). This 59

PAGE 67

Gyles Iannone suggests that a degree of power sharing coincided with the poli tical decentralization referred to above (Culbert 1991:325-326; Iannone 1996:447, 2005; Stuart 1993:336). Importantly, Culbert (1991:326) notes that the proliferation of royal titles is particularly telling, as this practice occurs in both China (Fash 1985), and Mesopotamia (Yoffee 1977), during periods of sociopolitical instability leading up to the demise of royal dynasties. Of utmost importance to the current discussion is the fact that these aforementioned signs of weakness were manifest prior to any projected droughts. An additional culture historical fact that the drought hypothesis cannot explain is why the city-states of the Petexbatun region fell prior to the earliest documented drought? As Demarest (2001:106) points out, the major kingdoms in the western Petn region of Guatem ala disintegrated into warfare and migrations beginning in the mid-eighth century, ex tending back the collapse process. He goes onto to state that, The Petexbatun region, where much recent collapse research ha s been conducted, already was largely depopulated by the beginning of the ninth century, when Gills proposed volcanically induced droughts would have just begun. In terms of other weaknesses of the drought hypothesis, Robichaux (2002:341-342) underscores that Gills use of last monuments to date the fall of individual Maya centers has two main flaws: 1) the monument data suggests that the centers did not fall simultaneously, but rather over a protracted, 150 year period. For Robichaux, this co ntradicts the drought hypothesis. Specifically Robichaux argues that, regardless of th e different hydrological settings, and the patchy nature of droughts themselves, the long time span suggests that other causal factors must also have been at play the drought was thus not overwhelming; 2) the three phases of drought, from southwest, to southeast, and then to the central po rtion of the southern lowlands, also flies in the face of the hydrological backdrop. As Robichaux points out, the fact that southwestern centers such as Piedras Negras a nd Yaxchilan both of which are located along the prolific Usumacinta drainage declined before more aquatically-challenged centers in the southeast and central re gions, also calls into question the underpinnings of Gills drought hypothesis. Even Gill (1994:327) indicates that there is no evidence to suggest that the Usumacinta ever went dry. Robichaux (2002:344) concludes that the issues discussed above suggest the likelihood that cultural factors and events also played important roles in the development of the collapse. It should be noted here that Gills reliance on last monuments to build his collapse chronology is also problematic because the failure to erect monuments cannot be used as a proxy for total abandonment (see also Haug et al. 2003:1734; Peterson and Haug 2005:327). Ultimately, the cessation of monument erection simply signifies when the rulers of a particular center stopped erecting monuments. In other words, it indicates when Late Classic style kingship institutions died. It does not, however, specify when centers were abandoned. Finally, Gills (1994, 2000) study must also be considered flawed because he only includes upper tier centers in his analysis (Haug et al. 2003:1734; Peterson and Haug 2005:327). Centers without monuments need to be worked into the collapse equation, or our perspective on the collapse will continue to be biased towards the most politically prolific centers. This also creates a situa tion wherein it becomes easy to mistake the demise of the elite sector of Maya society for the collapse of society as a whole. 60

PAGE 68

Drought and Collapse of the Ancient Maya Drought, Famine, Severity, and Cultural Response The final set of criticism s revolves around the effects that the droughts had when they did arrive. According to Nancy Farris (1984), less br utal droughts during the ethnohistoric period clai med the lives of up to 50% of the indigenous population. The situation may have been even more devastating for the Terminal Classic Maya. Unfortunately, actual evidence for such a deadly famine is lacking, something that Gill (2000:328) himself admits. Of significance here is the f act that a number of environmental and biological studies conducted in the Petexbatun region the area that was hit earlie st and hardest by the collapse have failed to find any evidence for drought-induced famine before, during, or after the late eighth century political disintegration... (Demarest 2001:106). Demarest concludes that evidence such as this, and similar data from other parts of the southern lowlands, provide a serious challenge to Gills pr oposed simultaneous, famine-caused, ninth century collapse. There are also problems with Gills (2000:350) assertion that as a result of the severe droughts, Most sites in the southern and central lowlands were abandoned after the collapse. Those few sites not abandoned had reduced populations (Gill 2000:350). This is clearly not the case in all instances. Belizean centers such as Lamanai (Pendergast 1986), and Caracol (D. Chase and A. Chase 2004), do not appear to have been hit by severe demographic changes during the periods wh en the droughts are postulated to be most severe. Our current research at Minanha al so suggests that there was a sizable population at that center during the Terminal Classic. This data alone calls into question the actual severity of the Terminal Classic droughts. Nevertheless, it is still plausible that these eastern centers exhibit strong Terminal Classic settlement simply because of the patchy, and protracted, nature of the droughts. It is also possible that the apparent demographic resilience of these eastern centers actually foreshadows their impending doom. In particular, it may reflect severe demographic problems elsewhere in the southern lowlands. The events in the Petexbatun are especially important to the overall collapse sequence because the 8th century migrations from this troubled region may have affected the stability of the city-states located to the east (Demarest 2004:119; Demarest et al. 2004a: 568). Eventually, due to increasing populations, the southern lowland Maya may have been unable to react effectively to the new series of droughts because they may not have been able to migrate into new areas to ease the pressure (Peterson and Haug 2005:328). Importantly, our own research at Minanha actually indicates that in the early part of the 8th century migrants were already moving into areas which had previously only supported low population densities, and which had markedly less agricultural potential, than the areas from which they had originated (Iannone 2005). In the Minanha case, the evidence suggests that migrants likely came from the area around Caracol if not Caracol itself and they may have included both peasants and elites. The fact that these migrants established a new royal court in an area that had never supported one before implies that the Maya socioeconomic and sociopolitical systems were likely already stretched to their limits before the Terminal Classic period, and likely before the onset of the droughts. Another archaeologi cal data set that presents challenges for the drought hypothesis is the infillin g and/or termination of elite residential courtyards prior to total site abandonment. Excavations at Belizean centers such as Dos Hombres, Blue Creek, 61

PAGE 69

Gyles Iannone and possibly Punta de Cacao, have exposed large Terminal Classic deposits of ceramic sherds and other items on the floors and stairs associated with elite courtyards (Adams et al. 2004:337). These deposits, some of which block the entrances to the courtyards themselves, are thought to represent acts associated with the termination of the elite inhabitants themselves, as opposed to the actual architectural features th at they are associated with (Adams et al. 2004:338). Even more dramatic is the total infilling of royal residential courtyards, and their associated throne rooms and residential buildings, during the early part of the Terminal Classic period. This practice has been recorded at La Milpa (Hammond 1999a, 1999b; Hammond and Thomas 1999), Lamanai (Graham 2004), and Minanha (Iannone 2005, 2006). The aforementioned data provide a serious challenge to Gills model of droughtinduced famine. For example, Gill (2000:374) argues that, Famine is generally a class affliction. It hits the poor hardest and first...The disintegratio n begins, then, with the lowest levels of society, including the peasants, and it proceeds from the bottom up. Elsewhere, he surmises that, In the case of the Maya, due to the length and severity of the droughts, the flow of food and water was shut off, the people died from hunger and thirst, and the social system collapsed from the bottom up (Gill 2000:371). The data presented above, however, suggests that the exact opposite happened in the Maya case. The first segment of society to meet their demise was apparently the elite, and particularly the royalty. That the surviving communities were still relatively strong is implied by the amount of labor and organization that went into the various infilling and termination events. Conclusions Gill (2000:363-368) argues that the collapse was not the result of bad managerial skills it was unpredictable and unavoidable. The collapse was thus an external process in its entirety (2000:367). I disagree. The data sugges ts that as is true in almost every other example of collapse the ancient Maya of the southern lowlands had likely put themselves in a situation that left them highly vulnerable to adverse climatic change long befo re the onset of the great Maya droughts. For this reason, I argue here that the current temporal and spatial data sets from the Late Classic to Terminal Classic era imply a secondary role for climatic factors, su ch as drought, in the fall of the southern lowland city-states (Demarest et al. 2004a: 568). Specifically, there is considerable evidence to suggest that before the Maya of the southern lowlands would have felt any of the effects of drought, this portion of the Maya subarea had likely already entered a period of political fragmentati on. This was possibly stimulated by economic stresses brought on by migrations, population increases, concomitant declining agricultural yields, increased competition, and in some cases endemic warfare. When the droughts did hit, their effects were difficult to overcome because the economic and political powerbases of the various rulers had already been severely eroded. The subsequent failure of the water managements systems during periods of severe dr ought would have only served to further undermine the efficacy of the rulers who had always proclaimed ideological, political, and economic control over such resources (Lucero 2002). In conclusion, I concur with others who have argued that: A one drought fits all model seems too simplistic, given that the collapse apparently happened at different places at different ti mes, while affecting 62

PAGE 70

Drought and Collapse of the Ancient Maya som e centers hardly at all (Peterson and Haug 2005:326). Nevertheless, I do agree that the great Maya droughts likely provided the coup de grace the final killing blow for many of th e city-states in the southern Maya lowlands. However, I wish to stress that these city-states, and their rulers, were already highly vulnerable to such a blow. The writing was on the wall, so to speak, long before the rains stopped coming. Acknowledgements My gratitude is extended to the Social Sciences and Humanities Research Council of Canada for their continued suppor t of the Minanha research upon which this paper is partially based. I also wish to acknowledge the Institute of Archaeology in Belize for hosting the annual symposium during which the ideas discussed herein were first presented. The final paper has benefited from discussions with a number of individuals, includi ng John Morris, Jaime Awe, Arlen Chase, Lisa Lucero, Norbert Stanchly, Claudia Zehrt, Carmen McCormick, Emma Dunkley, Jesse Phillips, Matthew Mosher, Trudy Kirschner, Jason Seguin, Rachel Dell, Natalie Thornhill, Anthony Beardall, Joe Martinez, and Lazaro Martinez. Many thanks are also extended to Miriam Khalil, and both Landy and Erva Espat, for the technica l, and nutritional, support they provided during the writing of the conference paper upon which this chapter is based. The errors in fact and logic that remain in the final product are fully attributable to the author. References Cited Adams, Richard E. W., Hubert R. Robichaux, Fred Valdez Jr., Brett A. H ouk, and Ruth Mathews 2004 Transformations, Periodicity, and Urban Development in the Three Rivers Region. In The Terminal Classic in the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 324-341. University of Colorado Press, Boulder. Chase, Diane Z., and Arlen F. Chase 2004 Hermeneutics, Transitions, and Transformations in Classic to Postclassic Maya Society. In The Terminal Classic in the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 12-27. University of Colorado Press, Boulder. Culbert, T. Patrick 1991 Maya Political History and Elite Interaction: A Summary View. In Classic Maya Political History: Hieroglyphic and Archaeological Evidence edited by T. Patrick Culbert, pp. 311-346. Cambridge University Press, New York. Curtis, Jason H., David A. Hodell, and Mark Brenner 1996 Climate Variability on the Yucatan Peninsula (Mexico) during the Past 3500 years, and Implications for Maya Cultural Evolution. Quaternary Research 46:37-47. Dahlin, Bruce H. 1983 Climate and Prehistory on the Yucatan Peninsula. Climate Change 5:245-263. Demarest, Arthur A. 2001 Climate Change and the Classic Maya Collapse: The Return of Catastrophism. Latin American Antiquity 12:105-107. 2004 The Ancient Maya: Rise and Fall of a Rainforest Civilization Cambridge, New York. Demarest, Arthur A., Prudence M. Rice, and Don S. Rice 2004a The Terminal Cl assic in the Maya Lowlands: Assessing Collapses, Terminations, and Transformations. In The Terminal Classic in the Maya Lowlands edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 545-572. University of Colorado Press, Boulder. Demarest, Arthur A., Prudence M. Rice, and Don S. Rice (editors) 2004b The Terminal Classic in the Maya Lowlands University of Colorado Press, Boulder deMenocal, Peter B. 63

PAGE 71

Gyles Iannone 2001 Cultural Responses to Climate Change During the Late Holocene. Science 292: 667-673. deMenocal, Peter B., and Edward R. Cook 2005 Perspectives on Diamonds' Collapse: How Societies Choose to Fail Or Succeed. Current Anthropology 46: S91-S92. Fash, William L. 1985 The Evolution of the Copan Polity. Paper presented at the 50th Annual Meeting of the Society for American Archaeology, Denver. Farriss, Nancy M. 1984 Maya Society Under Colonial Rule Princeton University Press, Princeton. Folan, William J., Joel Gunn, Jack D. Eaton, and Robert W. Patch 1983 Paleoclimatological Patterning in Southern Mesoamerica. Journal of Field Archaeology 10:453-468. Gill, Richardson B. 1994 The Great Maya Droughts Unpublished Ph.D. dissertation, Department of Anthropology, University of Texas, Austin. 2000 The Great Maya Droughts. University of New Mexico Press, Albuquerque. Graham, Elizabeth A. 2004 Lamanai Reloaded: Alive and Well in the Early Postclassic. In Research Reports in Belizean Archaeology, Volume 1 edited by Jaime Awe, John Morris, and Sherilyne Jones, pp. 223-241. Institute of Archaeology, National Institute of Culture and History, Belmopan, Belize. Gunn, Joel, and Richard E. W. Adams 1981 Climatic Change, Culture, and Civilization in North America. World Archaeology 13:87-100. Gunn, Joel D., William J. Folan, and Hubert R. Robichaux 1995 A Landscape Analysis of the Candeleria Watershed in Mexico: Insights into Paleoclimate Affecting Upland Horticulture in the Southern Yucatan Peninsula. Geoarchaeology 10:3-42. Gunn, Joel D., Ray T. Matheny, and William J. Folan. 2002 Climate-Change Statistics in the Maya Area: A Diachronic Analysis. Ancient Mesoamerica 13:79-84. Hammond, Norman 1999a A Pillar of State... Majestic, Though in Ruin : The Royal Acropolis of La Milpa. Context 14(1):11-15. 1999b A Maya Throne Room at La Milpa. PreColumbian Art Research Institute Newsletter 28:12-13. Hammond, Norman, and Ben Thomas 1999 Another Maya Throne Room at La Milpa. Context 14(1):15-16. Haug, Gerald H., Detlef Gnther, Larry C. Peterson, Daniel M. Sigman, Konrad A. Hughen, and Beat Aueschilmann 2003 Climate and the Collapse of Maya Civilization. Science 299:1731-1735. Hodell, David A., Jason H. Curtis, and mark Brenner 1995 Possible Role of Climate in the Collapse of Classic Maya Civilization. Nature 375:391-394. Hodell, David A., Mark Brenner, Jason H. Curtis, and Thomas Guilderson 2001 Solar Forcing and Drought Frequency in the Maya Lowlands. Science 292:13671370. Iannone, Gyles 1996 Problems in the Study of Ancient Maya Settlement and Social Organization: Insights from the Minor Center of Zubin, Cayo District, Belize Unpublished Ph.D. dissertation, Institute of Archaeology, University College London, London. 2005 The Rise and Fall of an Ancient Maya Petty Royal Court. Latin American Antiquity 16:26-44. 2006 Investigations in the Buried Royal Residential Courtyard at Minanha, Belize. In Research Reports in Belizean Archaeology, Volume 3 edited by John Morris, Sherilyne Jones, Jaime Awe, and Christophe Helmke, pp. 149-160. Institute of Archaeology, National Institute of Culture and History, Belmopan, Belize. 64

PAGE 72

Drought and Collapse of the Ancient Maya 65 Jackson, Sarah E. 2005 Deciphering Classic Maya Political Hierarchy: Epigraphic, Archaeological, and Ethnohistoric Perspectives on the Courtly Elite Unpublished Ph.D. dissertation, Department of Anthropology, Harvard University, Cambridge. Lowe, John W. G. 1985 The Dynamics of Apocalypse: A Systems Simulation of the Late Classic Maya Collapse. University of New Mexico Press, Albuquerque. Lucero, Lisa 2002 The Collapse of the Classic Maya: A Case Study for the Role of Water Control. American Anthropologist 104:814-826. Messenger, Lewis C. Jr. 1990 Ancient Winds of Change: Climatic Settings and Prehistoric Social Complexity in Mesoamerica. Ancient Mesoamerica 1:21-40. Pendergast, David M. 1986 Stability through Change: Lamanai, Belize, from the Ninth to the Seventeenth Century. In Late Lowland Maya Civilization: Classic to Postclassic edited by Jeremy A. Sabloff and E. Wyllys Andrews V., pp. 223-249. University of New Mexico Press, Albuquerque. Peterson, Larry C., and Gerald H. Haug 2005 Climate and the Collapse of the Maya. American Scientist 93:322-329. Robichaux, Hubert R. 2002 On the Compatibility of Epigraphic, Geographic, and Archaeological Data, with a Drought-Based Explanation for the Classic Maya Collapse. Ancient Mesoamerica 13:341-345. Stuart, David 1993 Historical Inscriptions and the Maya Collapse. In Lowland Maya Civilization in the Eighth Century A.D. edited by Jeremy A. Sabloff and John S. Henderson, pp. 321354. Dumbarton Oaks Research Library and Collection, Washington, D.C. Yoffee, Norman 1977 The Economic Role of the Crown in the Old Babylonian Period Undena, Malibu.

PAGE 73

6 LATE AND TERMINAL CLASSIC OCCUPATION OF PUSILHA, TOLEDO DISTRICT, BELIZE SITE PLANNING, BURIAL PATTERNS, AND COSMOLOGY Geoffrey E. Braswell Archaeological investigations at Pusilha, Toledo District, Belize have revealed an occupational history beginning shortly before A.D. 600 and lasting well into the Terminal Classic period. Site planning, architecture, and even burial patterns of the Late and Terminal Classic periods reflect important cosmological principles shared by the Classic Maya. Nevertheless, in southern Belize, a rather distinct and isolated archaeological region, the way these shared principles were manifested was unique. Introduction Since 2001, the Pusilha Archaeological Project has investigated ancient economy and political history at Pusilha, Toledo District, Belize (Bill and Braswell 2005; Braswell et al. 2004, 2005; Braswell and Gibbs 2006). Located in the southwestern corner of the country, merely 1,500 m east of the border with Guatemala and within the Qeqchi village of San Benito Poite, Pusilha is the largest ancient Maya city of southern Belize (Figure 1). To date, project members have surveyed nearly 3 km2 of the settlement, which we estimate to have covered approximately 5 to 6km2 (Figure 2). During survey, 105 structural groupscontaining a total of about 500 mound and terrace featureshave been mapped. In 2002, we conducted a testpitting program designed to recover a representative sample of ceramics and other artifacts from different portions of the ancient city. In 2002, 2004, and 2005, we conducted extensive horizontal excavations in seven structures and tested an eighth platform. Throughout our project, Christian Prager, co-director and project epigrapher, has drawn, cataloged, and analyzed the extensive hieroglyphic corpus of the site, consisting of 23 stelae a hieroglyphic stair, an inscribed ballcourt marker, and 18 miscellaneous fragments (Figure 3). The Political History of Pusilha Our interest in Pusilha grew out of research conducted by co-director Cassandra Bill and I at Copan, Honduras. Since the rediscovery of Pusilha in 1927, several scholars have posited a close relationship between the Belizean city and the CopanQuirigua region. Marc us (1992) argued that Pusilha began as a small and independent chiefdom during the Early Classic, was annexed by the rapidly expanding Copan polity, and later regained its independence during the 8th century in a manner very similar to that proposed for Quirigua. But as we have described at previous symposia, Marcus (1993) Dynamic Model of state formation does not seem to apply to Pusilha quite the way we once thought, and evidence of both political and economic connections with Copan are quite limited. For this reason, we have developed a new narrative for the political history of Pusilha, one that emphasizes local development without foreign control (Braswell et al. 2004). An evanescent Early Classic presence in the Pusilha region is demonstrated only by pottery collected from the numerous caves found in the hills near the site. Permanent settlement began late in the 6th century A.D., when colonists arrived from the southwestern Peten. Bills (in Bill and Braswell 2005) analyses reveal the Research Reports in Belizean Archaeology Vol. 4, 2007, pp.67-77. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 74

Geoffrey Braswell Figure 1. The ruins of Pusilha, Toledo District, Belize. Bo rder with Guatemala is shown in yellow, the modern village of San Benito Poite is visible. Figure 2. Partial map of Pusilha. Contour interval is 1m, s cale in meters as measured from datum in Stela Plaza. strongest ceramic ties with that region (and much stronger ties, in general, with Tepeu sites in the Peten rather than with the southeastern Maya pe riphery). Moreover, Pragers epigraphic st udies have discovered names in the Pusilha corpus that were also used in the Petexbat un region (Braswell et al. 2004, 2005). Prager has identified 39 individuals in the hieroglyphic corpus of Pusilha. Eight to ten of these are ajawob or divine kings. The dynastic founder of Pusilha, whom we call Ruler A and whose name is read kawil chan kinich was inaugurated in A.D. 570. He signaled his status as the founder of a dynasty by employing the ochkin 68

PAGE 75

Terminal Classic Occupation at Pusilha Figure 3. Seven stelae from the Stela Plaza, Pusilh a (drawing by Christian M. Prager). kalomte title. During his long rule, Ruler A appears to have formed limited economic and ambassadorial ties with Copan. Among the ceramics recovered by the British Museum and by our own project are a small number of sherds th at exhibit motifs known principally from western Honduras and eastern El Salvador. Moreover, the dynastic founder of Pusilha named his firstborn son Pusilha Ruler B after Copan Ruler 11. But it is important to stress that these ceramic and epigraphic ties are limited, not only in the frequency of their appearance, but also in time. After the mid-seventh century, evidence for this interaction ceases. Again, ceramic data, as well as obsidian procurement patterns and iconographic evidence, reveal much closer affiliations with the Peten than with any other region throughout the Late Classic period (Braswell et al. 2005). 69

PAGE 76

Geoffrey Braswell Ruler F, nam ed ix ichak ... kinich was a female ajaw who ruled Pusilha in her own right. It is likel y that she ruled only long enough for her son, whose father was not a divine king, to acc ede to the throne. In an attempt to legitimize his rule, Ruler G, the son and successor of this woman ajaw used the important title ochkin kalomte again implying the founding of a new royal line. Ruler G died between A.D. 731 and 751 (Braswell et al. 2005). Three late monuments describe the actions of two or possibly three later kings of Pusilha who ruled between A.D. 751 and A.D. 798. But the epig raphic record for the end of the Late Classic and the beginning of the Terminal Classic is scanty compared to earlier periods. What happened to the population of Pusilha during the 9th century? To begin with, it is quite possible that population levels dropped at this time, as they did throughout the southern and central Maya lowlands. Nonetheless, excavations have revealed a robust Terminal Classic occupation. During this period, many important groups, including the Gateway Hill Acropolis, remained populated. Moho Plaza, located in the southwestern periphery of the site and where the Hieroglyphic Stair and Ballcourts III and IV are found, was built and occupied at this time. The larger of these two ballcourts is constructed on an East-West axis, an orientation characteristic of the Terminal Classic and later periods. In some ways, Moho Plaza seems to have replaced much of the Late Classic city and conflated within its boundaries architectural features that in earlier times were dist ributed in different structural groups. Excavations in the Moho Plaza and elsewhere have revealed significant evidence, in the form of imported Belize Red vessels, of economic ties with western Belize during the Terminal Classic (Braswell et al. 2004). Evidence of interaction with the Belize Valley is not present before this time. Ceramic forms and a few sherds of Fine Orange ware demonstrate new relations with the northwest Maya region, as does the importation of obsidian from the Mexican source of Zaragoza, Puebla. It is also distinctly possible that by about A.D. 800, large segments of the population of Pusilha moved to a new location in southern Belize. Lubaantun, a site whose major architecture was built near the end of the Late Classic and a location occupied well into the Terminal Classic period (Hammond 1975), may have been one place where people from Pusilha moved. Certain aspect s of the utilitarian ceramic inventory an d the practice of making ocarinas in the form of figurines suggest relations between that site and Pusilha. More importantly, the basic site plan of Lubaantun, which also is a hilltop citadel, is very similar to that of Pusilhas Gateway Hill Acropolis. The two most imposing structures of each acropolis are located on their eastern edges and are both west-facing platforms. The western portions of each acropolis consist of a more open arrangement of patio groups built on different levels. The acropolis of Pusilha is similar enough to that of Lubaantun that I suspect the former may have been the architectural inspirat ion for the latter. Excavations in a platform at Pusilha that we call the Bulldozed Mound reveal an even later Postclassic occupation (Braswell et al. 2004). Ceramics dating to this period are crude and un-standardized in ways reminiscent of the Ejar complex of Copan, the Newtown complex of the Belize Valley, and various other Early Postclassic ceramic complexes from the southern Peten (Bill and Braswell 2005). Thus, at least small numbers of people occupied Pusilha at a time well after the Maya Collapse of the late 8th and early 9th centuries. 70

PAGE 77

Terminal Classic Occupation at Pusilha One of the most strik ing aspects of the political history of Pusilha is that the extensive hieroglyphic corpus never once mentions Tikal, Calakmul, Caracol, Copan, Quirigua, or even nearby Nimli Punit (Braswell et al. 2004). Rival polities and toponyms are named in connection with warfare events, but not one is a place whose location is known. Although many of the original inhabitants of Pusilha almost certainly came from the southern Peten, the lack of references to large and powerful sites in that region suggests that the royal family of Pusilha chose to distance itself from the political machinations that permeated the Maya lowlands during the 7th through 9th centuries. We do not, therefore, interpret the history of Pusilha in terms of Marcus Dynamic Model (in which the development of state-level organiza tion at Pusilha could be seen as the result of first incorporation into and later fragmentation of the Copan state), nor do we consider to be relevant Martin and Grubes Superstate Model (in which the political affairs of many lowland polities were manipulated in a hegemonic fashion by Tikal or Ca lakmul). Instead, Pusilha is best viewed as a nonaligned polity that developed independently in a generally under-populated and somewhat peripheral region of the Maya lowlands. That is, we propose that there was a third way to Maya political development, one that entailed neither hegemonic nor direct political control. Some years ago, Leventhal (1990) described southern Belize as an archaeological region distinct from the rest of the lowlands. One reason that material culture may have deve loped in a unique and somewhat divergent manner in southern Belize is that the inhabitants of the region maintained political independence from their larger neighbors throughout the Late Classic period. Site Planning at Pusilha I would like to em phasize several aspects of material and ideological culture at Pusilha that link the city in some respects to the Peten, but also help to define southern Belize as a region quite di stinct from the rest of the southern and central Maya lowlands. In particular, I would lik e to stress an aspect of our research that has not yet been described in detail, the results of our settlement survey. I will also discuss several burials excavated by the project with a special emphasis on how they relate to site planning and cosmological principles. The map shown in Figure 2 is still incomplete. The vacant north-south strip running down the center of the map has yet to be surveyed, and contains at least three large groups of structures. Moreover, the limits of the site still need to be defined, particularly on the northern side of the Poite River and towards the Guatemalan border. Years ago, Morley ( 1938) noted that one important feature characteristic of inscriptions from southern Belize is that they include erroneous or otherwise distinctive Lunar Series data. In Leventhals (1990) definition of the southern Belize region, he chooses to focus on architectural features. He notes that the ballcourts of southern Belize tend to be located within walled enclosures that separate the entirety of the ballcourt as well as other auxiliary structures from outlying areas. Leventhal also discusses how natural features of the landscape, particular ly hills and large boulders, were incorporated into architectural constructions. The effect, he writes, is rather like a Hollywood set, consisting primarily of cut-stone facades and a more limited use of fill than is found in many other regions. Such construction is often viewed as a way to save labor. But it is important to stress that the most timeconsuming aspect of co nstruction is cutting stone blocks to face structures or natural 71

PAGE 78

Geoffrey Braswell features, not the gathering and placing of fill. Moreover, at Pusilha, the deliberate incorporation of natura l features such as boulders into architect ure seems to have served a religious function. Finally, Leventhal (1990) also mentions that tombs in the southern Belize region were often re-used, a pattern also noted at nearby Caraco l and at Minanha. Although we have excavated no tombs at Pusilha that were clearly used for sequential burials, at least one major tomb was reopened and resealed, and another elaborate crypt is a secondary burial. Equally important in defining the southern Belize region are features common at other Maya sites that are lacking or rare in southern Belize. An important and wellknown feature of southern Belize architecture is the lack of vaulted superstructures and tombs. A second feature that is quite rare in southern Belize is the Egroup or astronomical commemoration group. The paucity of E-groups, many of the best known of which are Late Preclassic in date, may not seem particularly surprising in a region where virtually all known architecture dates to the Late and Terminal Classic periods, but it s hould be recalled that in the neighboring southern Peten, particularly in the Dolores Valley, Juan Pedro Laporte has discovered and mapped more than a hundred E-groups dating to precisely this period. Related to the lack of E-groups is a difference in burial patterns. In western Belize and eastern Peten, a common pattern is to place the principal tomb in a pyramidal structure on the eastern edge of a plaza group. In many cases, these ancestor shrines are also part of an E-group. In southern Belize, burials are indeed present in and in front of eastern structures. But in many cases these are range platforms that likely supported residences or other buildings. Moreover, important burials and offerings are also found associated with the northern, southern and western structures of a group. Figure 4. Three Special Function Groups at Pusilha, shown at the same scale. Contour interval is 1 m, except in (A) where it is 20 cm. Special Function Groups at Pusilha Survey has revealed a distinct architectural pattern at Pusilha, what I call the Special Function Group (Figure 4). Unlike habitation groups, built around one or more open patios oriented with the natural topography, Special Function Groups are built on a NNW-SSE alignment. As Ivan prajc (personal communication, 2004) has noted, this particular alignment is quite unusual for Mesoamerica; a more typical pattern is NNE-SSW. The archetypal pattern for a Special Function Group has three structures defining the eastern side of a patio. The northern tw o of these eastern platforms may have supported range structures. The southernmost (not the central) of the eastern platforms is generally the highest, and may be square rather than rectangular in plan. The two principal structures (often square at the base) are located to the north and south of the patio, and a single range stru cture is found on the western edge of the platform. The western side of the patio is qui te open in comparison 72

PAGE 79

Terminal Classic Occupation at Pusilha to the eas tern side. At least three Special Function Groups are known at Pusilha: The Lunar Group, near the northeast corner of our map, the Stela Plaza, and the Moho Plaza, located just off the far southwest corner of the map. Although built on the same axis and with the same general plan as the Stela Group and Lunar Group, the Moho Plaza is somewhat aberrant from the other two examples in four respects. First, the northern principal structure is an east-west ballcourt rather than a pyramidal mound. Second, the southern pr incipal structure, although quite high, is rectangular and contains a hieroglyphi c stair. Third, an additional large structure is located in the center of the open plaza. Finally, the Moho Plaza is built on low-lying terrain rather than on a commanding hilltop. I have called these Special Function Groups not only because of their layout, but also because of the artifacts found in the two examples where we have conducted test-pitting operations. First, and most obvious, the Moho Plaza and the Stela Plaza contain carved sculpt ure with hieroglyphic inscriptions. In addition to the Hieroglyphic Stair, three carved ba llcourt markers were found in the ballcour t. Second, ceramics from these groups are generally not utilitarian in character. The large and numerous jars, tinajas basins, ollas and serving plates found in most of our excavations are uncommon by comparison in the Stela Group and Moho Plaza. Instead, incense-burner fragments are quite common. Third, animal bones and river-snail shells -nearly as ubiquit ous as pottery sherds in residential areas of the site are entirely absent from our collections from the Stela Plaza and Moho Plaza. Thus, the preparation and consumption of food do not seem to have been important functions in the two Special Function Groups where we have conducted excavations. One conjecture that needs to be tested thro ugh further survey is that the Special Function Groups somehow indicate the corners and center of the site, in a manner similar to the U-shaped groups of Copan. Figure 5. Lower Groups I and II, located 150 m southwest of the Gateway Hill Acropolis. Contour interval is 1 m. Directionality and Height Although the Special Function Groups of Pusilha are built on a plan and according to an axis not known outside of southern Belize, other more complex aspects of site planning do reflect cosmological notions shared throughout the Maya lowlands (Braswell et al. 2004, 2005). The clearest example can be seen in how the Stela Plaza Group articulates with Ballcourt I. Here, the Stela Plaza, with its row of monuments depicting ru lers and describing their exploits, is located to the northwest and is conceptual linked to the heavens, divine kingship, and the northern side of the world. For the Maya, ballcourts were associated with death, the south, and the underworld. The Stela Group is connected to the ballcourt by a 150m long sacbe. On both sides of the middle of the sacbe and at an intermediate height are residential groups that may represent our own world. The sacbe, therefore, connects and separates the three worlds much like the World Tree of 73

PAGE 80

Geoffrey Braswell Maya m ythology. An important concept of the plan of this part of the site is that not only does it reflect Maya cosmological notions of directional ity and the universe, but also it expresses these notions in three dimensions. The Stela Group (representing the heavens) is some 12m higher on the hill than Ballcourt I (representing the underworld), and the residential groups are at an intermediate altitude. Other Important Groups At least three other groups share some but not all of the features of the Special Function groups. In these cases, the portion of each larger group that shares one or more of the features of a Special Function Group is located on the highest terrain and forms the northernmost cluster of structures defining the greater group. Again, both altitude and the north are associated with the heavens, so it is plausible that these other groups (rather than eastern shrines of the sort found in the Belize Valley) were places of ancestor veneration. We have excavated in two of these: Lower Group I and the Pottery Cave Group and ample evidence of household and ritual activities were recovered, so their use was more general than that of the Special Function Groups. We conducted extensive excavations of two structures in Lower Group I: the eastern range platform (called the Op. 5 Structure), and the southern pyramidal platform (or Op. 6 Structure; Figure 5) (Braswell et al. 2005). Both structures contained two burials. The most elaborate one was found not in the eastern structure, but in the southern one. This crypt contained the remains of an individual with inlayed teeth, but many of the important skeletal elements were missing. For this reason, we interpret it as a re-interment or other secondary burial. Symbols of high office found in the crypt include a white limestone baton, a slate paddle or wrench, and a pyrite mirror most often associated with warriors. It is likely the burial of a Late Classic non-royal functionary who lived below and in the shadow of the acropolis (Braswell et al. 2005). The Gateway Hill Acropolis The most important architectural group at Pusilha is the Gateway Hill Acropolis (Figure 6). An ancient toponym known from the Pusilha hieroglyphic corpus consists of a set of st airs and the glyph witz or mountain. Rising vertically some 79m from the base of the ancient Maya bridge, the Gateway Hill Acropolis is indeed a stepped mountain. Faced with cut-stones and viewed suddenly from a canoe passing beneath the Maya Bridge, it must have been one of the most imposing acropoli in the Maya world. Figure 6. Gateway Hill Acropolis, showing locations of Operation 8 and Operation 9 Structures. Contour interval is 1 m. Large boulder features are shown in gray and pink The orientation of the Gateway Hill Acropolis follows that of the natural hill on which it is built: NNW to SSE. The 74

PAGE 81

Terminal Classic Occupation at Pusilha orientation of the Sp ecial Function Groups, therefore, mimics the natural axis of the acropolis. Two large boulders are centered on the northern face of the acropolis just south of the Maya bridge and on the climb leading up to the first terrace. In front of each boulder is a platform where perhaps religious activities were conducted. Numerous small and shallow cavities have been dug beneath each boulder, and it is easy to imagine that incense was burned and offerings were placed in them, in a manner similar to that practiced today in the Maya highlands. The natural orientation of these boulders also reflects the axis of both the Gateway Hill Acropolis and of the Special Function Groups. Two sets of stairs and terraced ramps symmetr ically flank the boulders and lead up to the first terrace. A short sacbe or ramp leads 20m down from this first terrace to the south end of Ballcourt II, which is bordered immediately to the west by the Machaca or Pusilha River. Thus, Ballcourt II (associated with the underworld) is at the lowest level of the acropolis, it is connected by a sacbe to a terrace with range platforms (which may have supported dwellings), and some 50m above the terrace are the large structures where royal burials ar e located. Although the compass directionality is reversed, the vertical pattern of ballcourt/living space/ancestor worship is the same as found in the Stela Plaza-Ballcourt I complex. From the first terrace, an eastern stair rising to the second terrace provides access to the higher portions of the acropolis. Here, three platforms the highest of which is in the south, as in a Special Function Group flank the eastern side of the terrace. Two more large boulders form the natural gateway that gives the hill its modern name. A long set of poorly pres erved stairs leads to the fourth terrace and eventually to the base of the northernmost and lowest of three high structures whose westward orientation, axis, and relative size agai n reflect the pattern seen in the Special Function groups. The northernmost platform has been subjected to horrific looting and at least three (but probably more) royal tombs have been opened and robbed. During the 2005 field season, we excavated both the central platform, which we call the Op. 9 Structure, and the southern pyramid or Op. 8 Structure. The Op. 9 Structure is not a true pyramid, but is the natural top of the hill. Its western side is faced with cut masonry and does not contain a stair; instead, there are steps on its northern and narrow face. The Op. 9 Structure served, therefore, as an elevated passageway linking the top of the northern pyramid to the Op. 8 Structure. In 2005, we also excavated the Op. 8 Structure, the highest point and largest platform in Gateway Hill Acropolis (Braswell et al. 2005). Two important burials were found associated with the structure, and show how the same cosmological principles that determined site planning also are reflected in mortuary patterns. The Op. 8/3 Crypt and Op. 8/4 Tomb In front of the stair of the Op. 8 Structure and on the center line of the building, we discovered a simple crypt burial containing two individuals (Figure 7). The single ceramic vessel in the interment dates to the late 8th or early 9th century. The principal figure was encountered in an extended position, and an accompanying figure was flexed. As in all but one of the burials we have excavated, the heads of each figure point north, or more correctly are on the same alignment as the Gateway Hill Acropolis and the Special Function Groups. This orientation probably reflects not only ideas about the heavens, but also the concept of verticality that conceptually links north and up. Grave goods include an incised pyraform vessel to the east of the principal 75

PAGE 82

Geoffrey Braswell individual, and the accom panying figure located to the north. This pattern, of grave goods to the north and east, is common at Pusilha. Figure 7. Operation 8 Structure, showing locations of double crypt Burial 8/3 and royal tomb Burial 8/4. In 2005 we also discovered a royal tomb in the Op. 8 Structure. Photographs have been published of some of the artifacts found in and around the tomb (Braswell et al. 2005), including eccentrics made of chert and obsidian, a multitude of jade beads, the three diadems that formed a royal saq hunal headdress, and a polychrome tripod plate. Iconographic elements borrowed from Teotihuacan link some of the items in and around the tomb most clos ely to the first and the seventh rulers. Th e 14 vessels are late Tepeu 2 in form and decoration. They are most consistent in date with Ruler G, who died some time between A.D. 731 and 751. If our identification is correct, this is the first Belizean ruler to be excavated whose name is known and whose expl oits are described in hieroglyphics texts. With the exception of items of personal adornment, the distribution of grave goods within th e tomb follows the same pattern described for the crypt. Thirteen of the ceramic vessels were found along the eastern side of the tomb, a direction associated with resurrection. These 13 vessels and the food they contained therefore awaited the rebirth of the king from the underworld and his subsequent journey to the 13 levels of heaven. A second interesting pattern in the linear arrangement of these vessels was noted by Cassandra Bill (personal communication 2006). Polychrome and light-colored vessels were placed toward the north while black and dark brown vessels were located toward the south end of the eastern wall of the tomb. In this case, color associations may also indicate directions: north associated with light and color, south with darkness. The fourteenth vessel, a large and crudely made basin cont aining more than 80 items of jade, was discovered on the north side of the tomb: a direction associated with the heavens. Two small obsidian eccentrics were also found in the no rth. In contrast, no offerings were encount ered along either the west or south walls, di rections associated with death and the underworld. A final observation is that the tomb was reopened in antiquity. Rather than entering from the antechamber, the capstones spanning the tomb were removed, and some items including one of the three jade diadems forming the royal headdress were moved. Although there is no evidence of tomb reuse, a pattern cited by Leventhal (1990) as characteris tic of the southern Belize region, but also known at Caracol, the royal tomb of Pus ilha and the secondary burial in the elaborate crypt of Lower Group I indicate that interments were revisited. Conclusion Pusilha is a southern Peten site, with close ceramic, epigraphic, and iconographic ties to the Rio Pasion and Petexbatun 76

PAGE 83

Terminal Classic Occupation at Pusilha 77 regions. It also shar es certain cosmological ideas the association of north and up with the heavens, south with the ballgame and underworld, and east with the emerging sun and resurrection with most of the Maya region. But, like other sites in somewhat isolated and peripheral southern Belize, the inhabitants of Pusilha developed a distinct regional tradition that is expressed in burial practices, architectural forms, and site planning. The long independence of the southern Belize regi on no doubt contributed to the development of this tradition. Reference Cited Bill, Cassandra R., and Geoffrey E. Braswell 2005 Life at the Crossroads: New Data from Pusilha, Belize. In Archaeological Investigations in the Eastern Maya Lowlands : Papers of the 2004 Belize Archaeology Symposium, Volume 2. Institute of Archaeology, NICH, Belmopan, Belize, edited by J. Awe, J. Morris, and S. Jones, pp. 301-312 Braswell, Geoffrey E., & Sherry A. Gibbs 2006 In the Land of the Avocado: Recent Archaeological Investigations at Pusilh, Toledo District, Belize. In Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2005 Belize Archaeology Symposium Volume 3. Institute of Archaeology, NICH, Belmopan, Belize. edited by J. Morris, S. Jones, J. Awe, and C. Helmke, pp. 271-286. Braswell, Geoffrey E., Christian M. Prager, and Cassandra R. Bill 2005 The Kingdom of the Avocado: Recent Investigations at Pusilh, a Classic Maya City of Southern Belize. Anthropological Notebooks 11(1):59-86. Braswell, Geoffrey E., Christian M. Prager, Cassandra R. Bill, Sonja A. Schwake, and Jennifer B. Braswell 2004 The Rise of Secondary States in the Southeastern Periphery of the Maya World: A Report of Recent Archaeological and Epigraphic Research at Pusilh, Belize. Ancient Mesoamerica 15:219-233. Hammond, Norman 1975 Lubaantun: A Classic Maya Realm Monographs of the Peabody Museum, No. 2. Peabody Museum of Archaeology and Ethnology, Cambridge, MA. Leventhal, Richard M. 1990 Southern Belize: An Ancient Maya Region. In: Flora S. Clancy and Peter D. Harrison (eds.), Vision and Revision in Maya Studies pp. 124-141. University of New Mexico, Albuquerque. Marcus, Joyce 1992 Dynamic Cycles of Mesoamerican States: Political Fluctuations in Mesoamerica National Geographic Research and Exploration 8:392-411. 1993 Ancient Maya Political Organization. In: Jeremy A. Sabloff and John S. Henderson (eds.), Lowland Maya Civilization in the Eighth Century A.D, Washington, D.C.: Dumbarton Oaks, pp. 111-183. Morley, Sylvanus G. 1938 The Inscriptions of Peten Volume IV. Publication No. 347. Carnegie Institute of Washington, Washington, D.C.

PAGE 84

7 COMMONER RITUAL AND SOCIO-POLITICAL LIFE IN A LATE CLASSIC NEIGHBORHOOD: ARCHAEOLOGICAL INVESTIGATIONS AT THE NORTHEAST GROUP, CHAN, SITE, BELIZE Chelsea Blackmore While increased attention has been given to smaller-scale settlements, rarely is commone r ritual understood in the wider socio-political context of ancient Maya society. Investigations at the Northeast Group, a Late Classic neighborhood, have examined how the centralization of neighb orhood-wide ritual was key to the negotiation and formation of social identity among its occupants. Eviden ce recovered from excavations in 2004 and 2005 indicate that two groups, Clusters NE-1 and 3, were the focus of social and ritual activities within the neighborhood. Excavations have identified considerable diversity between households, suggesting that neighborhood occupants shaped social status and identity through the control of feasting and farming ceremonies. This ability to actively manipulate status forces archaeologists to consider how commoner activities, no matter how mundane, may have impacted wider social, political, and economic networks. Introduction The elaborate art and iconography of the Classic Maya have long provided testament to the importance of ritual in elite ancient Maya identity-formation and politics. But what of ordinary Maya people? Recent critiques (e.g. Iannone and Connell 2003; Lohse and Valdez 2004; McAnany 1995; Pyburn 1997) have illustrated the variability that exists between and within settlement types, yet the role that commoners played in creating and maintaining this variability remains underdeveloped. Although we speak confidently of them as laborers and producers, we usually ignore commoners as socially diverse individuals. If commoners are only as important as the materials they produce, then we can never understand the impact these populations had on wider social and political processes. If archaeological interpretations of ancient Maya society are based on the lives of the elite, then how can we accurately describe commoner populations, let alone ancient Maya society as a whole? Over the last two years, household research at the Northeast Group, part of the ancient agrarian village of Chan, Belize has investigated how its occupants built and manipulated social status and identity through the control of ritual materials and activities (Figure 1). Posthole testing and excavations have demonstrated considerable diversity between households based on architectural style, differences in household debris, and the identification of ritual caches and burials. Because the Northeast Group approximates a neighborhood, it would have acted as a locu s of intra-community interaction, maintained through social, economic and ritual ac tivities. If these spaces were the loci of community interaction, they were also the arenas in which identity and status were expressed and manipulated. Commoner Identity and Status Social identity is not simply a category we use for discerning differences in social position, it also refers to the way in which people interact and define themselves within a set of cultural rules. Whether we refer to households, neighborhoods, or communities, archaeological interpretations tend to focus on the in tegration of social Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 79-87. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 85

Investigations at the Northeast Group, Chan, Belize Figure 1 : Location of Chan Archaeological Site Figure 2 : The Northeast Neighborhood Group 80 80

PAGE 86

Investigations at Chan Site, Belize behaviors and activities. Yet integration does not necessarily negate differentiation; it involves the social negotiation of difference and sam eness, and often entails larger tensions between individuals, the group, and the state (Meskell 2001:89). Since people vary in their economic strategies, status posi tions, occupations, and access to goods and resources, there is usually little room for social homogeneity. The effects of state and class-based societies are not limited to seats of government, economic power, or religious authority. Those living in the periphery, margins or rural areas of society are part of the infrastructure and as a result play a crucial role in the maintenance and reproduction of society. Figure 3: Midden 2 of Mound Group 2, Cluster NE-1 Archaeologists often equate status and power with displays of wealth, the control of space, and ritual activities. Because the built environment reflects cultural attitudes, an examination of spatial arrangement, specifically the organization of space, may reflect aspects of social identity (Kent 1990). Similarly, control over ritual spaces and domestic activities therein allowed emergent elites to build and consolidate power (e.g. Lucero 2003; McAnany 1995). In the Maya area, ancestor veneration and the dedication and termination of structures were important aspects of domestic life, related to the cycles of the cosmos and acts of birth and renewal. Fire ritual and censing served to renew the community and household through the activation of its sacred spaces (Rice 1999; Stuart 1998). Feasting activities were dedicated in the commemoration of gods as well as the sacrifice and transformation of food into a sacred element (LeCount 2001: 941). While these revitalized and ensured the survival of the community, they also served as political and social currency. Lineage heads of both commoner and elite households would have been expected to host feasts, enact fertility rituals, and renew sacred spaces; thereby, legitimizing their position within a neighborhood, community, or region. Dissertation research at the Northeast Group has examined how interior and exterior spaces were utilized and defined by the groups occupants as a means to establish social standing within the community. By examining the function and organization of space and the corresponding distribution of material goods across the neighborhood, we can define the activities and behaviors in which people engaged; how these varied in relation to resource access; as well as the restriction and elaboration of particular spaces. If there is a direct connection between social status and the centralization of ritual activities, then evidence from higher-status households should include increased densities of ritual deposits, elaboration of architectural spaces, 81

PAGE 87

Investigations at the Northeast Group, Chan, Belize and greater access to econom ic resources and prestige items. Figure 4 : Location of Midden 5, Cluster NE-3 (confirmed by test-pit on western edge The Chan Community and the Northeast Group The Chan archeological site is an ancient agrarian village located in the Upper Belize River Valley, approximately 4km southeast of the minor polity-center of Xunantunich (Robin 1999, 2002) (see Figure 1). The Northeast Gr oup, one of a number of neighborhoods that characterize the Chan landscape, includes 6 mound clusters of variable size a nd spatial arrangement (Figure 2). Its size and organization fits definitions of neighborhoods found in Maya archeological and ethnographic resources. Bullard (1964:281), for example, defined neighborhoods as a collection of households organized in a hamlet-like cluster of from 5-12 structures. Vogt (1976), meanwhile, relates neighborhoods to waterhole groups of modern-day Zinacantan. Although thes e places had little formal spatial organization, scholars suggest (Blanton et al 1982; Pyburn 1997; Vogt 1976) that neighborhoods were organized on the basis of lineage, status, and possibly occupation. Set within broader social networks, neighborhoods bound people through various interactions that would have defined identity both within the neighborhood and across settlements within the Chan community. Investigations at the Northeast Group Household research at the Northeast Group has investigated how occupants built and manipulated social identity, specifically through the control of ritual materials and activities. Excavatio ns in 2004 focused on the systematic coverage and sampling of both on and off-mound areas while those in 2005 focused on more extensive horizontal and penetrating excavations of architecture in Clusters NE-1, 3 and 6 (Blackmore 2004, 2005). In 2004, a post-hole and test-pitting program encompassed approximately 40,000m2 of area. Posthole tests were placed at 6-m intervals along north-south transects throughout the neighborhood to delimit houselot boundaries. An additional twenty five 2x1m test-pits were excavated, placed in each of the six architectural clusters to elucidate construction history and investigate significant fi nds identified from posthole tests. Of the 812 postholes completed, 46% contained evidence of cultural remains and 13 of these were considered unusually productive. Further e xploration of these areas via test-pits resulted in the identification of at least 5 middens. Most notable among these were Middens 2 and 5, both of which produced a considerable diversity of material, both domestic and ritual in nature. Mate rial recovered from Midden 2, located directly behind mound Group 2, Cluster NE-1 included ceramics, lithics, groundstone, incensario fragments, and obsidian prismatic blades (Figure 3). Based on excavations along the western edge of Cluster NE-3 and the identification of productive post-holes near the cluster, Midden 5 was identified as a series of refuse deposits disposed along the perimeter of the 82 82

PAGE 88

Investigations at Chan Site, Belize group (Figure 4). Its th ick lens of refuse, (approximately 60cm thick in comparison to 20-30cm in Midden 2), suggests that it was used frequently through the lifecycle of the cluster. The diversity and quality of artifacts recovered from Clusters NE-1 and 3 in 2004 were unique when compared to elsewhere in the neighborhood (Blackmore 2004). The identification of highe r concentrations of middens, obsidian prismatic blades, polychrome ceramics, and effigy censer fragments suggested that Clusters NE-1 and 3 may have been the focus of social or ritual activities within the neighborhood. Research in 2005 focused on extensive horizontal and penetra ting excavations in Clusters NE-1, 3, and 6 to confirm whether these differences were in fact real. Excavations in Cluster NE-1 focused on Group 2, one of twothree mound patio groups (Structures 4, 5, & 6) that characterize the cluster (see Figure 3). Excavations focused on Structures 4 and 6, exposing both terminal architecture and internal stratigraphy through a series of twenty suboperations. Ceramic analysis conducted in 2006 indicated that Structure 6 was founded in the Early Classic and continuously occupied through the Late and Terminal Classic period. This contrasts with most other households in the Northeast Group, which are not es tablished until the Late Classic period. Prior to excavations, I suggested that Structure 4 was ritual in nature rather than an ancillary structur e based on its square shape and location on the groups eastern side. Based on the recovery and identification of two burials and the dearth of utilitarian items that would have indicated a storage or kitchen area, the excavations supported my initial thesis. Two interments were dedicated during its construction, a capped cist buried under the western staircase, and a looted burial recovered from the structures interior. In contrast to Structure 4, Structure 6 was a residential living space as implied by the identification of three benches, one three-course rectangular bench in the center, and two-Lshaped benches on either end. As is common to ancient Maya households, we also recovered a simp le stone-lined crypt found directly in fron t of the structures staircase. Although evidence suggests that Structure 6 is a residenc e, other architectural features imply a potential ritual focus of this structure in a manner pa rallel to that of Structure 4. While seeking the southeast corner of the structure, we identified a stonelined pit, most likely an emptied corner cache, adjacent to the easternmost bench. Another unusual architectural feature, a niche, was identified on the southeast corner of Structure 6, set into the stairs leading out of the patio group. Nothing was found in association with the niche, making functional assessment difficult. Its location on the platforms eastern edge which faces away from the platform suggests that its contents were visible to people living outside of the group. Cluster NE-3, a tightly arranged three-mound patio grou p, appears to have been occupied continuously from the early Late Classic through the Terminal Classic period (see Figure 4). Excavations in 2005 focused on the horizontal exposure of Structure 3, a small ancillary structure, and the eastern half of Structure 2, a larger residential building. In Structure 2s ultimate phase, it can be described as a 1.5m-high rectangular s ubstructure with an eight-course staircase and a 3.5 m rectangular bench at its summit. What distinguish Cluster NE-3 are its architectural detail and the concentration and diversity of remains and deposits recovered from a relatively modestly size d space. Three areas stand out; first, a small alleyway between 83

PAGE 89

Investigations at the Northeast Group, Chan, Belize Structure 2 and 3, second, a hearth and possible ritual or feasting deposit and finally a series of burials and caches identified in the center of the patio. In the alleyway between Structures 2 and 3, we recovered a number of worked pieces of limestone; shell, and jade (Figure 5), interspersed between concentrations of unslipped jar rims and necks. More than li kely, this material is an accumulation of trash, swept into the alleyway from the two adjoining structures. The second area of note is a niche built into the western edge of Structure 3 (Figure 6). Identified as a hearth based on the recovery of ash and animal bone, it appears to have been in use and modified during the last two construction episodes of Cluster NE-3. When first built, the hearth was 30cm deep and 45cm wide. Although the hearth was decreased in depth with the addition of the ultimate patio floor, it remained in use by the groups occupants as indicated by the recovery of ash in its final context. What makes the hearth important, however, is its capped portion, which became a ritual deposit sealed by the ultimate patio construction. This deposit was divided into two main segments: those materials found within the hearth and those scattered beyond it. The densest portion of the deposit was located in the hearth and included burnt animal bone, shell fragments, ceramic and lithic fragments, and a 20cm lens of ash. Outside of the hearth, the deposit extended an additional 85cm and included pockets of ash mixed with the surrounding fill context. The primary concentration of artifacts in this scatter, however, was located directly in front of the hearth including a single 20-cm unslipped calcite dish (see Figure 6). The entire deposit was then capped by the ultimate patio floor, a well-pre served plaster surface, 4cm thick. Figure 5 Alleyway trash from Cluster NE-3 The final area of interest in Cluster NE-3 is a series of burials and deposits recovered in a trench, measuring 4m long by 3.5m wide that cut through the approximate center of Cluster NE-3s patio (Figure 7). In 2004, a relatively large (2.3m NS x 1.0m EW) and well-constructed stone crypt was identified at the base of test-pit excavations (B1 in Figure 7). Osteological analysis indicates that two individuals were interred; the primary individual is a 35 to 50 year old male, placed underneath a female between the ages of 20 and 35 years old (Briggs 2004). In 2005, three additional burials were recovered. The first of these was a crypt burial containing at least three individuals (B2 in Fi gure 7). Identified adjacent to this was a second burial, a 65cm deep cist cut into bedrock. The interment itself, a single individual lying face down, was found at the base of the cist (underlying 40cm of fill) and was incredibly well preserved (B3 in Figure 7). To the east of this, we also identified a deposit of two lipto-lip vessels, which were covered by a third larger, unslipped plate (D1in Figure 7). Finally, while drawing the western section of our trench, we identified an additional burial, a small and poo rly preserved crypt. 84 84

PAGE 90

Investigations at Chan Site, Belize Figure 6 Ritual Deposit from Hearth, Str. 2, Cluster NE-3 The final group excavated in 2005 was that of Cluster NE-6. Identified as a one of the four lower-status domestic households in the neighborhood, these excavations provided an important comparative base line for understanding the unique complex of materials identified in Clusters NE-1 and 3. Cluster NE-6 is a small, two-mound patio group (both structures under 1m in height), located at the northern end of the neighborhood. Architecturally and materially Cluster NE-6 can be characterized as a strictly domestic house mound group with simple platform construction and domestic debris. In contrast, Clusters NE-1 and 3 were considerably more complex; excavations identifying staircases, benches, masonry architecture, multiple construction sequences, constructio n bins, jade, carved shell fragments, grave goods, niches, and corner caches; materials otherwise not found in Cluster NE-6. The identification of domestic and ritual material in Cluster NE-1 and NE-3 suggest they each had a specific ritual purpose. Given its cl ose proximity to a series of agricultural terraces, mound Group 2 of Cluster NE-1 may have been linked to farming and fertility ceremonies. The recovery of extensive midden remains, a hearth, and possible feasting deposit suggests that the occupants of Cluster NE-3 were lineage heads who hosted neighborhood-wide social activities. The burials and deposits of this cluster reaffirm this notion. The focus of activity in the center of the cluster reflects Maya concepts of directionality and reinforces this space as an axis mundi between the three levels of the universe. Whereas Cluster NE-1 included ritual and domestic activities, Cluster NE-3 appears to be focused on public displays and social activities. Conclusion The identification of Clusters NE-1 and 3 provide a unique opportunity to explore how the control of ritual and neighborhood activity may have been a principal component in the expression and definition of social identity. Preliminary assessments of the quality and diversity of household debris, architect ural style, and the identification of ritual caches and burials, suggests that the occupa nts of Clusters NE-1 and 3 were considerably more complex than other groups in the neighborhood. 85

PAGE 91

Investigations at the Northeast Group, Chan, Belize Figure 7: Burials and Deposits underlying the Patio of Cluster NE-3 In particular, Cluster NE-3 may have been the focus of feasting or other social and public activities for the neighborhood. Without question the quantity and concentration of burials within the cluster suggest that its occupant s were vested in the power of ancestor veneration. Cluster NE-6, in comparison, appears to be a simpler, more strictly domestic house-mound group. In the competition for social standing and power, residents of Clusters NE-1 and 3 were ascendant hosts, while neighbors in Cluster NE-6 appear to be guests or observers. As Hendon (2000:9) notes On the one hand, we tend to speak confidently of the elite and non-elite, rulers and ruled, without doing much to dissect the internal differences in these large groupings or their interactions. Because of the various limitations of archaeological data, we tend to focus on one aspect of social organization, while giving negligible treatment to intersecting facets. In particular, a focus on elite society has rendered commoners invisible and therefore unimportant to archaeological interpretations. While internal differentiation is an expected attribute of elite society, commoners are consiste ntly portrayed as simple farmers and part-time craft specialists bound together by community ritual. By basing our investigations on such assumptions, we lose sight of the internal variations that comprise social identity; the very root of interactions between individuals, groups, and societies. Acknowledgements: Funding for this project was made available by the National Science Foundation, Sigma-Xi, and the University of California, Riverside. I would like to thank the Belize Institute of Archaeology for their continued support of my dissertation research. This work would not have been possible without the guidance of my project director, Cynthia R obin, my dissertation advisor, Wendy Ashmore, and the Chan excavation team: Nester Alfaro, Yasmine Baktash, Bernadette Cap, Don Ventura Cocom, Don Virgilio Goodoy, Cirro Hernandez, Laura Kosakowsky, John Meierhoff, Glenis Smith, Elvis and Elmer Valdez, and Andrew Wyatt. As always, I owe a debt of gratitude to Shankari Patel for giving me the space and time to conduct my research so far from home. 86 86

PAGE 92

Investigations at Chan Site, Belize 87 References Cited Blackmore, Chelsea 2004 Preliminary Testing and Excavations at the Northeast Group, 2004. In The Chan Project: 2004 Field Season, edited by C. Robin. Northwestern University, Department of Anthropology. 2005 Excavations in the Northeast Group: Clusters 1, 3, and 6. In The Chan Project: 2005 Field Season edited by C. Robin. Northwestern University, Department of Anthropology. Briggs, Margaret 2004 Analysis of the Chan Burials. In The Chan Project: 2004 Field Season edited by C. Robin. Pp. 97-128. Northwestern University, Department of Anthropology. Bullard, William 1964 Settlement Pattern and Social Structure in the Southern Maya Lowlands during the Classic Period. In XXXV Congreso Internacional de Americanistas, vol. 1. Mxico City. Hendon, Julia 2000 Having and holding: Storage, Memory, Knowledge and Social Relations. American Anthropologist. 102, 42-53. Iannone, Gyles and Samuel V. Connell 2003 Perspectives on Ancient Maya Rural Complexity. In Perspectives on Ancient Maya Rural Complexity edited by Gyles Iannone and Samuel V. Connell, pp. 1-6, Cotsen Institute of Archaeology, Los Angeles. Kent, Susan 1990 Activity Areas and Architecture: an Interdisciplinary View of the Relationship between Use of Space and Domestic Built Environment. In Domestic Architecture and the Use of Space edited by Susan Kent, pp.1-8. Cambridge; Cambridge University Press. LeCount, Lisa J. 2001 Like Water for Chocolate: Feasting and Political Ritual among the Late Classic Maya at Xunantunich, Belize. American Anthropologist 10(4):935-953. Lohse, John and Fred Valdez, eds. 2004 Ancient Maya Commoners. University of Texas Press, Austin. Lucero, Lisa J. 2003 The Politics of Ritual: The Emergence of Classic Maya Rulers. Current Anthropology 44(4): 523-558. McAnany, Patricia A. 1995 Living with the Ancestors: Kinship and Kingship in Ancient Maya Society. University of Texas Press, Austin. Meskell, Lynn 2001 Archaeologies of Identity. In Archaeological Theory Today edited by Ian Hodder, pp. 187-213. Blackwell, Oxford. Pyburn, Anne 1997 The archaeological signature of complexity in the Maya lowlands. In The Archaeology of City States: Cross-Cultural Approaches edited by T. Charlton and D. Nichols. Smithsonian Institution Press, Washington, DC, pp. 155. Rice, Prudence M. 1999 Rethinking Classic Lowland Maya Pottery Censers. Ancient Mesoamerica 10: 25-50. R obin, Cynthia 1999 Towards an Archaeology of Everyday Life: Maya Farmers at Chan Noohol and Dos Chombitos Cik'in, Belize. Ph.D. dissertation, University of Pennsylvania 2003 Community Development at the Agrarian Village of Chan, Belize. Northwestern University. National Science Foundation Grant Stuart, David 1998 The Fire Enters His House: Architecture and Ritual in Classic Maya Texts. In Function and Meaning in Classic Maya Architecture, edited by Stephen D. Houston, pp.373-425. Dumbarton Oaks Research Library and Collection, Washington, D.C. Vogt, Evan Z. 1976 Tortillas for the Gods: a Symbolic Analysis of Zinacanteco Rituals. Cambridge: Harvard University Press.

PAGE 93

8 AN UNEXPECTED DISCOVERY IN THE CHAU HIIX SETTLEMENT: SUGGESTIONS FOR TERMINAL CLASSIC RESIDENCES IN NORTH-CENTRAL BELIZE A. Sean Goldsmith Settlement reconnaissances at Chau Hiix during the past several seasons have identified a multitude of lowmounded features akin to those recorded at ancient cities throughout the Maya Lowlands. The additional discovery of both ground-level (non-m ounded) as well as entirely buried occupation surf aces at Chau Hiix highlights the probability that demographic studies which refer to mounded features alone may under-represent population trends at many Maya sites. At Chau Hiix, at least one such surface-invisible feature was recently identified during systematic subsurface testing, and has been dated to the Terminal Classic Period. An expanded subsurface testing program at the site would seem likely to encounter many more such features, a suggestion partly supported by several ground-level stone a lignments and notable artifact scatters which have been ob served in empty terrain throughout the settlement area. While population decline around the end of the Classic Period is well attested at Lowland sites, shifts in non-elite residential strategies at Chau Hiix, especially as regards architectural investment, may significantly mask demographic continuities during this tumultuous transition. Introduction Cultural transformations during the Terminal Classic Period in the Maya Lowlands are postulated to have been variable in both so cio-political and demographical result (Bey et al. 1997; LeCount 1999; Lucero 2002; Webster and Freter 1990; Wright and White 1996). While these transformations are often evidenced by data collected from elite contexts, Terminal Classic Period panregional shifts in ec onomic flow, political alliances, or systems of ideology are also likely to have impacted local non-elite communities in ways visible to archaeological scrutiny (Ashmore et al. 2004; LeCount 1999; Masson 1997). Better yet, discernible shifts in agricultural, residential, or specialized productive strategies at the household level may inform our understanding of the actual mechanics of change across the transition between the Classic and the Postclassic Periods (cf. Robin 2003). Both of these avenues of inquiry require detailed study of rural household material deposition and community-level settlement patterning for this time period goals toward which only tentative steps have been made at most sites in Belize. This paper is a presentation of data derived from settlement area research at the site of Chau Hiix, in north-central Belize (Figure 1), and is inte nded to add new data to the overall corpus of information regarding Terminal Classic household archaeology in the re gion. The Chau Hiix Houselot Project was de signed to investigate the spatial nature of refuse deposition among a group of proximate non-elite domestic compounds (houselots) at a single, relatively small Maya centre. The goals and results of this project have been described in a previous Belize Archaeology Symposium volume (Goldsmith 2005), and are therefore summarized only briefly below. The Chau Hiix Houselot Project Chau Hiix is a modestly sized Maya city in north-central Belize, relatively equidistant from Lamanai to the west, and Altun Ha to the east. The city has its origins near the beginning of the Middle Formative period (approximately 1000 BC), and was Research Reports in Belizean Archaeology Vol. 4, 2007, pp.89-99. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 94

A. Sean Goldsmith Figure 1: The location of Chau Hiix and other nearby sites probably occupied cont inuously up until the Colonial Period (Andr es n.d.; Andres and Pyburn 2004). Work at Chau Hiix has been ongoing since the earl y 1990s under the direction of Dr. Anne Pyburn of Indiana University, and with the gracious support and encouragement of the people of Crooked Tree Village, who have curated and protected the site throughout the 20th century. Settlement pattern research at Chau Hiix has been conducted sporadically since the mid 1990s (Cook 1997; Cuddy 2000). A systematic survey, posthole augering, and testpitting program within a large region to the north of the site ce ntre was conducted by the author and several assistants between 1999 and 2003, and was designated the Chau Hiix Houselot Project (Goldsmith 2006). The Chau Hiix Houselot Project was designed to explore th e spatial distribution of subsurface cultural material across a portion of non-elite settlement at Chau Hiix. The purpose of the project was to illustrate the high data-collection potential of systematic subsurface sampling over large areas. The project followed a three-stage methodology as follows: Stage 1 involved surface mapping of settlement features over a large portion of terrain to the north of the site centre (Figure 2), and resulted in the identification of numerous mounds, terraces, water catchment features, and so on. Stage 2 employed a systematic subsurface sampling program, using posthole augers, over a smaller area w ithin the Stage 1 mapping zone (Figure 3). The third and final stage of the fieldwork, conducted during the 2003 field season, involved the excavation of targeted 1 x 1 metre test pits in areas of subsurface artifact density as 90

PAGE 95

Investigations at Chau Hiix Figure 2 : Site centre and mapped settlement of Chau Hiix Figure 3 : Locations of all Stage 2 auger tests 91

PAGE 96

A. Sean Goldsmith identified by the auger testing program (Figure 4). Figure 4 : Locations of all Stage 3 testpits The auger-testing program resulted in the collection of substantial quantities of artifact material. Almost all of the postholes contained artifacts of some kind, and what became clear during the fieldwork was that there was a great deal of variation in the densities of artifacts from one part of the sampling area to the next. To illustrate this variation, all posthole artifact data were entered into an electronic database, and subsequently represented in graphic form using contour mapping software to show spatial artifact density distributions. Figure 5 shows one such example, and is a representation of the sp atial distribution of subsurface ceramic densities in the study area. Areas of greater contour indicate areas with higher numbers of ceramic sherds. The resulting maps of artifact density distribution allowed the tentative delineation of houselot boundaries within the study area. The concept of the houselot is enjoying increased popularity as a methodological tool for data collection in the Maya area (Deal 1985; Hayden and Cannon 1983; Killion 1990; Robin 1999). The dual intent of the Chau Hiix Houselot Project was (a) to delin eate houselots in the study area, and then (b) to draw representative artifact samples from each houselot in order to make inferences about household activity. The primary result of the Chau Hiix Houselot Project was the identification of five discrete houselots within the study area (Figure 6), including th ree that contained visibly mounded archite ctural remains, and an additional two that did not exhibit mounded features of any kind. Ceramic materials recovered from all five of these houselots suggested occupation primarily during the Early Cla ssic Period (see Goldsmith 2006 for further discussion of this Early Classic community). Data relating to other parts of the lengthy Chau Hiix chronology were limited. However, some evidence for Terminal Classic activity in this part of the site is suggested, and was recovered from the southwest portion of the study area. Figure 5 : Subsurface ceramic density variation in the study area 92

PAGE 97

Investigations at Chau Hiix Terminal Classic Period Evidence at Chau Hiix Architectural excava tion s in the elite monumental core of Chau Hiix have demonstrated not only that the city was occupied during the Late Classic to Terminal Classic transition, but also that the rulers undertook significant architectural projects at this time (Andres n.d.). Limited excavations into mounded remains immediately adjacent to the central architectural precinct have suggested the presence of a supporting population in the later parts of the Chau Hiix chronology, but unequivocal settlement evidence dating to the Terminal Classic period at Chau Hiix is virtually non-existent. Figure 6 : Extrapolated houselot boundaries as based on subsurface densities During Stage 2 of the Chau Hiix Houselot Project, it was not uncommon for excavated auger holes to contain cobble fill matrices of the sort commonly found as subfloor ballast in monumental architecture and housemounds alike (Figure 7). In most instances, the locations of cobble fill auger holes were explained by reference to visibly mounded features within the study area. Other locations were further explained by reference to what were ultimately found (through test pit excavati on in the final stage of the project) to be occupation floors entirely invisible to surface scrutiny. But within the southwest portion of the study area, the discovery of cobble fill material in two auger locations was not immediately explained. To address this enigma, a 1 x 1 metre excavation unit was placed in the vicinity of one of th e postholes in question (see Figure 4, Testpit 46). The terrain was highly unremarkable from the perspectives of both surface topography and surface evidence of cultural material (Figure 8), and were it not for the presence of cobble fill matrices in the auger holes themselves, there would have been no compelling reason to suspect cultural activity in this location whatsoever. Figure 7: Locations of auger holes containing cobble fill matrices Excavation of the 1 x 1 metre test pit proceeded in similar fashion to all test pits excavated during the 2003 season. Soil matrices were removed in contextual fashion, with the identification of contexts primarily based on the visual clues provided by soil color changes, changes in matrix consistency, the appearance of new material types, or marked changes in artifact 93

PAGE 98

A. Sean Goldsmith densities. All sediments were screened through m esh to ensure the recovery of smaller artifacts such as shell or obsidian fragments. The final ve rtical profile of Test pit 46 is shown in Figure 9. The transition between contexts 1 and 2 of this test pit almost certainly represents the shattered remains of a plastered occupation surface, with context 2 constituting subfloor cobble ballast of a nature highl y typical of subfloor contexts throughout the settlement area. In fact, in the absence of chronological markers to the contrary, this occupation surface would likely have been assigned an Early Classic period date on the basis of its similarity, and by proximity to several Early Classic period houselots within the study area. Figure 8 : Flat terrain in the vicinity of Testpit 46 Within the sub floor context, however, and presumably deposited there at the time of the construction of the occupation platform itself, were ceramic fragments dated to the Terminal Classic period (Figures 10 and 11), in particular sherds assigned generally to the type Daylight Orange (R. Fry, personal communication). The location of these ceramics in a sealed context below the occupation surface strongly suggests that the construction of the low platform itself took place in or around the Terminal Classic Period. The cobble fill matrices in the proximate auger holes as well as the recovery of evidence, in Testpit 46, for the construction of a low plastered platform is thus interpreted to represent the occupation of this locality by Terminal Classic inhabitants, who for one reason or another did not construct a residential platform of sufficient stature to be preserved as a mounded feature on the modern landscape. The very existence of this occupation surface, however, in a location where occupation was previously unsuspected, contains certain implications worthy of further discussion. Figure 9: Profile of Testpit 46, showing contexts and cobble subfloor ballast Discussion From the perspective of analytical cleanliness, the fact of having a Terminal Classic period houselot stuck as a palimpsest on top of an otherwise completely Early Classic community is, to say the least, somewhat inconvenient. One of the underlying goals of the Chau Hiix Houselot Project was to be able access the notion of community as a concept relating to agency and daily practice (sensu Bourdieu 1977, Giddens 1984), and th e Terminal Classic 94

PAGE 99

Investigations at Chau Hiix 95 material was clearly no t part of the Early Classic community. But there are certain implications in the discovery of this later residential space that are worth exploring. The use of numerical formulae in the 1960s and 1970s to estimate ancient populations was based primarily on counts of surface mounds as a basic quantifiable datum (Casselberry 1974; LeBlanc 1971; Naroll 1962). Household archaeological research during the pa st two decades has repeatedly shown that entire domestic areas remain hidden from surface view at numerous Maya sites (Chase 1990; Cliff 1988; Johnston 1994; Pyburn 1989; Wilk and Wilhite 1991), and as a result, have the potential to skew the accuracy of population estimates based on surface features alone. The most straightforward solution to this potential issue has been to add a percentage to the visible mound total, which would account for the features that were invisible on the surface. Significantly, however, the general trend has been to add the same percentage to estimates from all different time periods. The underlying assumption, whether a conscious one or not, is that the behaviours and processes which ultimately made some residences visible and others invisible were constant throughout PreColumbian times. That assumption is something we may want to consider more carefully, because while apparently logical on the surface, it fails to take into account factors that operate at the level of settlement behaviour. The reasons why some residential surfaces are visible as mounds and others are not depend on several factors. Some were certainly social as in a conscious attention to symbols of status. The equation of status with stature in architecture, while probably not valid in all cases, has numerous precedents throughout the Maya area. The increased labour needed to build a large platform as compared to a small one indicates the ability to garner such labour in the first place. On the other hand, other reasons for elevating a residential floor may have been more practical as in a preference to maintain elevation against flooding. Certainly at Chau Hiix, no prominent ridges or hills exist within the extended settlement zone to offer natural defense against seasonal inundation. Still other reasons are certainly related to ch ronology, as it is well known that rural domiciles were sometimes built upon earlier floors, and thus gained a mounded appearance through accretion. Yet others may be complete ly post-occupational, such as the variable ab ility of soil to build up in one area as compared to another, and thus not related to th e original residential strategy at all. The key point is that not one of these factors can be assumed to have remained constant throughout the history of the Maya civilization, even at single sites or within single local communities. Nevertheless, on the occasions that subsurface settlement features are considered in population estimates at all (and they usually are not), the methodology involves a certain percen tage being added to the number of people that lived in a community at any given time. Reference to population using that same method was even employed (albeit briefl y) by the Chau Hiix Houselot Project during discussions of the Early Classic Period community there: three houselots contained visibly mounded groups, and to that were added the two subsurface houselots discovered through posthole augering, with the conclusion that forty percent of the Early Classic population at the city of Chau Hiix might have lived in houselots that do not show on the surface today. But such an additive process to the analysis of population encounters more serious problems when the surface

PAGE 100

A. Sean Goldsmith Figure 10 : Terminal Classic Period plate fragm ents from Context 2 of Testpit 46 Figure 11: Daylight Orange Darknight Variety vessel fragment from Testpit 46 percentage is co mpletely missing. Specifically, in the case of the Chau Hiix Houselot Project ther e are no surface traces of Terminal Classic houselots in that same area nor in the settlement that immediately surrounds it. While this may sound like an issue of mathematical semantics, the addition of forty perc ent (or fifty, or one hundred percent) to a nonexistent dataset still results in nothing. So the formula calculation for Early Classic houselots in this part of Chau Hiix would be inappropriate for the Terminal Classic, where the entire sample size (and admittedly, the sample size is a single locality) comes from the subsurface universe. The discovery of evidence for a strictly Terminal Classic houselot in a place where only those dating to the Early Classic were suspected raises the possibility that many more invisible Terminal Classic houselots will be found at Chau Hiix. If such a phenomenon turns out to be the case, then at some point it will be necessary to reassess our images of economic and social process at the end of the Classic Period in that city. Increasing percentages of ground96

PAGE 101

Investigations at Chau Hiix level dwellings would, in that cas e, represent a residence shift that begged interpretation as a function of a be havioural change, a change, furthermore, that would likely inform our understanding of the events surrounding the Classic to Postclassic transition. Does a shift to ground-level dwellings also reflect economic or wealth shifts among Maya non-elites? Does the apparent reluctance of at least one Terminal Classic household to e xploit the existing mounded platforms of a nearby abandoned Early Classic houselot have a meaning that we can discern? Ar e there any invisible houselots at Chau Hiix that date strictly to the Late Classic Period, or to the Postclassic Period, and if so, does the proportion of these non-mounded residences show any long term trends throughout the history of the city? The resolution to these and many other questions does not lie strictly in the realm of surface-based settlement pattern studies, because in at least one case at Chau Hiix, all of the available data have come from below the surface of the ground. Conclusions The results of the Chau Hiix Houselot Project demo nstrate strongly that systematic subsurface testing brings an important additional level of visibility to the total settlement universe. At Chau Hiix, even given the limited size of the area where such systematic augering was conducted, we were able to make infe rences about parts of the Chau Hiix chronology for which information had been lacking up to that point, the Terminal Classic being one of those time periods. Given the numerous reasons why people might decide to build their houses either on the tops of mounds or on flat ground, it seems reasonabl e to expect that one or the other preference might hold sway at some points in history and not in others. The intention of this discussion has certainly not been to suggest that the small sample size currently available for Chau Hiix should be held up as a model for population estimation in the Terminal Classic across the Maya Lowlands. The data are presented here primarily becau se they are thought provoking in their char acterization of a possible residence shif t at a time when populations are assumed to have been falling at most Lowland Maya centres. If Terminal Classic period residences simply became more difficult to find on the modern landscape, then a ny assumption of population decline at Chau Hiix at the end of the Classic Period would be based on a methodological gap rather than an actual settlement pattern. Naturally it will take a good deal more similar fieldwork at Chau Hiix to determine whether or not this is the case. With any luck, further subsurface research in the Chau Hiix settlement will bring more surface-invisible houselots into the light where they belong. Acknowledgements I would sincerely like to thank the Institute of Archaeology, National Institute of Culture an d History for inviting me to participate in the 4th Belize Archaeology Symposium. Support for the fieldwork of the Chau Hiix Houselot Project was provided by the Indiana University Chau Hiix Archaeological Project, and by a research grant from the University of Calgary. This project would not have been possible without the assistance of the students of the 1999, 2001, and 2003 Indiana University Chau Hiix Fieldschools, and through the enthusia stic contributions of fieldworkers from the village of Crooked Tree. My research has benefited greatly from discussions with several of my colleagues on the Chau Hiix project, including especially Christopher Andres, Sarah Wille, Gabriel Wrobel, Alfredo Minnetti, Thomas Cuddy, and Leslie Sering. Dr. Anne Pyburn, Director of the Chau Hiix 97

PAGE 102

A. Sean Goldsmith Archaeolog ical Project, has been a constant source of guidance and mentorship, and I thank her for her continuing support and suggestions. References Cited Andres, C. and A. Pyburn 2004 Out of Sight: The Postclassic and Early Colonial Periods at Chau Hiix, Belize. In The Terminal Classic in the Maya Lowlands edited by A. Demarest, P. Rice, and D. Rice. Westview Press, Boulder. Ashmore, W., J. Yaeger, and C. Robin 2004 Commoner Sense: Late and Terminal Classic Social Strategies in the Xunantunich Area. In The Terminal Classic in the Maya Lowlands edited by A. Demarest, P. Rice, and D. Rice, pp. 302323. University Press of Colorado, Boulder. Bey, G., C. Hanson, and W. Ringle 1997 Classic to Postclassic at Ek Balam, Yucatan: Architectural and Ceramic Evidence for Defining the Transition. Latin American Antiquity 8(3):237-254. Bourdieu, P. 1977 1977 Outline of a Theory of Practice Cambridge University Press, Cambridge. Casselberry, S. 1974 Further Refinement of Formulae for Determining Population From Floor Area. World Archaeology 6(1):117-122. Chase, D. 1990 The Invisible Maya: Population History and Archaeology at Santa Rita Corozal. In Precolumbian Population History in the Maya Lowlands edited by D. Rice and P. Culbert, pp. 199-213. University of New Mexico Press, Albuquerque. Cliff, M. 1998 Domestic Architecture and the Origins of Complex Society at Cerros. In Household and Community in the Mesoamerican Past edited by R. Wilk and W. Ashmore, pp. 199-226. University of New Mexico Press, Albuquerque. Cook, P. 1997 Basal Platform Mounds at Chau Hiix, Belize: Evidence for Ancient Maya Social Structure and Cottage Industry Manufacturing Ph.D. Dissertation, University of Arizona, UMI Dissertation Services, Ann Ar bor, Michigan. Cuddy, T. 2000 Socioeconomic Integration of the Classic Maya State: Political and Domestic Economies in a Residential Neighborhood. Ph.D. Dissertation, Columbia University. Deal, M. 1985 Household Pottery Disposal in the Maya Highlands: An Ethnoarchaeological Interpretation. Journal of Anthropological Archaeology 4:243-291. Giddens, A. 1984 The Constitution of Society: Outline of a Theory of Structuration. University of California Press, Berkeley. Goldsmith, S. 2005 Explorations of an Early Classic Community at Chau Hiix. Research Reports in Belizean Archaeology 2: 2006 The Houselots of Chau Hiix: A Spatial Approach to the Study of Non-Elite Domestic Variability at a Small Maya City Ph.D. Dissertation, Department of Archaeology, University of Calgary. Hayden, B. and A. Cannon 1983 Where the Garbage Goes: Refuse Di sposal in the Maya Highlands. Journal of Anthropological Archaeology 2:117-163. Johnston, K. 1994 The Invisible Maya: Late Classic Minimally Platformed Residential Settlement at Itzan, Peten, Guatemala Ph.D. Dissertation, Yale University, UMI Dissertation Services, Ann Arbor, Michigan. Killion, T. 1990 Cultivation Intensity and Residential Site Structure: An Ethnoarchaeological Examination of Peasant Agriculture in the Sierra de los Tuxtlas, Veracruz, Mexico. Latin American Antiquity 1(3):191-215. 98

PAGE 103

Investigations at Chau Hiix 99 LeBlanc, S. 1971 An Addition to Narolls Suggested Floor Area and Settlement Population Relationship. American Antiquity 36(2): 210-211. LeCount, L. 1999 Polychrome Pottery and Political Strategies in Late and Terminal Classic Lowland Maya Society. Latin American Antiquity 10(3): 239-258. Lucero, L. 2002 The Collapse of the Classic Maya: A Case for the Role of Water Control. American Anthropologist 104(3): 814-826. Masson, M. 1997 Cultural Transformation at the Maya Postclassic Community of Laguna de On, Belize. Latin American Antiquity 8(4): 293316. Naroll, R. 1962 Floor Area and Settlement Population. American Antiquity 27(4): 587-589 Pyburn, A. 1989 Prehistoric Maya Community and Settlement at Nohmul, Belize BAR International Series #509, Oxford. Robin, C. 1999 Toward an Archaeology of Everyday Life: Maya Farmers of Chan Noohol and Dos Chombitos Cikin, Belize. Ph.D. Dissertation, University of Pennsylvania, UMI Dissertation Services, Ann Arbor, Michigan. 2003 New Directions in Classic Maya Household Archaeology. Journal of Archaeological Research 11(4):307-356. Webster, D. and A. Freter 1990 Settlement History and the Classic Collapse at Copan: A Redefined Chronological Perspective. Latin American Antiquity 1(1): 66-85. Wilk, R. and H. Wilhite 1991 The Community of Cuello: Patterns of Household and Settlement Change. In Cuello: An Early Maya Community in Belize, edited by N. Hammond, pp. 118 133. Cambridge University Press, Cambridge. Wright, L. and C. White 1996 Human Biology in the Classic Maya Collapse: Evidence from Paleopathology and Paleodiet. Journal of World Prehistory 10(2): 147-198.

PAGE 104

9 THE CURSE OF THE WARE: USING CERAMIC SYSTEMS IN BELIZE James J. Aimers The Terminal Classic period was one of rapid cultural change across the Maya lowlands. The subsequent Postclassic period was also one of intense interregional interaction across Mesoamerica, but this surely had roots in the Terminal Classic and earlier. Ceramics are one of the most important artifact classes for the identification of intersite and interregional interaction. Petrographic and chemical analyses can help trace the movement of vessels, but most ceramics were made and used locally, and the style of such ceramics can help map the exchange of ideas across the Maya world. In Belize, type-variety is the domi nant method of ceramic classification for intersite stylistic comparison, but there are currently substantial inconsistencie s in type-variety methods, especially in the use of ware classifications, that are hindering our ability to understand sty listic interaction. Here, I advocate the use of ceramic systems in Belize. Part of the original formulation of type-variety in the 1950 s, but never adopted in Belize, ceramic systems can group stylistically similar ceramics from across the Maya world and help solve some of the problems of ware in type-variety. I illustrate the value of ceramic syst ems in assessing intraand inter-regional interaction with specific Terminal Classic examples. Introduction For those of us involved in the arcane and often solitary study of ceramic method and theory, the 2005 Society for American Archaeology (SAA) symposium Squaring the Spheres was unusually exciting For two hours a number of archaeologists engaged in a lively, complex, and often impassioned discussion that made one thing clear: the ware concept, once a somewhat fuzzy but benign part of type-variety ceramic classification has been transformed since the 1960s into a snarling beast that now threatens our attempts to use Maya ceramics for anything other than description and chronology, and may even threaten those basic roles. In res ponse, I would like to advocate the adoption of ceramic systems in Belize. In ceramic systems one uses style (especially surface finish and decoration) to group similar ceramics without reference to their compositional and technological characteristics (i.e., paste or fabric). Furthermore, ceramic systems designations can provide a preliminary classification of individual pots or sherds while ware, group, and type designations are pending. Ultimately, systems designations tell us something about intersite and interregional interaction. Petrog raphic and chemical analyses can help trace the physical movement of vessels but since the vast majority of Maya ceramics were made and used locally (see e .g., Arnold 1980; Foias 2002; Fry 2003; LeCount, et al. 2002), we must continue to refine our use of the style of locally-made ceramics to map the exchange of ideas between sites and regions. Although the ceramic system concept was part of the original formulation of typevariety in the 1950s (see Wheat, et al. 1958), to my knowledge it has never been used in Belize. Ceramic systems will not kill the beast that is ware, but it may at least provide a way to compare ceramics among sites and regions without depending upon ware. The Nature of the Beast Prudence Rice (1976) identified the basic problem with ware three decades ago: Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 101-110. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 105

James Aimers Figure 1 : Pabellon Modeled-carved from Uaxactn (Smith 1955:Figure 88 i-n ) Figure 2 : Balantun Black-on-slate from Chichn Itz (Smith 1971:Figure 14a-c) wares have been defined in terms of both paste and surface, yet these are technologically independent and should be analysed separately. Since then, Rice has promoted the concept of paste ware (Rice 1987) to address this problem and her work with Leslie Cecil in the Petn Lakes has shown that paste wares can help address questions about ancient Maya social, political, and economic organization (see e.g., Cecil 2004). 102

PAGE 106

James Aimers Although most people seem to agree that surface colour is an unreliable criterion upon which to base ware designations, and Rices 1976 article is widely cited, not everyone agrees with the emphasis she places on paste. This was brought into focus at the 2005 SAA symposium when I raised a problem that occurs frequently in ceramic analysis: How does one classify two vessels that are stylistically very similar, or identical, but have visibly different pastes? This happens both within assemblages from single sites and, more often, between sites. I was interested to find that there were two an swers given to this question, and both methods are currently in use in Belizean archaeology. Basically, the differences revolve around whether ware is used as a hierarchical category above group, or is used modally, like shape, and thus nonhierarchically. Lumpy and Splitty The lumpy answer was to place the two pots in the same type (and thus the same group and ware), but to note the paste variation as part of the type description. Thus, Lumpers treat paste as a modal characteristic that cros scuts types, as vessel form or lip shapes do, and there is plenty of evidence that pastes of ten do crosscut types. I asked if the fabric variation should then be moved down to the varietal level and some people felt this was reasonable, others did not. Willey et al. (1967:304) foreshadowed this approach when they expressed reservations about ware as a hierarchical concept above group. Later, Ball (1977:3) noted that: it is taxonomically conceivable that variety differences within a type could be of technological (ware) magnitude, i.e., that a single type could crosscut two or more wares via two or more technologically distinguishable varieties." The splitty contingent noted that by de-emphasising paste, the lumpy approach downplays important productionrelated questions such as the identification of multiple production locales. They also noted that type-variety is a taxonomic classification system and thus must be hierarchical and the goal of classification is to create high within-group homogeneity (Rice 2005:4). The Spli tters would place the two stylistically similar pots in different wares, and therefore in different groups and types (and possibly varieties). This is the dominant hierarchical type-variety method (see e.g., Gifford 1976; Sabloff and Smith 1970; Willey, et al. 1967:304) but Lumpers have abandoned it in part because it has the end result of creating so many types. All this splitting, the Lu mpers countered, just makes understanding conceptual links between pottery types more difficult. At least two observations flow from this apparent impasse. The first is that the way one approaches type-variety has much to do with the questions one is asking. If you are interested in the stylistic choices consumers make and their implications for intersite interaction you are likely to be a Lumper; if you are interested in productionrelated questions like the number of workshops producing a given style around a site you are likely to be a Splitter. The nature of the sample is also relevant (e.g., eroded sherds without slips are likely to lead to splitting based on paste). Several participants acknowledged these issues and many felt comfortable that the ware problem has led to at least two separate methods in type-variety. I was not as comfortable with this because it suggested to me that we are creating confusion about exactly what groups, types and varieties represent and endangering our ability to consistently compare ceramics from different sites. A third possible solution to the Curse of the Ware is to disregard ware entirely (Willey, et al. 1994). I agree with Rice that the old surface-plus-paste version of ware must be discarded, but I worry about throwing the baby out with the bathwater. 103

PAGE 107

James Aimers It seem s to me that observable differences in paste should not simply be ignored because they undoubtedly mean something and these differences are going to be considered eventually if one is looking at the pottery closely. My opinion is that if some Mayanists continue to deal with paste variation at the ware level while others place it at the type or variety level, or not at all as they choose, we are creating a situation that creates more problems than it solves. Essentially we will no longer know what others mean when they say Type X and one of the major advantages of typevarietyperhaps THE major advantage-is that it provides common language (or a form of shorthand) that al lows archaeologists at different sites to talk to each other about ceramics. I suspect we are about to create a monster of a problem that will be very difficult to solve once the results of these varied methods become fossilized in print, especially since methods are rarely described in detail. Wither Ware at Lamanai? On the Lamanai project we have been discussing these issues intently (e.g., Aimers and Graham 2005). From the start of the project in 1998 the intent was to approach the ceramics in a number of different ways and to employ multiple, independent analyses and then to review the results before actually proposing any one or more kinds of classifications. Our approaches include contextual analysis along the lines of Pendergast and Grahams already published work (e.g., Graham 1987; Pendergast 1985), petrographic analysis (Howie 1996), iconographic analysis (John 2007), and type-variety/ modal analysis (Powis 2002). In the next year or so we will compare our results and see where there are points of contrast and overlap. Lumping While Splitting In terms of the type-variety classification at Lamanai, I believe I have come to a compromise between lumping and splitting, although my thoughts on the method described here are evolving and open to revision. Curren tly, I plan to place stylistically similar vessels in pastes that are macroscopically different along at least two axes of variation (e .g., coarse tan paste versus fine red paste) in different wares (and thus sometimes create new groups, types, and possibly varieties) following the dominant splitty type-variety method. But what about macroscopically identical pastes that have microscopic or chemical differences; that is, diff erent paste recipes? These different paste recipes have tremendous potential to tell us about production location and even about the presence of different pottery making groups (for example, workshops) in or around the site. I want to acknowledge these recipes in my classification, but I simply cant place these microscopic differences at the traditional high level of ware for at least two reasons: 1. We would have to conduct petrography on virtuall y our entire sample and although this is an admirable goal, it is not feasible in terms of time and money, 2. Creating wares for each paste recipe identified by petrography would create an unmanageable number of groups, types, and varieties. Thus, microscopic paste differences will be split at the lowest hierarchical level, variety, which also means they are lumped at the type level as some Lumpers do. Nevertheless, I pulled back from placing all paste variation as modal qualities of types, or at the varietal level wh en I revisited Rice and Cecils work in the Petn Lakes. There, they have several stylistically similar types of Postclassic red dishes, in at least three macroscopically different pastes. These macroscopically identifiable paste wares (in 104

PAGE 108

James Aimers Rices term s) align well with ethnohistorically known socio-political groupings around the lakes. It appears that the three groups were interacting enough to share styles but maintained different paste recipes and production techniques (Cecil 2001). If Rice and Cecil had simply ignored macroscopic paste variation and placed all these similar red dishes in one ware as the Lumpers might, they would have missed a valuable opportunity to tease out social groupings in the Petn Lakes (Cecil 2001, Rice 2005). One more crucial issue is that if I try to use paste ware as an important criterion in my classification, I cannot reasonably identify a type in the Lamanai collection as one named at, say, Mayapn, without having physically examined sherds from Mayapn. Furthermore, Howies work with clay sampling and petrographic analysis at Lamanai suggests that most clays used were from near the site, a lthough in at least one case temper may have been imported (Howie, et al. 2005). Although these results are from preliminary reports, it seems likely that most of the pottery at Lamanai was made locally. Does it make sense, then, to give ceramics from Lamanai names from other sites when the paste wares are likely to differ due to different production locations, techniques, and raw materi als? This at times has seemed an almost irresolvable issue in my analysis. However, following the work of Patricia Urban and others in Honduras, Ive adopted ceramic system as a partial solution to this problem. Ceramic Systems A ceramic system was defined in an unpublished 1959 paper by Phillips and Gifford (quoted in Forsyth 1983:9) as a "horizontal arrangement of roughly contemporaneous pottery types that are related in particular fr om the standpoint of design style and surface manipulation and which range over a wide area." Therefore, I use overall stylistic sim ilarities to link the ceramics of Lamanai to ceramics elsewhere through system. Even if I give some types Lamanai-specific names because they have distinctive pastes, systems designations will tell other archaeologists where to look for stylistically (conceptually) analogous types. I may eventually further split the Lamanai examples into one or moreLamanai specific varieties based on microscopic or chemical technological qualities, but my initial paste ware categorization is based on what can be detected by touch, the naked eye, and a 10X lens. This is in part a practical decision: to create paste ware de signations based on qualities that are not readily observable in the field is to require petrography by anyone wishing to compare their ceramics to those of Lamanai, and also runs the risk of creating a huge number of wares, groups and types. The rationale is also theoretical my questions at Lamanai at this point concern the sites chronology and its relationship with other sites. Howie is addressing questions of manufacture at the site and for her paste recipes are more important. Below, I discuss two well-known Terminal Classic types in a systems perspective. Pabellon Modeled-carved and its Imitations Pabellon Modeled-carved (Figure 1) was initially classified as part of the Altar Group of Fine Orange Ware at Uaxactn (Smith 1958), but similar ceramics have been found in most of the lowlands as far southeast as the Ula Valley (Lopiparo, et al. 2005) in a variety of pa stes other than Fine Orange and in shapes and designs that also differ from the Petn. In the Petn lakes district imitation Pabellon Modeled-carved occurs in four distinct pastes (Rice and Rice 2004). Setting aside formal (or stylistic) differences for a moment, if you follow the standard splitty ru les of type-variety, ceramics that look like these in anything 105

PAGE 109

James Aimers other than the distinct ive Fine Orange W are paste must be give different ware, group, type, and variety names. Awe (1985), Graham (1987), Helmke et al. (1998), and others have proposed th at the Belize Valley versions of Pabellon Modeled-Carved are distinctive enough to deserve a new designation in part because they are generally made of calcite paste. There is also a similar type in a fine red paste in the Cehpech sphere called Sahcaba Modeledcarved, but this type also occurs in the Petn (including Uaxactn) a nd could occur here in Belize (Ball 1976). Further complicating the issue is the fact that Sahcaba Modeledcarved is placed in the Teabo Group of Puuc Red Ware at Uaxactn (Smith and Gifford 1966) and Mayapn (Smith 1971) whereas it is placed in the Terminal Classic Tinaja Group of Petn Gloss Ware just over the border from Belize in the Mopn drainage (personal observation, Atlas collections, Dolores, Guatemala; Laporte 2006:18). Thus, one question raised for analysts who find similar samples in Belize is, simply, what to call them. This is where I think ceramic systems have great potential for type-variety. Even using drawings or photographs it is easy to tell if a vessel is st ylistically analogous to Pabellon Modeled carved. One can then place the new sample in the Pabellon Modeled-carved System and thenand this is the part that I think is really advantageousleave it at that until a more detailed paste analysis can be conducted and more accurate names assigned (or created, if the paste is new). Even the few examples of modelledor molded-carved sherds and vessels I have seen from the Belize come in a range of pastes. Some appear to be volcanic ash tempered, others appear to be calcite tempered and some are Fine Orange (i.e., true Pabellon Modele d-carved). However, systems assignments alone tell us a great deal about what these ceramics represent: some sort of interaction between the Belize Valley and the northwestern Petn and probably the Gulf Coas t. Nevertheless, system designations carry absolutely no connotations regarding production locale and they dont require ware, group, type or variety designations. Yet, I believe that systems designations alone have the added potential to be used in creating sphere designations, which ultimately describe interaction among sites and between regions. Black-on-Slate Types The following is an experiment with a more detailed classification that includes both system and paste ware. I offer this for discussion as a way to avoid the surfaceplus-paste problem inherent in the traditional use of ware but this is not meant to be the definitive classificatory schemejust one of several possibilities. Smith (1971) created two different black-on-slate types at Chichn Itz and Mayapn based in part on now-outdated chronological distin ctions between the Cehpech and Sotuta complexes. Thus, Sacalum Black-on-slate was placed in the Slate Muna group of Puuc Slate Ware, while Balantun Black-on-slate was placed in the Slate Dzitas group of Ch ichn Slate Ware. These are minor spatial variants of the same stylistic idea, yet their different ware designations obscure this connection. Furthermore, there are black-on-slate ceramics in coastal Belize (personal observation, Caye Coco collections), in the Petn, and possible eroded sherds at Lamanai these should be linked to the types from northern Yucatn. In the specific case of Balantun Black-onslate from Chichn Itz, (Figure 2) a systems perspective could create a designation like the following (although variations in the classificatory strategy below the system level are possible): System : Black-on-Slate System, based on surface, form, decoration, etc. The 106

PAGE 110

James Aimers nam ing conventions for systems normally dictate that the system be named after the first published type. In this case both types were published simultaneously, so I have used only the more descriptive part of the name they have in common. This is a minor issue that may have to be addressed further, perhaps in reference to some of the ware names from Brainerd (1958) (e.g., Holoctun). However, Brainerd did not assign type names so this minor issue is rather complicated. Paste Ware : Chichn Slate Paste Ware. Generally, paste wares should be based on naked eye and 10 X lens observation in terms of texture, colo r, hardness, temper (kind, quantity and size ), presence-absence of firing core, thickness, and porosity (see Rice 1976). Smith (1971:28) stated that although Chichn Slate Ware differs little from Puuc Slate Ware superficially, the paste differences are macroscopic, and I observed them in samples at INAH Mrida in 2006. Thus, this distinction is based on paste color and texture (as in paste ware), not on surface. Group: Slate Dzitas Group. Type: Balantun Black-onslate. Variety: Balantun Variety, and possibly other varieties based on petrography. The point here is that black-on-slate ceramics from Belize can be placed in the Black-on-slate System right away, and then assigned to existing paste wares, or given new paste ware names (and thus group, type and possibly varietal names) after comparison with the established types. Silver Bullets and Conclusion Can we kill the ware-beast? On one hand the answer is easy: abandon the ambiguous and problematic notion of ware that Rice described in 1976. On the other hand the solution is much more difficult. While we should surely disconnect paste and surface in ware, I dont think we can just ignore the visible paste qualities of the sherds or vessels we study so I dont think paste can beor ever is in a thorough typevariety analysis-abandoned entirely. For example, as similar as Balantun Black-onSlate and Sacalum Black-on -Slate are superficially, there are visible differences in the pastes suggesting differences in production choices that may relate to differences between Puuc and Itz people. The same is probabl y true of Pabellon Modeled-carved, Sahcaba Modeled-carved and Belize Modeled(or Molded-) carved. And we cant just wait for petrography to sort all this out. To start, when identifying types at a site in term s of previously named types, it makes sense to check that the pastes are at least broadly similar to the naked eye or with a magnifying glass. Although this seems obvious it is not an easy task to locate and handle sherds from different sites, and in some cases it is impossible. In the summer of 2006 I visited many collections all over the peninsula, but th e planning took over a year. If there is general agreement on all of this at Lamanai it is that there probably isnt a single solution to ceramic analysis generally or even type-variety specifically. One of the basic steps, though, for any project using type-variety is to tame the ware concept back into so mething manageable and once again helpful. I am convinced that ceramic system designations do much of what ware was meant to do, without most of its problems. My ideas continue to evolve, but I believe that ceramic systems can help in resolving problems with type-variety that, as the quotation that en ds this paper shows, have been running amok for almost half a century: During the past 20 years, the number of named pottery types in the Southwest has increased at such a rate that it is now virtually impossible for an archaeologist to know and be familiar with more than a small 107

PAGE 111

James Aimers percentage of them. Some of these types are totally new, but surprisingly large number represent refinements in terminology or segregations from more inclusive categories, resulting from further study and increasingly complex technological analyses. This proliferation of named types has alarmed many archaeologists. There can, however, be no legitimate doubt that if the intricate ceramic history of the Southwest and other areas of the New World is to be understood, research analysts must be free to break down their material to as fine a point as necessary in order to localize in time and space the infinitesimal variants of pottery which constitute, with other aspects of material culture, the documents of regional prehistory. It is equally clear that some method must be found to integrate the smallest units into meaningful groupings of a larger order. Groupings of this kind are necessary in the derivation of cultural interpretations from archaeological materials (Wheat, et al. 1958, italics mine) Acknowledgements I would like to thank the Belize Institute of Archaeology for supporting my research for close to 20 years, and Dr Elizabeth Graham, Dr. Linda Howie, and the other members of the Lamanai Archaeological Project for many (surprisingly) stimulating discussion of ceramics typology since 2002. Comments on the content of this paper by Dr. Arlen Chase and Dr. Patricia McAnany at the BAS were most useful. Finally, I would like to acknowledge the many archaeologists in Belize, Mexico, and Guatemala who have co-operated in this research, including Dr. Jaime Awe, Dr. Prudence Rice, Dr. Leslie Cecil, Dr. Joseph Ball, Dr. Marilyn Masson, Dr. Juan Pedro Laporte, Dr. Carlos Peraza, Dr. William Folan, Dr. Susan Milbrath, Dr. Adriana Velasquez Morlet, Dr. James Garber, and Dr. Rafael Cobos. Reference Cited Aimers, J. J. and E. Graham 2005 Type Variety on Trial: Experiments with Classification and Meaning at Lamanai, Belize. Paper presented at the Society for American Archaeology 70th Annual Meeting, Salt Lake City. Arnold, D. E. 1980 Localized Exchange: An Ethnoarchaeological Perspective In Models and Methods in Regional Exchange edited by R. E. Fry, pp. 147-150. Society for American Archaeology SAA Papers No. 1, Washington. Awe, J. J. 1985 Archaeological Investigations at Caledonia, Cayo District, Belize. Unpublished Master's Thesis, Trent University. Ball, J. W. 1976 Ceramic Sphere Affiliations of the Barton Ramie Ceramic Complexes In Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley edited by C. A. Gifford, pp. 323-330. Memoirs of the Peabody Museum of Archaeology and Ethnology Volume 18, Harvard University, Cambridge. 1977 The Archaeological Ceramics of Becan, Campeche, Mexico 43. Middle American Research Institute, Tulane University, New Orleans. Brainerd, G. W. 1958 The Archaeological Ceramics of Yucatn. Anthropological Records, University of California 19. Cecil, L. 2001 Technological Styles of Late Postclassic Pottery from the Central Petn Lakes Region, El Petn, Guatemala. Unpublished Ph. D. Dissertation, Southern Illinois University. 2004 Inductively Coupled Plasma Emission Spectroscopy and Postclassic Petn Slipped Pottery: An Examination of Pottery Wares, Social Identity and Trade. Archaeometry (46):385-404. 108

PAGE 112

James Aimers Foias, A. E. 2002 At the Crossroads: The Economic Basis of Political Power in the Petexbatun Region In Ancient Maya Political Economies edited by M. A. Masson and D. Freidel, pp. 223248. Altamira Press, Walnut Creek CA. Forsyth, D. W. 1983 Investigations at Edzna, Campeche, Mexico, Volume 2: Ceramics Papers of the New World Archaeological Foundation No. 46., Provo, Utah. Fry, R. E. 2003 Social Dimensions in Ceramic Analysis: A Case Study from Peripheral Tikal. Ancient Mesoamerica 14:85-93. Gifford, J. C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley Memoirs of the Peabody Museum of Archaeology and Ethnology Volume 18. Harvard University, Cambridge. Graham, E. 1987 Terminal Classic to Early Historic Vessel Forms from Belize In Maya Ceramics: Papers from the 1985 Maya Ceramic Conference edited by P. M. Rice and R. J. Sharer, pp. 73-98. BAR International Series 345(i), Oxford. Helmke, C., P. R. Colas and J. J. Awe 1998 Comments on the Typology, Epigraphy, and Iconography of the Actun Tunichil Muknal Vase and Belize Valley ModeledCarved Vessels In The Western Belize Regional Cave Project: Report of the 1997 Field Season edited by J. J. Awe, pp. 93140. Department of Anthropology Occasional Paper Number 1, University of New Hampshire., Durham, New Hampshire. Howie, L., T. Powis and E. Graham 2005 Can you Judge a Pot by its Color? Typological Groupings Under the Microscope at Lamanai, Belize Paper presented at the Society for American Archaeology 70th Annual Meeting, Salt Lake City. Howie, L. A. 2006 Ceramic Production and Consumption in the Maya Lowlands During the Classic to Postclassic Transition: A Technological Study of Ceramics at Lamanai, Belize. Unpublished Ph.D. Thes is, University of Sheffield. John, J. R. 2007 Postclassic Iconography at Lamanai, Belize, Central America. Unpublished Ph.D. Dissertation, University College London. Laporte, J. P. 2006 Secuencia Cermica Atlas Arqueolgico de Guatemala Sureste de Petn, Unpublished Manuscript, Dolores, Peten. LeCount, L. J., J. Yaeger, R. M. Leventhal and W. Ashmore 2002 Dating the Rise and Fall of Xunantunich, Belize: A Late and Terminal Classic Lowland Maya Regional Center. Ancient Mesoamerica 13(1):41-63. Lopiparo, J., R. Joyce and J. A. Hendon 2005 Terminal Classic Pottery Production in the Ula Valley, Honduras. In Geographies of Power: Understanding the Nature of Terminal Classic Pottery in the Maya Lo wlands edited by S. L. Varela Lopez and A. E. Foias, pp. 107-120. British Archaeological Reports International Series 1447, Oxford. Pendergast, D. M. 1985 Stability Through Change: Lamanai, Belize, from the Ninth to the Seventeenth Century. In Late Lowland Maya Civilization: Classic to Postclassic edited by J. A. Sabloff and E. W. Andrews V, pp. 223-249. School of American Research, University of New Mexico Press, Albuquerque. Powis, T. G. 2002 An Integrative Approach to the Analysis of the Late Preclassic Ceramics at Lamanai. Unpublished Ph.D. Dissertation, University of Texas, Austin. Rice, P. M. 1976 Rethinking the Ware Concept. American Antiquity 41(4):538-543. 1987 Macanche Island, El Petn, Guatemala: Excavations, Pottery, and Artifacts University of Florida Press, Gainesville. 2005 Type-Variety: What Works and What Doesn't Paper presented at the Society For 109

PAGE 113

James Aimers 110 American Archaeology 70th Annual Meeting, Salt Lake City. Rice, P. M. and D. S. Rice 2004 Late Classic to Postclassic Transformations in the Petn Lakes Region, Guatemala In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by A. Demarest, P. M. Rice and D. S. Rice, pp. 125-139. University Press of Colorado, Boulder. Sabloff, J. A. and R. E. Smith 1970 Ceramic Wares in the Maya Area: A Clarification of an Aspect of the TypeVariety System and Presentation of a Formal Model for Comparative Use. Estudios de Cultura Maya 8:97-115. Smith, R. E. 1955 Ceramic Sequence at Uaxactn, Guatemala Publication 20. Middle American Research Institute, Tulane University, New Orleans. 1958 The Place of Fine Orange Pottery in Mesoamerican Archaeology. American Antiquity 24(2):151-160. 1971 The Pottery of Mayapn Papers of the Peabody Museum of Archaeology and Ethnology 6(4). Harvard University, Cambridge. Smith, R. E. and J. C. Gifford 1966 Maya Ceramic Varieties, Types, and Wares at Uaxactn: Supplement to "Ceramic Sequence at Uaxactn, Guatemala" Publication 28, Middle American Research Institute, New Orleans. Wheat, J. B., J. C. Gifford and W. W. Wasley 1958 Ceramic Variety, Type Cluster, and Ceramic System in Southwestern Pottery Analysis. American Antiquity 24(1):34-47. Willey, G. R., T. P. Culb ert and R. E. W. Adams 1967 Maya Lowland Ceramics: A Report from the 1965 Guatemala City Conference. American Antiquity 32(289-315). Willey, G. R., R. M. Leventhal, A. Demarest and W. L. Fash 1994 Ceramics and Artifacts from Excavations in the Copan Residential Zone Papers of the Peabody Museum of Archaeology and Ethnology Vol. 80. Harvard University, Cambridge.

PAGE 114

10 FRESH VOLCANIC GLASS SHARDS IN THE POTTERY SHERDS OF THE MAYA LOWLANDS Anabel Ford and Frank Spera Pristine, fresh volcanic ash has been known as a dominant tempering agent in the Late Classic Maya (AD 600-900) ceramics. Archaeologist Anna O. Shepard first identified vo lcanic glass in Maya pottery sherds and struggled to solve the mystery of its origin as the lowland Maya lived on carbonate bedrock with th e closest volcanic sources 350 km away. How did relatively large volumes (~ 106 m3) of volcanic ash become available for manufacturing of ceramic products for the entire Late Classic Period? This question has never been answered. While it has been accepted that the ash was of non-local or igin, this anomaly has never been expl ained. Identification of the source of the volcanic ash used by the ancient Maya has implications for the economy of ceramic production as well as the ecology of the Maya forest. Resolving the origin of the unexplained appearance of volcani c ash in the Late Classic, coincident with the apex of Maya civilization, will contribute directly to our understanding of the development of the Maya society and address the impacts of volcanic activity at a distance. Introduction: It was 1930 and the researcher Anna O. Shepard, then at the Carnegie Institution of Washington, applied innovative geological techniques to archaeological problems, and so doing discovered that there was volcanic ash in pottery from the nonvolcanic limestone Maya lowlands (Shepard 1937). This incongruity geologically recent volcanic glass used as temper with clay derived from geologically ancient marine limestone presented an anomaly that followed her throughout her career. Shepards challenge to identify the source(s) for the volcanic ash in the limestone Maya lowlands and recover evidence of cultural and environmental influences preserved in the potters craft (S hepard 1936 cited in Cordell 1991) was the basis of our research endeavour. We have begun an inquiry to evaluate the origin of the volcanic ash used in ancient Maya pottery production with the use of 21st century geochemical tools and have completed the first phase of this research, combining microprobe with classical petrographic techniques to reveal new dimensions to the issue of origins of the ash temper. The resolution of the problem of volcanic ash shards in Maya pottery sherds bears significantly on our understanding of the development of the Maya civilization. Historical View of Volcanic Ash and the Maya Shepards identification of diagnostic pristine volcanic glass shards is unexpected for clay deposits derived from chemical weathering of the Cretaceous/Eocene carbonates of the Maya area. Consequently vo lcanic ash has been interpreted as exotic tempering agent (Shepard 1937, 1951, 1956, 1962; Sunahara 2003). According to Shepard, and confirmed by all subsequent research, ceramics tempered with volcanic ash occurred predominantly in association with the Late Classic Period (c. AD 600-900). The consistency of the pastes pointed to a reliable and steady procurement source over the long period of use (Shepard 1939, 1942; in Smith 1955). Decades have passed since Shepard discovered this incongruity. We seek to rectify this. Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 111-118. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 115

Anabel Ford and Frank Spera Figure 1 Photomicrograph of Sherd #1902 site 272-136 L 10x (2mm field of view) and R 40x (field of view is 0.2 mm). Glass is abundant, visible as the sharp edged white shards embedded in the clay matrix of the sherd. Glass is chemically and mechanically unstable ma terial; the abundance of pristine glass shards is consistent with aeolian deposition (air fall); the glass shapes have no evidence of weathering or fluvial transport. Figure 2 Volcanoes of the Central American Guatemala High lands and Central Maya Lowlands of El Pilar and Tikal Late Classic Maya ceramics have been well-described in the literature beginning with the Carnegie studies that involved Anna O. Shepard, through the intensive Barton Ramie project (Willey et al. 1965; Gifford et al. 1976), and up into the active projects in the region today (e.g., Jones 1984; LeCount 1996; Lucero 2001, Sunahara 2003). Ceramics from the earliest periods of the Preclassic, from c. 1000-800 BC through AD 250 are dominated by the local calcite (limestone) tempering as are those of the Early Classic (c. AD 250-600). The Maya mastered the unstable additive and produced finely made ceramics as well as the necessary cooking and storage vessels. It is in the Late Classic that there is a distinction of ceramic wares with the 112

PAGE 116

Investigations of Pottery Sherds in the Maya Lowlands abrupt presence of serving bowls and plates as well a s water jars tempered with volcanic ash, leaving the cooking and storage vessels consistently tempered with the local calcite. Ceramic pastes tempered with volcanic ash are remarkably uniform in appearance and texture permitting a rough identification based on inspection and simple methods such as reactivity with hydrochloric acid (Ford and Glicken 1987; Ford and Rose 1995; LeCount 1996:360; Gifford et al. 1976:255). Gifford marked ash-tempered ceramics as British Honduras Volcanic Ash Ware (Gifford et al. 1976:255-267). These pastes are found widespread at every residential unit occupied in the Late Classic from the environs of the major center of Tikal to the major center of El Pilar, used in the production of everyday wares such as bowls, jars, and plates, as well as in the production of specialized elite polychromes. These distinctive wares of Maya ceramics, when examined petrographically (Figure 1), display fresh, unaltered volcanic glass, and retain all the characteristics of pristine volcanic air fall tephra (see Ford and Glicken 1987; Ford and Rose 1995). Without a doubt, the ultimate source of the volcanic ash in the Late Classic Maya ceramics must have been the active volcanoes themselves. Yet how did the volcanic ash get to the limestone lowlands, more that 350 kilometres away? (Figure 1) Volcanic ash may be chemically sourced to the Central American Highlands (CAH) or the Mexican Volcanic Belt (MVB), just as obsidian is sourced, but it is unlikely that the distant deposits more than 350 km away were the source of ash in the Maya ceramics (Figure 2). Arnolds (1985) research on resource procurement for ceramic production provides a benchmark. Clays are heavy and make up the principal material of pottery; 84 % of clay sources are collected within 7 km of production. Similarly temper is collected with 97% of the sources coming from under 9 km and none more than 25 km away. Only slips, paints, and dcor are a comparatively minor component of the potters craft, and sources can come from much greater distances; 56% of the resources were found within 30 km distance of where the ceramics were made (Arnold 1985:50-52). What were the sources of volcanic ash that provided a ready and continuous supply for the Late Classic Maya? Importing volcanic ash from long distances would be prohibitive (Santley 2004) and the volcanological analyses support the hypothesis of an ash fall source (Ford and Rose 1995). There may have been a source or sources unidentified today but accessible and used by the Late Classic Maya. An ash fall source is logical and ash has fallen in the lowland Maya region from evidence in the Petn lake cores (Leyden et al. 1993) and records of the 1982 El Chichn (Espindola et al. 2000). Ash fall at once resolves procurement issues and provides a ready explanation for the morphology of the glass shards in the Maya ceramics. Geological and Archaeological Issues Geological Advantages of volcanoes may be overlooked when considering immediate and direct impacts of volcanism. Benefits of volcanic eruptions include long-term positive effects related to soil fertility, well known from geothermal resources in volcanic regions all over the world (Williams and McBirney 1979; Ping 2000; Arnorsson 2000); and the availability of volcanic products for commerce and industry (Dehn and McNutt 2000). Both prehistoric and modern societies in the vicinities of volcanoes have used these products in various ways: volcanic ash as temper in ceramics (Arnold, 1985:59; Shepard, 1956:4,378381), obsidian (volcanic glass) for cutting implements or as 113

PAGE 117

Anabel Ford and Frank Spera 114 semi-precious stone in jewellery (Gaxiola and Clark 1989), and ot her extrusive rocks have been used in construction, as stone implements or abrasives (Dehn and McNutt 2000). When considering the impacts of volcanic eruptions from a distance, the byproducts of eruptions, su ch as volcanic ash, can present an opportunity and this appears to be the case for the Late Classic Maya. The Maya lowlands are underlain by lithified marine sediments geologically older than 60 million years (Instituto Geogrfico Nacional 1972). The sites where the volcanic ash tempered ceramics are recovered, essentially wi thin the central core area of the Maya, lie more than 350 air kilometres north of the active volcanic vents in the of Central America Highlands and 550 air kilometres from the closest volcanoes of the Mexican Volcanic Belt. There are multiple candidates for volcanic sources that could have used by the ancient Late Classic Maya (Ford and Glicken 1987; Ford and Rose 1995; Carr et al. 2003; Espindola et al. 2000; Mercado and Rose 1992). Moreover, while the amount of volcanic ash used in th e ceramics of the Late Classic is considerable if one regarded it as a trade item (Ford and Glicken 1987), such as obsidian, the total amount is easily accounted for in a deposit of a single eruption of about 0.007 km3 (Ford and Rose 1995:152). For about 200 years during the Late Classic, volcanic ash was dependably available in sufficient quantity for use in ceramic vessels made for domestic purposes. You cannot make chocolate cake without chocolate; therefore you cannot make volcanic ash tempered pottery without volcanic ash. With this the central issue is clearly articulated: (1) there was a single eruption of significant volume to accumulate in the lowlands in sufficient amount for use throughout the Late Classic period or (2) there was repeated influx of volcanic ash into the lowland region that allowed for the continual use of ash as temper. If there is a single source, the glass elemental qualities should have a single compositional signature that links to one specific volcanic source. Multiple sources should demonstrate multiple compositional signatures and link to a sequence of eruptions over the Late Classic period at candidate volcanic source sites. Analysis of Volcanic Ash Shards in Maya Pottery Sherds Our findings demonstrate the challenge of research to source the volcanic ash in the Maya lowlands. We began with the petrographic descriptions, gaining an appreciation for the detailed work that Anna Shepard undertook. On average, from 2050% of the pastes were composed of the volcanic ash and the glass shards ranged in size from 10 to 300 microns, with a mode near to 50 microns. Associated igneous phenocrysts, particularly biotite and plagioclase feldspar (Ford and Spera 2005; Ford and Rose 1995) reflect air fall qualities, with no evidence of reworking by water or other mechanical means (see Figure 2) supporting Shepards early assessments. The pastes speak to the potters craft, the recipe used for the ceramic pastes confirms the cultural influence in the production of the Late Classic volcanic ash wares. Our electron microprobe of glass shards from four unique pottery sherds of different residential sites in the El Pilar area provided encouraging resu lts. Four separate compositional groups with virtually no overlap were found based on the major elemental analyses. In particular the mass fraction ratios SiO2/CaO and Na2O/K2O were especially useful as discriminates (Figure 3). Although intra-sherd volcanic glass shard compositions are relatively homogeneous, inter-sherd glass shard

PAGE 118

Investigations of Pottery Sherds in the Maya Lowlands Figure 3 Ratio Distribution of Major Elements of Volcanic Ash Shards in Four Example Late Classic Maya Pottery Sherds compositions are distinctly different. Our exploratory results also revealed that the glass shards used as ceramic temper by the ancient Maya have uniformly high silica content (75-77 wt %), typical of rhyolitic volcanic ash. This seems to rule out the El Chichn volcano (Espindola et al. 2000; Weintraub 1982) that covered the areas around El Pilar with an andesitic composition ash (~ 60 wt % silica). Further, even though the volume of ash needed for everyday ceramic ware in the interval between 600-900 AD seemingly requires a consistent supply that would imply more than a single ash fall event (Ford and Rose 1995), our initial compositional microprobe analyses have not succeeding in fingerprinting the source or sources. The material used by the ancient Maya was unlike the typical composition of CAH volcanic sources (58-72 wt % silica). Silica, the primary component in rhyolitic glass, is resistant to thermal effects, yet other elements are more mobile. The prehistoric ceramic firing is at low temperatures with respect to melting (Ford and Lucero 2000), but there remains the potential to modify glass shard compositions by reactions between the rhyolitic glass and its surrounding matrix. Calcite (CaCO3) is an omni-present component in the Maya lowland soil derived from the chemical and physical weathering of lowland limestone and carbonate-bearing siltstones. Chemical weathering of carbonates in Maya lowland limestone leaves insoluble clays as a residue. Calcite decarbonates at 850 C and is reactive with sili cate glass at ceramic firing temperatures-800 C (Rice 1987:87; Ford and Lucero 2000). It is possible, indeed likely, therefore, that there was chemical exchange of CaO between glass shards and surrounding clay-carbonate matrix. This effect would reduce the silica/lime ratio in glass shards because CaO reacts with and dissolves into pristine glass. To test the nature of the changes in shard composition due to reaction with matrix and alkali element volatility, we initiated firing experiments with analyzed pumice, collected from the Moho Cay area of Belize, set in the Maya clay matrix starting material. Using clay collected in the field area, we prepared samples of clay and volcanic ash for firing. Controlled experimental tests were conducted at time 115

PAGE 119

Anabel Ford and Frank Spera and tem perature intervals that parallel the firing conditions reporte d in the literature by Shepard (1963:83-91) and Rice (1987:9093). We discovered that there were changes in the specific elemental levels and that the changes would affect the SiO2/CaO and Na2O/K2O ratios in pa rticular. We observed a decrease in the ratio of both silica/lime and soda/potash with increasing firing temperature and the duration of the firing event. That is, the composition of the fired laboratory shards did no t match the starting glass composition for the long durationfiring experiments. These results have expanded the scope of our investigation, to leave the central problem of source unresolved. We clearly need to better understand the effects of firing on the composition of the glass shards. We are now at a point where we have a birds-eye-view of the problem. We know the major oxide compositional span for ~ 1000 shards in the Maya ceramic sherds established vi a electron microprobe of analysis. We have an initial study of the candidate volcanoes from the CAH and MVB regionsboth of which represent possible source areas. We have learned that there are elemental changes that increasingly shift the composition of glass shards with maximum firing temperature and firing duration. The problem remains: Where did the ash originate? To determine the volcanic ash source or sources, we need to know: 1) The range of volcanic ash compositions from Late Classic Maya ceramics, 2) The composition changes of ash with firing time and temperature, 3) The location of candidate eruptions, and 4) The distribution of ash from candidate volcanoes. Our aim then is to develop accurate descriptions of the volcanic ash shards in the Late Classic Maya ceramic sherds and to determine the potential matches that exist with known volcanic ash falls for the region. We are examining information on the active volcanoes of Central America and Mexico. We are also compiling data on the environmental impacts of the ash at a distance. Summary Pristine volcanic ash has long been known as a specific tempering agent in the Late Classic Maya ceramics. While it has been accepted that the ash was of non-local origin, this anomaly has never been explained. Identification of the source of the volcanic ash used by the ancient Maya has implications for the economy of ceramic production as well as the ecology of the Maya forest. For seventy-five years, archaeologists of the ancient Maya have recognized the presence of volcanic ash in the Late Classic Maya ceramics. Left dormant, there was no acknowledgement of the implications of the ash in the ceramics. Sourcing the exotic tempering materials will impact our essential grasp of the Late Classic Maya economy. Were the Maya opportunist when a single eruption deposited ash in their midst to exploit over several centuries? Or did a sequence of ash falls concentrate in the Late Classic promoting new ceramic production strategies and perhaps improving local production capacities for the soil? Our research brings together data from the distinct disciplines volcanology and archaeology into focus with an anthropological question that can only be answered with volcanological data. Our results will have far reaching implications for understanding the rise and fall of the ancient Maya civilization 116

PAGE 120

Investigations of Pottery Sherds in the Maya Lowlands Reference Cited Arnold, D.E. (editor) 1985 Ceramic Theory and Cultural Process Ashmore W. (ed) ed. Cambridge University Press, New York. Carr, M.J., Feigenson M.D., Patino L.C. and Walker J.A. 2003 Volcanism and Geochemistry in Central America: Progress and Problems. In Inside the Subduction Factory, edited by J. Eiler and G. Abers, Rutgers University. Geophysical Monograph Volume 138: 153179. Cordell, Linda S. 1991 Anna O. Shepard and Southwestern archaeology: ignoring a cautious heretic In Ceramic Legacy of Anna O. Shepard edited by R. L. Bishop, pp. 132-153. Colorado University Press, Niwot. Dehn, J. and S. R. McNutt 2000 Volcanic Minerals in Commerce and Industry In Encyclopaedia of Volcanoes edited by H. Sigurdsson, pp. 1272-1299. Academic Press San Diego, San Diego, CA. Espndola, J.M., J.L. Macas and R.I. Tilling 2000 Volcanic History of El Chichn Volcano (Chiapas, Mexico) during the Holocene, and its impact on human activity. Bull Volcanol 62(2000):90-104. Ford, Anabel and Harry Glicken 1987 The Significance of Volcanic Ash Tempering in the Ceramics of the Central Maya Lowlands. Paper presented at the 1985 Maya Ceramics Conference, Washington D.C. Ford, Anabel and Lisa Lucero 2000 Malevolent Demons of Ceramic Production: Where have all the Failures gone? Estudios de Cultural Maya 21:57-74. Ford, Anabel and William I. Rose 1995 Volcanic Ash in Ancient Maya Ceramics of the Limestone Lowlands Implications For Prehistoric Volcanic Activity in the Guatemala Highlands. Journal of Volcanology and Geothermal Research 66(1-4):149-163. Gaxiola, Margarita, John E. Clark and Instituto Nacional de Antropolog a e Historia (Mxico). Centro Regional de Hidalgo 1989 La Obsidiana en Mesoamrica 1. ed. Instituto Nacional de Antropologa e Historia, Mxico, D.F. Gifford, James C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley Memoirs of the Peabody Museum of Archaeology and Ethnology, Harvard University; v. 18. Peabody Museum of Archaeology and Ethnology Harvard University, Cambridge, Mass. Instituto Geogrfico Nacional 1972 Atlas Nacional de Guatemala Instituto Geogrfico Nacional, Guatemala City. Jones, L. 1984 A Petrologic Analysis of Lowland Maya Ceramics. Ph.D. dissertation, University of London. LeCount, Lisa J. 1996 Pottery and Power: Feasting, Gifting, and Displaying Wealth Among the Late and Terminal Classic Lowland Maya. Ph.D. dissertation, University of California Los Angeles. Leyden, B. W., M. Brenner, D. A. Hodell and J. H. Curtis 1 993 Late Pleistocene Climate in the Central American Lowlands. In Climate in Continental Isotopic Records edited by P. K. Swart, pp. xiii, 374. 78 ed. Geophysical Monograph. American Geophysical Union, Washington, D.C. Lucero, L. J. 2001 Social Integration in the Ancient Maya Hinterlands: Ceramic Variability in the Belize River Area. Tempe, AZ, Arizona State University. Mercado, Reinaldo and William I. Rose 1992 Reconocimiento geolgico y evaluacin preliminar de peligrosidad del volcn Tacaa, Guatemala/Mexico. Geofisica Internacional 31(3):205-237. 117

PAGE 121

Anabel Ford and Frank Spera 118 Ping, C. L. 2000 Volcanic Soils In Encyclopaedia of Volcanoes, edited by H. Sigurdsson, pp. 1259-1270. Academic Press San Diego, San Diego, CA. Rice, Prudence M. 1987 Pottery Analysis: A Sourcebook. University of Chicago Press, Chicago, IL. Santley, Robert S. 2004 Review Article: What Was Teotihuacan Doing in the Maya Region? Journal of Anthropological Research 60:379-396. Shepard, Anna O. 1937 Ceramic Technology In Carnegie Institution of Washington Year Book pp. 144-145. Vol. 36. Carnegie Institution of Washington, Washington, D.C. 1939 Technological Notes on the Pottery of San Jose In Excavations at San Jose, British Honduras edited by J. E. S. Thompson, pp. Appendix. Vol. 506. Carnegie Institution of Washington, Washington, D.C. 1942 Classification of Painted Wares In Late Ceramic Horizons at Benque Viejo, British Honduras edited by J. E. S. Thompson. Vol. 528. Carnegie Institution of Washington, Washington, D.C. 1951 Ceramic Technology In Carnegie Institution of Washington Year Book pp. 241-244. Vol. 50. Carnegie Institution of Washington, Washington, D.C. 1956 Ceramics for the Archaeologist. Vol. 609. Carnegie Institution of Washington, Washington, D.C. 1962 Ceramic Development of the Lowland and Highland Maya Paper presented at the 35th International Congress of Americanists. 1963 Ceramics for the Archaeologist Carnegie Institution of Washington, Washington, D.C. Smith, Robert Eliot 1955 Ceramic Sequence at Uaxactun, Guatemala, pp. 2 v. vo l. I. Middle American Research Institute Tulane University, New Orleans, LA. Sunahara, Kay Sachiko 2003 Ancient Maya Ceramic Economy in the Belize River Valley Region: Petrographic Analyses, pp. 176, Hamilton, ON (Canada). Weintraub, Boris 1982 Mexican Volcano Spews Death and Devastation (El Chichn in Chiapas) In National Geographic pp. 654-671, 676-684. Vol. 162. Willey, Gordon Randolph, W.R. Bullard, J.B. Glass and J.C. Gifford 1965 Prehistoric Maya Settlements in the Belize Valley In Papers of the Peabody Museum of Archaeology and Ethnology edited by G. R. Willey, pp. xxix, 589 p.: ill., maps (1 fold. in pocket) ; 27 cm. and portfolio (36 tables) 28 cm. vol. 54. Peabody Museum, Cambridge. Williams, H. and A.R. McBirney 1979 Volcanology. Freeman, Cooper, and Co., San Francisco.

PAGE 122

11 THE CLASSIC TO POSTCLASSIC TRANSITION IN THE SIBUN VALLEY, BELIZE: DEFINING THE TERMINAL CLASSIC CERAMIC ASSEMBLAGE Eleanor Harrison-Buck and Patricia A. McAnany Recent archaeological investigations in the Sibun Valley of Belize have identifie d Maya populations that survived the collapse of Classic Maya civilization. Excavations of several sites lo cated in different parts of the Sibun Valley have revealed stratified deposits that help to furthe r define the Terminal Classic ceramic assemblage. As a frontier zone, the Sibun Valley appears to have been impacted by two competing core areas the Petn and Belize Valley to the west where large centers like Naranjo were highly influential in the Classic period and northern Yucatan where Terminal Classic centers such as Chichn Itz fueled a trade network along the Caribbean coast that extended as far south as Belize. Potential spheres of interaction are discussed based on comparative analyses with ceramic complexes from other sites in Belize, namely the Belize Vall ey. The results of the study shed light on shifting spheres of interaction during the Terminal Classic period with varied social, political and economic processes occurring at different sites throughout the valley at the time of the collapse Introduction The Terminal Classic periodthe transition from the Classic to Postclassic marks an important period of political, economic, and social transformation throughout Mesoamerica (Diehl and Berlo 1989). During this transitional period from around AD 750 to 900, Maya centers in the Petn core area were declining, and northern Yucatec polities, such as Chichn Itz, were on the rise. Both the architecture and ceramic data suggest that the Sibun River valley was a frontier z one, interacting with these two competing core areas of the Maya region during the Late-Terminal Classic transition. Positioned in the central part of Belize, the Sibun River valley (Figure 1) is linked to the Petn areaand its many powerful Classic Maya centers, such as Tikal and Naranjovia the Belize Valley, which is the proximate drainage immediately to the north. To the east, the Sibun River flows into the Inner Channel of the Caribbean Sea, providing access via maritime transport to northern Yucatec capitals such as Chichn Itz. These geographic links enabled the Sibun inhabitants to interact with both of these core areas. Figure 1. Map of the Maya Lowlands (map prepared by B. Thomas). At certain sites in the Sibun Valley, we note the introduction of new architectural styles and associated ceramic types that depart from the Petn-affiliated (Tepeu/Spanish Lookout) tradition and show Research Reports in Belizean Archaeology Vol. 4 2007, pp. 119-133. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 123

Eleanor Harrison-Buck and Patricia A. McAnany 120 influence stemming from northern Yucatan (Harrison-B uck 2007). We discussed some of this archaeological evidence in last years Belize Archaeology Symposium (see Harrison-Buck and McAnany 2006). In particular, we examined finds of Yucatecstyle circular shrinessmaller versions of the Caracol Building at Chichn Itz (Ruppert 1935). We also touched upon some of the northern-style ceramic material, namely the Kik Group types, which we found in the Sibun Valley at sites with Yucatec-style architecture. Here, we expand this discussion of Terminal Classic occupation in the Sibun Valley, focusing on the diagnostic ceramic material that we find throughout the valley during this time. Figure 2. Map of the Sibun Valley settlements (map prepared by B. Thomas). The study presented is based on a typological analysis of ceramic material from a series of excavations conducted at Pechtun Ha, Oshon and Obispothree Maya settlements containing Yucatec-style circular architecture in the Sibun Valley (Figure 2). Additionally, we compare ceramics from two other sites in the middle and upper reaches of the valleyPakal Na and Hershey. Archaeological excavations at these five sites in the Sibun Valley yielded a series of well-stratified deposits that allow ceramic material to be examined contextually. In this way, Terminal Classic diagnostic ceramic material is identified in the assemblage using the Type-Variety format. The primary ceramic types found in the Sibun Valley during the Terminal Classic period includ e Roaring Creek Red bowls and dishes, Sibun Red Neck jars, Indian Creek Polychrome bowls, Dolphin Head bowls and dishes, and Cayo Unslipped utility jars. As part of the Sibun household assemblage, the Sibun Red Neck and Cayo Unslipped jar forms are the most common type of storage and cooking vessels, while Roaring Creek Red, Indian Creek Polychrome, and Dolphin Head Red represent the most co mmon serving dishes and bowls found in th e valley during the Terminal Classic period. Other less common diagnostics of the Terminal Classic assemblage in the Sibun Valley are found in the Garbutt Creek, Mountain Pine, Vaca Falls, Mount Maloney and Belize Ceramic Groups (for a full discussion see HarrisonBuck 2007). In this paper, the Terminal Classic diagnostic ceramic types, as well as some of the special trade wares found in the Sibun Valley, are reviewed and the results of a distributional study are presented. Distribution patterns reflect discrete social networks and spheres of interaction that are notably different when assemblages are compared between the upper and lower reaches of the valley. Sites in the upper reaches of the Sibun Valley, like Hershey, display none of the northern-style (Kik Group) ceramics and Yucatec architecture, but rather, retain an affiliation with the Belize Valley and Petn region to the west during the Terminal Classic period. Meanwhile, the Sibun Maya living in the lower reaches of the valley seem to shift

PAGE 124

Classic to Post Classic Investiga tions in the Sibun River Valley 121 their political and economic focus away from this western region by this time and establish a new interaction sphere along the coast with northern polities like Chichn Itz. Defining Terminal Classic Diagnostic Ceramics Mount Maloney Black and Belize Red Ceramicists, including James Gifford (1976), Lisa LeCount (1996 and 2005), Jim Aimers (2002) and others working in the upper Belize Valley have helped to define the Terminal Classic diagnostics for this central-western part of Belize. Several key ceramic types define the ninth and tenth century Maya occupation in this area, referred to as the late facet of the Spanish Lookout Ceramic Complex or Tepeu 3 period. Mount Maloney Black is a key Late to Terminal Classic ceramic diagnostic, initially defined by Gifford (1976) for Barton Ramie. LeCount (1996) has been able to identify and de fine certain varieties of Mount Maloney that serve as markers of the Terminal Classic period. While bowls are the most common form, jars are also present in the Belize Valley assemblage. LeCount (1996:160) notes that Terminal Classic Mount Maloney jars typically have overhanging, angled profiles, whereas the Late Classic II jars generally have smooth contours. Overall, Mount Maloney Black (Figure 3) is a rare type in the Sibun Valley assemblage, but when found they are invariably jar forms with overhanging, angled profiles and appear with other ceramics that date to the Terminal Classic, such as the Belize Molded-Carved type (see discussion of this type further below). Belize Red is another type that has been identified in Terminal Classic contexts in the Sibun Valley (Figure 4). This redslipped ware contains a high percentage of volcanic ash temper, which is largely comprised of laths of SiO2, or glass. The ash paste is distinctive and readily identifiable in the Sibun assemblage. The paste of eroded sherds is chalky and rubs off on ones hands. Li ke Mount Maloney Black, Belize Red is relatively rare in the Sibun Valley. It is more frequent in the Late Classic II and shows a sharp decline in frequency during the Terminal Classic period at Hershey and Pakal Na (Figure 5). These two sites have a strong Late Classic II occupation that could be separated from later Terminal Classic contexts. In Figure 5, frequencies (quantified based on the total count of all rim and body sherds) show a marked decrease over time when Late Classic II and Terminal Classic assemblages are compared side-by-side. LeCount (1996:160-161, Table 5.9) documents a similar pattern in the Belize Valley at Xunantunich and San Lorenzo. The Belize Red type dominates the assemblage in both the Late Classic II and Terminal Classic periods, but shows a slight decline during the latter period. Figure 3. a-d Mount Maloney Black: Mount Maloney Variety: a) Obispo RIP 1152, b) Hershey RIP 1469, c) Hershey RIP 1470, d) Hershey RIP 1471 (illustrations by E. Harrison-Buck). Both Belize Red and Mount Maloney Black were most likely imported from the Belize Valley, given their abundance in this region (Aimers 2002; Gifford 1976; LeCount 1996). One exception may be a partially reconstructible

PAGE 125

Eleanor Harrison-Buck and Patricia A. McAnany Figure 4 a-c. Belize Red: a) Oshon, RIP 1192, b) Oshon, RIP 1189, c) Hershey, RIP 1190, d) Obispo, RIP 1194, e) Pakal Na, Vessel 214 (illustrations by C. Cesario and E. Harrison-Buck). vessel (Figure 4e) from a Late Classic II burial deposit at Pakal Na. A thin-section of this vessel examined under a microscope revealed no ash temper, only calcite inclusions and it appear s to be an imitation Belize Red platter. Figure 6 displays the total percentage of Mount Maloney Blac k jars and the Belize Red ash wares in the Terminal Classic assemblages from the Sibun Valley settlements. Ceramic percentage is calculated based on the total number of rim sherds. Quantities are representative of Terminal Classic contexts and do not include the Late Classi c II ceramic counts, discussed above. The five sites analyzed in this study are presented from west (Hershey) to east (Oshon) in Figure 6. Both Mount Maloney Black and the Belize Red are surprisingly rare in the Sibun Valley given its close proximity to the Belize Valley where these ceramics were undoubtedly manufactured. Overall, Hershey and Pakal Na contain a heavier percentage of both types compared with Pechtun Ha, Obispo, and Oshon. In sum, ceramics from Hershey and Pakal Na (sites closest to the Belize Valley) reflect the strongest Tepeu/Spanish Lookout affiliation when compared to other sites in the valley. Roaring Creek Red and Sibun Red Neck Two of the most common Terminal Classic ceramic types found throughout the length of the Sibun Valley is Roaring Creek Red and Sibun Red Neck jars (Figures 7 and 8). Sibun Red Neck is a new type defined in this study. Figure 9 shows the total percentage of these tw o types represented in the Terminal Classic ceramic assemblage from each of the five sites in the Sibun Valley. Notably, the lowest percentage of Roaring Creek Red is found at Hershey, which seems to align with the ceramic distribution patterns of the Belize Valley. Roaring Creek Red is part of the Spanish Lookout Ceramic Complex defined by Gifford (1976) for Barton Ramie. However, the frequency of this Terminal Classic type is actually relatively low at Barton Ramie and other sites in the Belize Valley, especially when compared to other outflaring, red slipped dishes, such as the 122

PAGE 126

Classic to Post Classic Investiga tions in the Sibun River Valley 0 20 40 60 80 100 120 140 160 Late Classic IITerminal ClassicTotal Counts (rim and body sherds) in Hershey & Pakal Na Ceramic Assemblage s Hershey Pakal Na Figure 5. Frequencies of Belize Red at Hershey and Pakal Na in the Late Classic II and Terminal Classic periods (based on total counts of rim and body sherds). 4.85 7.96460177 1.39 0.91 3.94 9.71 4.424778761 0 1.82 0.3 0 2 4 6 8 10 12 14 16 HersheyPakal NaPechtun HaObispoOshon Sibun Valley SitesTotal Percent (%) of Terminal Classic Ceramic Assemblag e Belize Red Mount Maloney Black Figure 6. Total Percentage of Belize Red and Mount Maloney Black in the Terminal Classic ceramic assemblages from sites in the Sibun Valley. 123

PAGE 127

Eleanor Harrison-Buck and Patricia A. McAnany Belize Red type. Belize Red far exceeds th e presence of Roaring Creek Red in the Terminal Classic ceramic assemblage at Barton Ramie, Baking Pot, Tipu, and Xunantunich (Aimers 2002; LeCount 1996:Table 5.9). Belize Red and Roaring Creek Red share an equivalent percentage of the Terminal Classic assemblage (4.85%) at Hershey (Compare Figures 6 and 8). Elsewhere in the Si bun Valley, Roaring Creek Red is by far the dominant redslipped serving dish of the Terminal Classic ceramic assemblage. Figure 7. a-g. Examples of Roaring Creek Red: Roaring Creek Variety: a) Hershey, RIP 1005/1031; b) Oshon, RIP 1406; c) Pakal Na, RIP 1468; d) Pechtun Ha, RIP 1407; e) Obispo, RIP 1409, f) Pechtun Ha, RIP 1408; g) Oshon, RIP 1024 (illustrations by E. Harrison-Buck). Likewise, the Sibun Red Neck jar (Figure 8) represents a common ceramic type found throughout the Sibun Valley, but is more rare in the Belize Valley. Gifford (1976:235-237 and Figure 144k and j) designated this type as Vaca Falls Red, illustrating two examples in his Barton Ramie report. However, he admitted the jar form was not common and in his description he seems unsure of where to place the few examples that he had. Aimers (2002) and LeCount (1996) do not mention this jar form in their analyses of the ceramic assemblages from Baking Pot, Tipu and Xunantunich. In marked contrast, these jars are exceedingly abundant in the Sibun Valley. They appear quite distinct in form, paste and surface treatment from the Vaca Falls Red bowls. Therefore, in this study they have been given a new type name designated as Sibun Red Neck jars. Figure 8 a-d. Examples of Sibun Red Neck jars: a) Pechtun Ha, RIP 1004; b) Oshon, RIP 1377, c) Oshon, RIP 1015; d) Oshon, RIP 1006 (illustrations by E. Harrison-Buck). Red slip extending down to the shoulder is the character istic feature of these Terminal Classic jar forms. The remaining body of the Sibun Red Neck jar is unslipped. These vessels should probably be categorized as Uaxactun Unslipped wares, rather than Pine Ri dge Carbonate wares (where Gifford placed them as part of the Vaca Falls Group). Like Roaring Creek Red, this distinctive type is a solid indication of a Terminal Classi c context in the Sibun Valley and is never found in the Late Classic period. Other types of red neck jars do occur before the Terminal Classic period, but are distinct in form, paste and surface decoration. Several different types are found in earlier Classic period contexts in the Sibun Valley, including a jar form that has less of an outcurvi ng neck, more direct rim, with a light orange wash and a red slip that covers only the tip of the exterior rim. This type may also continue into the 124

PAGE 128

Classic to Post Classic Investiga tions in the Sibun River Valley 125 21.36 15.92920354 18.06 10.91 20 4.85 19.46902655 22.22 14.55 13.33 0 5 10 15 20 25 30 35 40 45 50 HersheyPakal NaPechtun HaObispoOshon Sibun Valley SitesTotal Percent (%) of Terminal Classic Assemblage Sibun Red Neck Roaring Creek Red Figure 9. Total Percentage of Sibun Red Neck and Roar ing Creek Red in the Terminal Classic ceramic assemblages from sites in the Sibun Valley Terminal Classic period. Another type is thinner-walled and has a darker paste with gray surfaces and a reddish-brown slip around the neck. Comparable types may be Duende Daub Striated and Petroglyph redrimmed, found in the Sibun Valley and Caves Branch District caves (Audet 2002: Fig. 7.4 a-d; Peterson 2006; Reents 1981), as well as the Stann Creek District (Graham 1994: Figs.5.17-5.18). These types are distinct from the Terminal Classic Sibun Red Neck type that is defined in this study. Occasionally, the unslipped body of the Sibun Red Neck jar has a thin wash of black and is lightly striated (see Figure 8d), similar to the Red Neck Mother Striated type defined by Diane Chase (1982) for the site of Nohmul. Pendergast (1974) identified both the plain and the striated varieties in his study of Actun Polbilche, a cave in the middle reaches of the Sibun Valley. Both varieties of these Terminal Classic red-necked jars also are found at Lamanai and Altun Ha (Graham 1987:7879; Pendergast 1990). Thompson (1939: Fig. 76) also illustrates identical jar forms, both plain and striated, in his ceramic report on the site of San Jose. Pendergast (1974:83-84) concludes that on typological grounds, the ceramics from the Sibun are linked most closely with San Jose, located only about 45 km to the west-northwest of the Sibun Valley. He acknowledges that what is most surprising is the Terminal Classic ceramics from the Sibun bare little resemblance to the ceramics from Xunantunich, Barton Ramie and Baking Pot, even though it lies on the edge of the Belize River Valley area. Indian Creek Polychrome and Fat Polychrome What is perhaps most strikingly dissimilar from the Belize Valley is the northern-style Kik Group ceramics introduced in the Sibun Valley during the

PAGE 129

Eleanor Harrison-Buck and Patricia A. McAnany Figure 10. a-g. Indian Creek Polychrome: a) Pakal Na b) Oshon, RIP 1286; c) Pakal Na, RIP 1089; d) Oshon, RIP 1313; e) Oshon, RIP 1312, f) Oshon, RIP 1285; g) Pechtun Ha, RIP 1297 (photograph & illustrations by E. Harrison-Buck). Terminal Classic period. The Kik Group ceramics, shown in Figures 10 and 11, are notably different from the ceramics of the Spanish Lookout Complex and, according to ceramic studies do not appear at sites in the upper Belize Valley. Li kewise, none of the Kik Group ceramics have been identified at the Hershey site in the upper reaches of the Sibun Valley. This further suggests that Hershey was perhaps more closely aligned with the Belize Valley and Petn overland trade network than the rest of the valley during the Terminal Classic period. In contrast, sites in the middle and lower reaches of the Sibun Valley that are closer to the coast Pakal Na, Pechtun Ha, Oshon, and Obispo have an abundance of Kik Group ceramics in terms of the total percentage represented at each site (Figure 12). The introduction of these foreign ceramic styles points to a different sphere of interaction developi ng along the eastern coast from which Hershey may have been excluded. The form, paste and surface treatment of these foreign ceramics suggest that they belong to the Kik Ceramic Group, established by Diane Chase (1982) at the site of Nohmul in northern Belize. The Kik Group ceramics are perhaps best defined as Puuc or Chichen Red Wares, showing strong northern attributes in terms of form and surface decoration. The Kik Group ceramics have a consistent yellowishorange paste and compri se a variety of large basin and bowl forms. These ceramics have a distinctive rim treatmenta bolstered rim with a rounded exterior fold in the lipthat makes these sherds relatively easy to identify in the Sibun assemblage. Figure 11. a-c. Fat Polychrome : a) Pechtun Ha, RIP 1007; b) Pechtun Ha, RIP 1087; c) Pechtun Ha, RIP 1300 (illustrations by E. Harrison-Buck & C. Cesario). 126

PAGE 130

Classic to Post Classic Investiga tions in the Sibun River Valley 0 0 0 0 0 0 0 0 0 0 0 2 4 6 8 10 12 14 XunantunichBaking PotHersheyPechtun HaOshon Archaeological SitesTotal Percent (%) of the Terminal Classic Assemblage Fat Polychrome Indian Creek Polychrome 7.96 12.50 8.18 8.79 Figure 12. Total percentage of Kik Group ceramics (Fat Polychromes and Indian Creek Polychromes) from sites in the Sibun and Belize Valleys. Indian Creek Polychro me (Figure 10) is a new type defined in th is study that has been placed in the Kik Group. The rounded bolster rim and interior of the vessel contain a hard, waxy, deep red slip that is generally well preserved. On the best-preserved examples, a reddish-orange slip with black and red polychrome painted designs is found on the exterior just below the rounded lip. Most sherds are considerably eroded, but faint traces of black and red painted designs, including cross-hatching and other abstract motifs, have been identified on some examples. While no complete vessels were recovered from the Sibun Valley excavations, a similar example of a basal break bowl containing a ring base and polychrome crosshatched design was found at Altun Ha (Gra ham 1987:Fig. 2g). Another Kik Group type found in the Sibun Valley is refe rred to as Fat Polychrome (Figure 11), which was defined by Shirley Mock at the site of Northern River Lagoon (NRL) on the northern Belize coast (Mock 1994:106-107 and Fig. 51; Mock 2005:128 and Fig. 7; Masson and Mock 2004: 387 and Fig. 17.7d-e). Mock (2005:128) notes that Fat Polychromes are found at Santa Rita on the Chetumal Bay in northern Belize and at Saktunja and the Salt Creek sites, located along the coast south of NRL. In addition, the type is found at San Jose in central Belize (Thompson 1939:124-125, Figs. 59 and 65). The basin form and pronounced bolster rim of the Fat Polychrome bare a strong resemblance to the Florescent Medium Puuc and Chichen Slate Ware basins (Brainerd 1958:52-53 and Figs. 41d, f, 43 a-c, and 73d 27-35; Smith 1971:Figs. 16d and g, and 27h-i). However, the paste characteristics, as well as the quantity and distribution of these northern-style ceramics 127

PAGE 131

Eleanor Harrison-Buck and Patricia A. McAnany Figure 13. a-e. Belize Molded-Carved: a) Pakal Na, RI P1001, b) Hershey, RIP 1002, c) Oshon, RIP 1032, d) Oshon, RIP 1034, e) Oshon, RIP 103 5 (illustrations by E. Harrison-Buck). suggest they were probably produced in Belize. In the Sibun assemblages, Fat Polychromes are generally less abundant than Indian Creek Polychromes in terms of total rim counts and it is possible the latter was produced in the Sibun Valley. Chemical studies are being performed on a sample of Kik Group ceramics from the Sibun, which will confir m where these types were produced. If not locally produced in the Sibun Valley, the di stribution pattern suggests they were probably manufactured somewhere near the coast in north-central Belize. Special Ceramics and Trade Wares Pabellon Modeled-Carved and Belize Molded-Carved Vessels Pabellon Modeled-Carved and Belize Molded-Carved ceramics are markers of the Terminal Classic period and have a wide distribution throughout the Maya Lowlands. The latter type appears in the Sibun Valley and elsewhere in Belize (Figure13). Belize Molded-Carved ceramics often are interpreted as imitations of the Fine Orange Pabellon Modeled-Carved ceramics (Aimers 2002:383; Mock 1994:318). However, Graham and others (1980:166) note the marked differences that separate Belize Molded-Carved from Pabellon Modeled-Carved, and suggest that it would probably be a mistake to view these vessels as imitations of the Fine Orange type. The Pabellon-Modeled type has two primary forms depicting two different scenes: a confrontation scene between two seated figures (vase with pedestal base) and a reclining figure (bowl with rounded sides) (Sabloff 1975:195 and Figs. 384386). Likewise, the Belize Molded-Carved ceramics show two fairly standardized vessel formsa pedestal-based pear-shaped drinking cup (see Thompson 1939:Fig. 84) and a barrel-shaped drinking cup with three hollow oven foot supports (Graham et al. 1980:Figs. 7 and 8; Thompson 1939:Fig. 83a and b). Graham and others (1980:166) have identified one set of scenes depicted on the Belize Molded-Carved vessels, referred to as Panels A and B. Panel A presents five people, while Panel B depicts seven people, 128

PAGE 132

Classic to Post Classic Investiga tions in the Sibun River Valley with two individuals standing in profile inte racting with a central elite figure. A portion of the scenes from Panels A and B are found on the examples illustrated in Figure 13, which are most similar to a complete vessel found at San Jose (see Thompson 1939:Fig. 83a). Glyphic bands occur in both Pabellon Modeled-Carved and Belize Molded-Carved. Adams (1971) and Sabloff (1973 and 1975) have noted northern Yucatan influence in the scenes on the Pabellon Modeled-Carved ceramics and argued this horizon marker signified an intrusive northern invasi on at sites, such as Seibal and Altar de Sacrificios, in the Maya heartland. Graham and others (1980:166) argue against the idea that the Belize Molded-Carved ceramics carry any northern influence, but Helmke et al. (1998) have identified the grammatical position and contexts of the Lu Bat glyph on Belize Valley vessels, which is similar to the use of the glyph on a Terminal Classic lintel at Chichn Itz. Both of these Terminal Classic ceramic types were produced using a mold or stamp. The transferred design on a leather-hard (pre-fired clay) vessel offered a template or guide for the potter who then delineated the lines with incising and carved out portions to create relief (Graham et al. 1980:164). Slipped red or orange, the rough edges of the carved design were rubbed smooth through burnishing, creating a polished surface and leaving the negative relief a rough matte surface. Although they do not represent the same scenes as depicted on the molded-carved ceramics, Brainerd (1958: Fig. 91g-o) illustrates a collection of stamps from Chichn Itz. Stamps and molds at Chichn Itz represent new ceramic production technology that was introduced during the Late-Termina l Classic period. The stamps would have facilitated the repeated imprinting of selected images on ceramic vessels, providing a standardized form and set of visual iconography. The use of stamps and mold made ceramics point to mass production of products in the Terminal Classic, manufactured explicitly for widespread distribution. Northern Imports There are several examples of basin forms found at Pechtun Ha, Obipso, and Oshon that have pronounced bolster rims like the Fat Polychrome type. However, the paste and surface treatm ent deviate from the Kik Group and may be direct imports from northern Yucatan, possibly Chichn Itz. One example is a basin form with a rounded bolster rim that was found in a Terminal Classic midden deposit associated with an elite household at the site of Pechtun Ha (Figure 14a). The form and reddish-brown paste with remnants of a light grayish slip match Smiths description of the Dzitas Slate ware basinsone of the most common slate ware types of the Sotuta Complex found at Chichn Itz (Brainerd 1958:5253; Smith 1971:174-177). Another basin fragment found in an elite residence at Oshon appears to be a northern import from the Cehpech Complex (Figure 14b). This basin form shows a small portion of a deeply incised design just below a pronounced rounded bolster rim. The vessel form, paste attributes, and inci sed decoration with remnants of gray slip on the exterior conform to Smiths (1971:177-178) description of Puuc Sl ate ware, possibly the Tekit Incised type. Smith (1971:168) notes that while Cehpech pottery is abundant at all Puuc sites in northwester n Yucatan, it is also quite well established at Chichn Itz. It is found in levels dating to the Terminal Classic period, coeval or at least partially overlapping with the Sotuta Ceramic Complex (Andrews et al. 2003). A distinctive basin form found on the floor of the interior room of th e circular structure at 129

PAGE 133

Eleanor Harrison-Buck and Patricia A. McAnany Obispo m ay represent another northern import. This vessel has a dark reddishbrown waxy slip and shows tooling on the exterior, underneath a pronounced triangular-shaped bolster rim (Figure 14c). The type is uniden tified, but the form, surface finish and paste characteristics do not resemble any of the other ceramics in the Sibun Valley assemblage. The basin form and triangular bolster r im point to a northern origin. Figure 14. a-c. Northern Imports (illustrations by E. Harrison-Buck and photographs by D. Buck). Concluding Thoughts The presence of northern imports at Pechtun Ha, Oshon, and Obispo the sites with Yucatec-style round structures points to a physical interactio n with northerners, possibly coming from Chichn Itz. We believe that, like the circular shrines, Yucatec imports and northern-style ceramics are materializations of a coastal sphere of interaction that developed during the Terminal Classic. Jeremy Sabloff (1977) and others (Kepecs et al. 1994; Masson and Mock 2004:367; Sabloff and Rathje 1975) have long noted the development of an extensive trade network along the eastern Caribbean coast, likely established and administered by Chichn Itz during the Terminal Classic. In this way, northern traits infiltrated coastal and riverine settlements as far south as the Sibun Valley in Belize, hundreds of kilometers away from this northern Yucatan regional center. The most common ceramic types found in the Sibun Valley the northernstyle Kik ceramics, Roaring Creek Red and Sibun Red Neckare the least common types in the Belize Valley. However, these types have been identified in Terminal Classic deposits at settlements located in the north-central part of Belize, such as Altun Ha, Lamanai, San Jose, Northern River Lagoon, and the Stann Creek District (Graham 1985 and 1987; Mock 1994; Masson and Mock 2004; Thompson 1939). Importantly, these sites are located either on the coast or along rivers and creeksin other words, at locales that are strategic for the movement of goods and ideas via canoe transport to and from the coast during the Terminal Classic period. The ceramic evidence points to two different spheres of in fluence that converged in this frontier zone of the eastern Maya Lowlands in the Terminal Classic period. Differential distribution patterns noted between the upper and lower reaches of the Sibun Valley suggests that not all settlements in the Sibun Valley were involved in this coasta l interaction sphere. Hershey, the site farthest inland, displays none of the Yucatec-style architecture and lacks the Kik Group ceramics. The Terminal Classic ceramic assemblage at Hershey shows the highest percentages of Tepeu/Spanish Lookout ceramic types, suggesting that this site continued to interact 130

PAGE 134

Classic to Post Classic Investiga tions in the Sibun River Valley with the Be lize Valley and Petn despite the failing political structures and economic networks in this cen tral region. Not surprisingly, the historical trajectory of Hershey mirrors many of the sites in the Belize Valley and Petn region, collapsing by the beginning of the ninth century. Meanwhile, sites such as Pakal Na, Pechtun Ha, Oshon and Obispo continue to thrive in the ninth and tenth centuries, mirroring the Terminal Classic florescence at Chichn Itz. The results of this study shed light on shifting spheres of interaction during the Terminal Classic period, with varied social, political and economic processes occurring at different sites in the Maya Lowlands at the time of the collapse. Acknowledgments The authors wish to acknowledge the financial support of the National Science Foundation (BCS0096603) and the Division of International Programs at Boston University. The project staff members, field school students, and local workmen who conducted the fieldwork reported upon in this analysis deserve a large note of appreciation. The Institute of Archaeology as part of the National Institute of Culture and Histor y (NICH) of Belize issued permission to conduct fieldwork in the Sibun Valley during 1997, 1999, 2001, and 2003, as well as la boratory analysis during 2002, 2004 and 2005. We thank the staff and directors at the Institute, particularly Dr. Jaime Awe, Dr. John Morris, Dr. George Thompson, Ms. Sherilyne Jones, Mr. Brian Woodye, and the entire Belize Archaeology Symposium Organizing Committee. These research reports are a testament of your hard work and commitment to preserving Belizes cultural heritage for the future. References Cited Adams, R. E. W. 1971 The Ceramics of Altar de Sacrificios Papers of the Peabody Museum of Archaeology and Ethnology, vol. 63, No. 1, Harvard University, Cambridge, MA. Aimers, J. J. 2002 Cultural Change on a Temporal and Spatial Frontier: Ceramics of the Terminal Classic to Postclassi c Transition in the Upper Belize River Valley Unpublished Ph.D. Dissertation, Tu lane University, New Orleans, LA. Andrews, A. P., E. W. Andrews V, and F. Robles C. 2003 The Northern Maya Collapse and Its Aftermath. Ancient Mesoamerica 14:151156. Audet, C. M. 2002 Pottery from the Caves of the Thumb District. In Sacred Landscape and Settlement in the Sibun River Valley: XARP 1999 Survey and Excavation edited by P. A. McAnany, pp. 79-102. SUNY Institute of Mesoamerican Studies Occasional Paper 8. Albany, NY. Brainerd, G. W. 1958 The Archaeological Ceramics of Yucatan University of California Anthropological Records, vol. 19. University of California Press, Los Angeles. Chase Diane Z. 1982 Spatial and Temporal Variability in Postclassic Northern Belize Ph.D. Dissertation, Department of Anthropology, University of Pennsylvania, Philadelphia. Diehl, R. A. and J. C. Berlo (eds) 1989 Mesoamerica after the Decline of Teotihuacan, AD 700-900 Washington, DC: Dumbarton Oaks Research Library and Collection. Gifford, J. C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley Peabody Museum Memoirs, vol. 18. Harvard University, Cambridge, MA. 131

PAGE 135

Eleanor Harrison-Buck and Patricia A. McAnany Graham, E. A. 1985 Facets of Terminal to Postclassic Activity in the Stann Creek District, Belize. In The Lowland Maya Postclassic edited by A. F. Chase and P. M. Rice, pp. 215-230. University of Texas Press, Austin. 1987 Terminal Classic to Early Historic Period Vessel Forms From Belize. In Maya Ceramics: Papers from the 1985 Maya Ceramic Conference edited by P. M. Rice and R. J. Sharer, pp. 73-98. BAR International Series 345(I). 1994 The Highlands of the Lowlands: Environment and Archaeology in the Stann Creek District, Belize, Central America Prehistory Press, Madison. Graham. E. L. McNatt, M. A. Gutchen 1980 Excavations in Footprint Cave, Caves Branch, Belize. Journal of Field Archaeology 7(2) L153-172. Harrison, E. 2003 A Circular Shrine and Repositioned Stelae at the Oshon Site (Operation 24). In Between the Gorge and the Estuary: Archaeological Investigations of the 2001 Season of the Xibun Archaeological Research Project edited by P. A. McAnany and B. S. Thomas, pp. 165-185. Submitted to the Department of Archaeology, Belmopan, Belize. http://www.bu.edu/tric ia/reports/xarp2003/in dex.shtml Harrison E. and K. Acone 2002 Further Investigations at Pechtun Ha: Feasting and Mass Importation of Cave Speleothems. In Sacred Landscape and Settlement in the Sibun River Valley: XARP 1999 Survey and Excavation edited by P. A. McAnany, pp. 123-140. SUNY Institute of Mesoamerican Studies Occasional Paper 8. Albany, NY. Harrison-Buck, E. 2007 Materializing Identity Among the Terminal Classic Maya: A Study of Architecture and Ceramics in the Sibun Valley, Belize Unpublished Ph.D. dissertation, Department of Archaeology, Boston University, Boston, MA. Harrison-Buck, E. and P. A. McAnany 2006 Terminal Classic Circular Shrines and Ceramic Material in the Sibun Valley, Belize: Evidence of Northern Yucatec Influence in the Eastern Maya Lowlands In Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2005 Belize Archaeology Symposium, Research Reports in Belizean Archaeology Volume 3, edited by J. Morris, S. Jones, J. Awe and C. Helmke, pp. 287-299. Belize Print Ltd., Belmopan, Belize. Helmke, C. G., P. R. Colas, and J. J. Awe 1998 Comments on the Typology, Epigraphy and Iconography of the Actun Tunichil Muknal Vase and the Belize Valley Modeled-carved Vessels. In The Western Belize Regional Cave Project: a report of the 1997 Field Season edited by J. J. Awe, pp. 93-140. Department of Anthropology Occasional Paper, No. 1. University of New Hampshire, Durham. Kepecs, S., G. Feinman, and S. Boucher 1994 Chichn Itz and Its Hinterland, a World-Systems Perspective. Ancient Mesoamerica 9(1):141-158. LeCount, L. J. 1996 Pottery and Power: Feasting, Gifting, and Displaying Wealth Among the Late and Terminal Classic Lowland Maya. Unpublished Ph.D. dissertation, UCLA. 2005 Continuity and Change in the Ceramic Complex of Xunantunich During the Late and Terminal Classic Periods. In Geographies of Power: Understanding the Nature of Terminal Classic Pottery in the Maya Lowlands edited by S. L. Lpez Varela and A. E. Foias, pp. 93-106. BAR International Series 1447, Oxford. Masson, M. A. and S. B. Mock 2004 Ceramics and Settlement Patterns at Terminal Classic-Period Lagoon Sites in Northern Belize. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation, edited by A. A. Demarest, P. M. Rice, and D. S. Rice, pp. 367-401. University Press of Colorado, Boulder. 132

PAGE 136

Classic to Post Classic Investiga tions in the Sibun River Valley 133 Mock, S. B. 1994 The Northern River Lagoon Site (NRL): Late to Terminal Classic Maya Settlement, Saltmaking, and Survival on the Northern Belize Coast Ph.D. Dissertation, The University of Texas at Austin. 2005 Pushing the Limits: Late to Terminal Classic Settlement and Economies on the Northern Belize Coast. In Geographies of Power: Understanding the Nature of Terminal Classic Pottery in the Maya Lowlands edited by S. L. Lpez Varela and A. E. Foias, pp. 121-133. BAR International Series 1447, Oxford. Pendergast, D. M. 1974 Excavations at Actun Polbilche, Belize Royal Ontario Museum Monograph 1. Toronto, Ontario, Canada. 1990 Excavations at Altun Ha, Belize, 19641970, Vol. 3. Royal Ontario Museum, Toronto, Canada. Peterson, P. A. 2006 Ancient Maya Ritual Cave use in the Sibun Valley, Belize Ph.D. Dissertation, Boston University, Department of Archaeology, Boston, MA. Reents, D. 1981 The 1978 Caves Branch Cave Archaeological Project Ceramic Analysis Manuscript on file, Department of Archaeology, Belmopan, Belize. Ruppert, K. 1935 The Caracol at Chichn Itz, Yucatn, Mexico Publication No. 454. Washington, DC: Carnegie Institution of Washington. Sabloff, J. A. 1973 Continuity and Disruption during Terminal Late Classic Times at Seibal: Ceramic and Other Evidence. In The Classic Maya Collapse edited by T. P. Culbert, pp. 107-132. University of New Mexico Press, Albuquerque. 1975 Excavations at Seibal: Ceramics Memoirs of the Peabody Museum of Archaeology and Ethnology Vol. 13, No. 2. Harvard University, Cambridge, MA. 1977 Old Myths, New Myths: The Role of Sea Traders in the Development of Ancient Maya Civilization. In The Sea in the PreColumbian World edited by E. P. Benson, pp, 67-97. Dumbarton Oaks Research Library and Collections, Washington, DC. Sabloff, J. A. and W. J. Rathje 1975 A Study of Changing Pre-Columbian Commercial Systems: The 1972-1973 Seasons at Cozumel, Mexico Monographs of the Peabody Museum of Archaeology and Ethnology, No. 3. Harvard University, Cambridge, MA. Smith, R. E. 1971 The Pottery of Mayapan: Including Studies of Ceramic Material from Uxmal, Kabah, and Chichn Itz. Papers of the Peabody Museum of Archaeology and Ethnology Vol. 66. Harvard University, Cambridge, MA. Thompson, J. E. S. 1939 Excavations at San Jose, British Honduras Publication No. 506. Carnegie In stitution of Washington, Washington, DC.

PAGE 137

12 THE TERMINAL CLASSIC IN THE THREE RIVERS REGION Lauren A. Sullivan, Brett A. Houk, and Fred Valdez, Jr. _____________________________________________________________________________ The Three Rivers Region was extensively occupied and utilized during the Terminal Classic. This paper focuses on the last occupation of the region, during the Terminal Classic, as represented by material remains, and the implications of interaction with neighboring areas. By the end of this period, the large centers and rural sites of the Late Classic were virtually abandoned with relatively little evidence of Postclassic habitation. The collapse of elite control and authority are discussed in light of recent arch itectural and ceramic data, which suggest a major cultural transformation at the end of the Terminal Classic. Introduction The data discussed in this paper were recovered from the Programme for Belize Archaeological Project (PfBAP) and the Chan Chich Archaeological Project (CCAP), both located in the Three Rivers Region of northwestern Belize (Figure 1). The Three Rivers Region encompasses parts of Belize, Guatemala, and Mexico (Adams 1995). Three escarpmentsthe Booths River escarpment, the Rio Bravo escarpment, and the La Lucha escarpment form the principal topographic feat ures of the area. Previous research has demonstrated that this area was extensively utilized, with at least five urban centers (La Milpa, Dos Hombres, Gran Cacao, Maax Na, and Great Savannah) and numerous smaller sites, groups, houses, terraces, and other features thriving by the end of the Late Classi c on the Belizean side of the region (Valdez and Sullivan 2006; Walling et. al. 2005). While occupation in the area began during the Middle Preclassic, the later half of the Late Classic (Tepeu 2-3, ca. A.D.700 to 850) had the highest population levels (Adams et al. 2004). At this time, there was an increase in the number of sites of all sizes and in the construction of monumental architecture. As has been suggested for the entire Maya Lowland area, this population increase throughout the Late Classic and into the Terminal Classic most likely stressed the agricultural system, taxed the natural resources, and led to increasing competition among the elite (Shimkin 1973; Willey 1973). By the end of the Terminal Classic/Tepeu 3, the Maya experienced a significant reorganization as documented through the lowland Maya area. The mechanisms that triggered the decline have long been debated, with climatic factors (Gill 1994), land exhaustion (Adams 1995), warfare (Sabloff and Willey 1967; Webster 1977), and internal strife (Buttles and Valdez 2006; Valdez 1989) all cited as contributing elements. Further complicating the issue is the fact th at the collapse was not uniform across the Maya area (D. Chase and A. Chase 2004). Many of the large centers of the Peten, such as Tikal, Uaxactn, and Yaxha, were virtually abandoned by AD 850, while sites in the northern lowlands thrived for several more centuries (Demarest et. al. 2004). Other problems in the discussion of the Terminal Classic stem from the origin and use of the term Terminal Classic. This term was originally used at the 1965 Maya Lowland Ceramic Conference to mark the transition from the Classic to Postclassic based on ceramic characteristics (Willey et al. 1967). The term eventually came to be associated with the idea of widespread Research Reports in Belizean Archaeology Vol. 4, 2007, pp.135-146. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 138

Terminal Classic Investigations in the Three Rivers Region Figure 1. Map of Three Rivers Regions Figure 2 A partially reconstructable Fine Orange vessel collapse, destruction, and the end of the cultural practices associated with the Classic Maya (Rice et. al. 2004). Extensive archaeological research over the past several decades across the Maya area has successfully demonstrated the highly variable nature of this time period. Data from the Three Rivers Region indicate that after a pr osperous Late Classic to Terminal Classic there was a cessation of monumental construction, a major decline in rural population (Robichaux 1995; Adams 1999), and a decrease in ceramic production (Sullivan and Valdez 2004). Evidence for Postclassic occupation in the Three Rivers Region is scarce, with little to no evidence of a rural population or construction at the large sites above the escarpments 136

PAGE 139

Sullivan et al. (Robichaux 1995). Jeff Durst (1996) reported Postclas sic ceramics from excavations at Gran Cacao; however, the nature and extent of Post classic at the site is not yet fully understood. Recent work at the site of Akab Muclil, a small settlement near the banks of the Rio Bravo, revealed an Early Postclassic constr uction phase (Padilla et al. 2006). However, to date we have yet to define Postclassi c occupation at sites above the escarpments. These data suggest a breakdown of elit e control and the subsequent abandonment of most of the Three Rivers Region by the end of the Terminal Classic/Tepeu 3 (ca. A.D. 850 900) Architectural Data Towards the end of the Terminal Classic, construction techniques in the region were clearly not what they once were. As previously stated, the construction of monumental architectur e virtually halted. The structures that are built appear to have been quickly made without the mortared retaining walls, care fully placed facing stones, thick plastered floors, or platform heights that were obs erved during the Late Classic. Another change was the construction of buildings in the middle of formerly sacred plazas. Low-walled structures made from stones stolen from existing Late Classic monumental structures and constructed on top of Late Classic plaza surfaces are observed or suspected at La Milpa (Structure 86) (Tourtellot and Rose 1993), Dos Hombres (Houk 1996), and Kinal (Hageman 1992). At the smaller site of Dos Barbaras, Lewis (2005) has documented a significant Late Classic occupation for most of the site with the addition of one Courtyard (Group G) that appears to have been constructed in the Terminal Classic. The primary structure associated with this courtyard had moderate amounts of cut-stone masonry and plaster flooring, while the foundation of the smaller structure was composed of small-medium sized cobble fill. Lewi s (2005) also notes a plain stela that appear s to be have been erected late in the development of Dos Barbaras and may represent an attempt by the elite of the site to exercise some degree of political autonomy during the political instabilities of the Terminal Classic (Lewis 2005). At Altun Ha, Pendergast (1992) suggests that the elite gradually lost control over labor forces, which resulted in a decrease in architectural construction, repair, and modification in central areas. Although the elite of Altun Ha did make a point to maintain and modify their own residential structures as if to keep up appearances, their leadership roles must have been drastically altered from what they once were (Pendergast 1992). At Tikal, when structures were renovated the difference in architectural quality and finish was notable (Valdes and Fahsen 2004: 151). These data all point towards a deteriorating economy, with the elites lacking the resources to maintain power and prestige at their former levels however hard they may have tried (Adams et. al. 2004; Sullivan and Valdez 2004). Some of the best architectural data for the Terminal Classic come from La Milpa, where Hammond and Tourtellot (2004) report major construction projects initiated but left unfinished in the early ninth century A.D. These half-finished buildings were silent witnesses to the collapse of the site but do little to explain the reasons for the centers abandonment. As Hammond and Tourtellot (2004:300) put it, the substantial, dense population documented throughout the settlement area seems to have melted away with no sign of violence and very little in the way of post-collapse squatter occupation. 137

PAGE 140

Terminal Classic Investigations in the Three Rivers Region Ceramic D ata In the Three Rivers Region a clear Tepeu 2-3 ceramic component has been defined with a very strong continuity between the Late Classic (Tepeu 2) and the Terminal Classic (Tepeu 3). However, one of the problems in the region is trying to define a clear break between the Late Classic (Tepeu 2) and the Terminal Classic (Tepeu 3). Part of this problem may be due to the fact that Terminal Classic ceramics might be related to occupations of earlier (Tepeu 2) structures without the addition of new construction layers to aid in clearer chronological sepa ration (Graham 1985; Lincoln 1985). Another issue is the absence of typical Terminal Classic ceramic markers such as Daylight Orange: Darknight variety, Fine Orange, Plumbate, and other Terminal Classic finewares at the majority of upland sites in the area. Unfortunately, these ceramic types are generally associated with elite groups, and their absence does not necessarily reflect a lack of Terminal Classic populations (A. Chase and D. Chase 2004). Smith (1955) had similar problems in his original description of Tepeu 1, 2, and 3 referring to a Tepeu General category, which included styles that could not be clearly associated w ith a Tepeu subphase (e.g. Tepeu 1, 2, and/or 3). While he was able to use well-preserved mortuary vessels to help clarify these divisions, he had a much harder time with utilitarian vessels. Smith (1955:24) suggested that Tepeu 1 and 2 ceramics were likely used by the people of Tepeu 3 and that precursors to Tepeu 3 styles could be identi fied in Tepeu 1 and 2 contexts, further obscuring firm divisions. Ceramic data from the Three Rivers Region do indicate the following trends from the Late to the Terminal Classic: an increase in utilitarian wares (specifically large unslipped and striated jars); an increase in the thickness of the these large vessels (especially at the lip) (Adams and Adams 2000); an increase in the use of face vessels with appliqud and modeled decoration; the appearance of slate ware and/or imitation slate ware; a decrease in the predominant Late Classic types such as Achote Black, Subin Red, and Tinaja Red; and a decrease in overall ceramic diversity and quality. By the end of the Terminal Classic, the ceramics were generally of poor quality, with crumbly pa ste, large, coarse, and poorly sorted inclusions, and little to no slip. Examples of these large utilitarian unslipped jars can be found at Dos Hombres and Dos Barbaras. One of the two Late Classic burials from the site center of Dos Hombres was placed in a lip to lip cache consisting of a Cayo Unslipped vessel covered by a Subin Red bowl (Houk 1996; Sullivan 2002). A similar pattern is observed in Tepeu 2-3 bu rials from the site of Dos Barbaras. A burial consisting of a large Cayo Unslipped vessel covered by a Subin Red bowl was found in front of the largest structure at Dos Barbaras (Lewis 2005; MeBar and Lewis 2005; Sullivan 2002). An increase in utilitarian wares has also been noted at othe r sites in the region (Sullivan and Sagebiel 2003). At Tikal, Burial 77 (A.D. 771) is thought to contain the remains of the latest known ruler as recorded on Stela 11 (A. Chase 1992). Coggins (1975) has observed that the pottery recovered from Burial 77 is the same in form and function as the pottery recovered from a typical Late Classic residential burial assemblage. This phenomenon was also noted at Uaxactn where Smith (1955:6) states: Tepeu buria ls contain no obvious trade pieces and no vessels specially made for interment (aside from a few exceptions). Similarly, Ball (1993:263) notes a de-emphasis of painted wares in the northern lowlands during the Late Classic. 138

PAGE 141

Sullivan et al. The increase in the num ber of human face effigy vessels may reflect a move away from more abstract religious symbolism and toward anthropomorphism similar to what was observed at Altar de Sacrificios (Adams 1973). Diane Chase (1992) suggests that this move away from abstract symbolism, which began in the Late Classic, was a trend not fully realized until the Postclassic with the Classic Collapse marking the religious transformation of Maya society. Two caches, recovered from the site of Dos Barbar as, consisted of two unslipped face vessels flanking the stairway of the largest structure at the site (Group B Structure 6). Similar face vessels were found at th e site of Maax Na below the stairway of an elite residential structure and may be related to the remodeling of the building. At the site of Bolsa Verde, a face vessel was recovered in a cache associated with the one of the latest construction phases of the largest structure at the site (King and Shaw 2003). An unslipped and appliqud style vessel that contained unmodified Spondylus shell and a piece of modified shel l was recovered from the stairway of the largest structure at Las Abejas The dating of these vessels is somewhat problematic in that they are associated with other Late Classic types and are not restricted to pure Terminal Classic deposits. In some cases (e.g., Las Abejas, Dos Barbaras) these vessels are associated with the last construction phase and may be associated with the termination and/or abandonment of a structure. Previous research in the area (Sullivan 2002; Sullivan and Sagebiel 2003) has suggested that throughout the Late Classic there was an increase in local autonomy in terms of ceramic production and distribution as well as a breakdown of the prestige good exchange system that characterized earlier times. Balls (1993) work in the Mopan-Macal area and including a range of si tes (i.e., Buenavista, Cahal Pech, Guerra, Nohock Ek, and Eden) also points strongly towards localized ceramic complexes during the Late Classic, correlating these changes to a more fragmented political environment which sets the stage for the Terminal Classic decline. While in other areas (small sites along the coast, Lamanai, Altar de Sacrificios, Seibal, and in the Yucatan) the Terminal Classic brings with it a prolif eration of new ceramic types hinting at incr easing interregional exchange and communication (Adams 1971; Graham 2004; Mock 2004; Sabloff 1975), this is clearly not the case for most of the Three Rivers Region. The majority of slate ware recovered most likely represent locally made imitations of Ticul Thin Slate and may represent a feeble effort on the part of the elite to associate themselves with the growing sites to the north. In terms of identifying any continuing Postclassic occupation through ceramics, only a few Postclassic sherds have been located at sites on the escarpment, including a whole censer from the summit of a mound at Chan Chich (Guderjan 1991), censer fragments from the base of Stela 2 at Dos Hombres (Houk 1996), and from Stela 7 and Stela 12 at La Milpa (Hammond and Bobo 1994). The presence of these few censers probably indicates pilgrimages to major Classic period centers and does not represent Postclassic occupation (Hammond and Bobo 1994). A Postclassic jaguarshaped plumbate foot was recovered by Stan Walling at Chawak Butoob, a Late to Terminal Classic commoner residential site along the Rio Bravo Escarpment. Recent excavations at Akab Muclil, which is at the base of the Rio Bravo escarpment in a much different ecological setting, suggest some Early Postclassic occupation of the region may have occurred (Padilla et al. 2006). 139

PAGE 142

Terminal Classic Investigations in the Three Rivers Region Ritual Deposits Further evid ence of Terminal Classic activities is found in the deposits recovered from the sites of Dos Hombres and Chan Chich that include larg e quantities of broken ceramics and other artifacts placed on elite residential courtyard floors or steps to buildings (Adams et al. 2004; Houk et al. 1999). These deposits are open to various interpretations until additional study is performed on them, but, at least superficially, they resemble middens in terms of their comp osition, but ritual termination deposits in terms of their contexts (Adams et. al. 2004; Houk 2003; Houk et. al. 1999). The deposition of broken pottery, dark soil, and other remains on the top of final co nstruction phases is seen at many sites in Belize, including Blue Creek (Clayton et. al. 2005), Blackman Eddy (Garber et. al. 1998; 2004), Punta de Cacao (Guderjan et al. 1991), and Cerros (Garber 1989), and has often been interpreted as specific ritual behavior aimed at terminating the structure. When viewed as a group, however, there is much variation in these features, and it is likely that multiple activities (some ritual, others not) are responsible for their formation. At Dos Hombres, approximately 4.3 cubic meters of a midden-like deposit (Problematic Deposit 2) were excavated in a test pit in a small, elite courtyard at the entrance to the elevated acropolis at the site. This deposit was approximately 40 cm thick and had been pla ced on the last floor of the courtyard. The vast majority of artifacts were broken, but the deposit contained some of the most exotic artifacts recovered from any context at the site, including numerous part ially reconstructable vessels, an eccentric bi face of chert from northern Belize, a roller stamp, a figurine head with an elaborate bird headdress, a ceramic animal face, a drilled jaguar tooth, an obsidian biface, numerous obsidian blade fragments, and an anthropomorphic whistle. The ceramics from the deposit comprised over 6,700 sherds, which included primarily utilitarian types (striated or unslipped) and a variety of finewares, including a Cubeta Incised sherd with hieroglyphs, Daylight Orange: Darknight variety plate fragments, and Palmar Orange Polychrome vessel sherds. The Daylight Orange: Darknight variety sherds provide the Tepeu 3, Terminal Classic, age assessment for the deposit. If this deposit were a midden related to Terminal Cl assic occupation of the Acropolis, faunal remains, in addition to artifactual material, would be expected. The only bone identified from the excavated sample, however, included approximately five fragments of unidentifiable animal bone, three pieces of worked bone, 10 fragments of turtle carapace (perhaps from a rattle), a drilled jaguar tooth, and five human cranial fragments (Houk 1996:Table C.10). At Chan Chich, Terminal Classic artifacts were encounter ed on the steps of two palace structures in the western half of the site (Houk 2001; Houk et al. 1999). One structure (Structure C-2) is located in Courtyard C-1 (Normans Temple); a tightly enclosed group built on a heavily modified hilltop. The other structur e (Structure C-6) is located in Plaza C-2, near the base of the same hill (Houk 2001). Partial vessel fragments, exotic artifacts, and human skeletal material were concentrated on the lower three steps on both of these structures (Structures C-2 and C-6). Although not as dense as the Dos Hombres deposit, these features are comparable in composition. They contained numerous partially reconstructable vessels, figurine fragments, shell artifacts, obsidian blades, manos, incised bone, and a whistle fragment. Faunal remains were absent or represented by very small samples, including a jaguar canine discovered during the 2001 excavations. A partially 140

PAGE 143

Sullivan et al. reconstructable Fine Orange vessel (Figure 2) and sherds from an imitation Fine Orange vessel date these features to the Terminal Classic (Houk 2002). The significance of the features from Chan Chich and Dos Hombres lies in their context and composition. They lack faunal remains that would suggest the deposits are middens, and they are found either on steps to palaces (as at Chan Chich) or physically blocking important access points (as at Dos Hombres). The frequency of artifacts that could be termed exotic bear similarities to termination events, but Houk (2001, 2003) has argued that these features are secular in nature and presumably tied to rather tumultuous last days at the two sites. Perhaps the case of Special Deposit 1 at Blue Creek demonstr ates the difficulties in interpreting these Terminal Classic deposits; two distinct and at odds interpretations have been put forth to explain the origin of that deposit by Guderjan (2004) and Clayton et al. (2005 ). Special Deposit 1 was a dense concentr ation of ceramic sherds and dark soil [that] was encountered on Structure 3s stairway at Blue Creek (Clayton et al. 2005:122). This feature was located along the buildings centerline and extended 1.2 m into the plaza and 2.2 m up the slope of the mound, and the material entirely covered the re mains of the stairway shrine and fanned out from its sides, burying the lowest four to five steps of the stairway (Clayton et al. 2005:122). Guderjan (2004:239) concluded, perhaps through a misunderstanding of the conclusions reached by Clayton et al. (2002), that the deposit was a large-scale termination ritual that included smashing pots against the facade of the [Structure 3]. Cl ayton et al. (2005:128), however, put forth a compelling argument that the feature repr esents a secondary deposit of midden material and appears to represent the remains of complex termination activities directly associated with the abandonment of the structure, and perhaps the site itself, but much of the material used in the Terminal Classic event had been produced by long-term feastingrefuse discard behavior. The feasting argument is an interesting one, and pe rhaps the deposit at Dos Hombres bears re-examination in light of it. However, Problematic Deposit 2 from Dos Hombres, with its high frequency of non-ceramic, elite artifacts, and its low frequency of faunal material, does not seem to be feasting debris in either primary or secondary context. Others have interpreted similar Terminal Classic deposits as middens of squatters (see Culbert 1973), and Helmke (2006) has documented Terminal Classic occupation debris features at Pooks Hill 1 in the Roaring Creek Valley, south of the Three Rivers Region. The deposits described by Helmke (2006:182) include terminal occupation de bris that comprises secondary, commingled deposits of artefactual remains. These deposits typically include a mixture of domestic artifacts, a few special finds, and the remains of food refuse and faunal materials, and were found either directly atop and/or abutting terminal phase architecture, usually in internal corners formed by terraces and stairs or corners between structures (Helmke 2006: 182). Helmke (2006) concludes reason ably that these may represent piles formed from sweeping clean steps or high-traffic areas. Conclusions The data presented here demonstrate that a drastic change in regional organization occurred during the Terminal Classic. The upland area of this region along the escarpments that was heavily utilized in the Late Classic was virtually abandoned. The quality of the architecture declined and construction ceased, ceramic quality and 141

PAGE 144

Terminal Classic Investigations in the Three Rivers Region quantity decreased, and the nature of the ritu al deposits hint at the destruction of elite culture as it was once known. Regardless of the specific mechanisms behind these changes it is clear that the elite lost the control and power they once had, resulting in the abandonment of ceremonial centers and a depopulation of the countryside The archaeological evidence of abandonment and its aftermath is problematic and debatable, as demonstrated in the multiple attempts to characterize and understand so called problematic Terminal Classic deposits and their contexts. As described above, the examples from Pooks Hill 1 south of the region, as well as those described at Blue Creek by Clayton et al (2005) seem fundamental ly different from the Terminal Classic deposits at Chan Chich and Dos Hombres. It is likely that different activities resulted in the formation of these features. Furthermore, similar deposits have yet to be found at other major centers, particularly La Milpa. This observation calls into question Houks (2000, 2001) hypothesis that a regional calamity befell the elite of the region at the end of the Terminal Classic, and it reinforces the notion that the timing and nature of elite abandonment of the major centers was highly variable. Each site may have followed its own unique trajectory along the path to collapse; some experiencing violence and others going out with a whimper, as Hammond and Tourtellot (2004:301) have proposed for La Milpa. Overall, the nature of the Terminal Classic demise of the major sites in the Three Rivers Region is poorly understood, and this situation is even worse for the rural settlement areas for which we have little to no data relating to their abandonment. Acknowledgments The ideas in the paper are the result of work by many individuals from multiple instituti ons associated with research in the Three Rivers Region, and we would like to thank the participants of not only the Programme for Belize Archaeological Project, but those of the Blue Creek Archaeological Project, the Blue Creek Regional Political Ecology Project, the La Milpa Archaeological Project, and the Chan Chich Archaeological Project, as well, for all their contributions to our understanding of ancient Maya society in the region. We thank the Institute of Archaeology (IA), especially John Morris and Jaime Awe, for supporting and encouraging our continued work in Belize. We also thank the IA for organizing this symposium and promptly publishing the papers every year. References Cited Adams, R.E.W. 1971 The Ceramics of Altar de Sacrificios. Papers of the Peabody Museum of Archaeology and Ethnology 63(1). Harvard University, Cambridge, Massachusetts. 1973 Maya Collapse: Transformation and Termination in the Ceramic Sequence at Altar de Sacrificios. In The Classic Maya Collapse edited by T. P. Culbert, pp. 133163. School of American Research, University of New Mexico Press, Albuquerque. 1995 The Programme for Belize Regional Archaeological Proj ect: 1994 Interim Report, Introduction. In The Programme for Belize Archaeological Project, 1994 Interim Report edited by R. E.W. Adams and F. Valdez, Jr., pp. 1-14. The Center for Archaeology and Tropical Studies and The University of Texas at San Antonio 1999 Rio Azul: An Ancient Maya City University of Oklahoma Press, Norman. Adams, R.E.W. and Jane Jackson Adams 2000 Rio Azul Ceramic Sequence Summary. In Rio Azul Reports Number 5, The 1987 Season edited by R.E.W. Adams, pp. 264272. The University of Texas at San Antonio. 142

PAGE 145

Sullivan et al. Adams, R. E. W., H. R. Ro bichaux, Fred Valdez, Jr., Brett A. Houk, and Ruth Mathews 2004 Transformations, Periodicity, and Urban Development in the Three Rivers Region. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 324341. University Press of Colorado, Boulder. Buttles, Palma J. and Fred Valdez Jr. 2006 Terminal Classic Ev ents: Colha, Belize and the Central Maya Lowlands. Paper presented at the 2006 Belize Archaeology Symposium. Belize City, Belize. Ball, Joseph W. 1993 Pottery, Potters, Palaces, and Polities: Some Socioeconomic and Political Implications of Late Classic Maya Ceramic Industries. In Lowland Maya Civilization in the Eighth Century A.D ., edited by J.A. Sabloff and J.S. Henderson, pp. 243-272. Dumbarton Oaks, Washington, D.C. Chase, Arlen F. 1992 Elites and the Changing Organization of Classic Maya Society. In Mesoamerican Elites: An Archaeological Assessment edited by D. Z. Chase and A. F. Chase, pp.30-49. University of Oklahoma Press, Norman and London Chase, Diane Z. 1992 Postclassic Maya Elites: Ethnohistory and Archaeology. In Mesoamerican Elites: An Archaeological Assessment edited by D. Z. Chase and A. F. Chase, pp. 118-134. University of Oklahoma Press, Norman and London. Chase, Arlen F. and Diane Z. Chase 2004 Terminal Classic Status-Linked Ceramics and The Maya Collapse. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, University Press of Colorado, Boulder. Chase, Diane and Arlen F. Chase 2004 Hermeneutics, Transitions, and Transformations. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 12-27. University Press of Colorado, Boulder. Clayton, Sarah C., Laura J. Kosakowsky, and W. David Driver 2002 Rubbish or Ritual? An Examination of Terminal Deposits at Blue Creek. Paper presented at the 67th Annual Meeting of Society for American Archaeology, 20 March, Denver, CO. Clayton, Sarah, W. David Driver, and Laura Kowsakowsky 2005 Rubbish or Ritual? Contextualizing a Terminal Classic Problematical Deposit at Blue Creek Belize. Ancient Mesoamerica 16: 119-130. Coggins, Clemency C. 1975 Painting and Drawing Styles at Tikal: An Historical and Iconographic Reconstruction. Unpublished Ph.D. Dissertation, Department of Anthropology, Harvard University. Culbert, T. Patrick 1973 The Maya Downfall at Tikal. In The Classic Maya Collapse edited by T. Patrick Culbert, pp. 63-92. University of New Mexico Press, Albuquerque. Demarest, Arthur A., Prudence M. Rice, and Don S. Rice 2 004 Assessing Collapse, Terminations, and Transformations. The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 545-572. University Press of Colorado, Boulder. Durst, Jeffrey J. 1996 Excavations at Gran Cacao (RB 43) Site Investigations in Northwestern Belize. Unpublished M.A. Thesis, Department of Anthropology, The University of Texas at San Antonio. Garber, James F. 1989 Archaeology at Cerros, Belize, Central America, Volume II, The Artifacts Southern Methodist University Press, Dallas. Garber, James F., W. David Driver, Lauren A. Sullivan, and David M. Glassman 143

PAGE 146

Terminal Classic Investigations in the Three Rivers Region 1998 Bloody Bowls and Broken Pots: The Life, Death, and Rebirth of a Maya House. In The Sowing and the Dawning: Termination, Dedication, and Transformation in the Archaeological and Ethnographic Record of Mesoamerica edited by Shirley Boteler Mock, pp. 125 134. University of New Mexico Press, Albuquerque. Garber, James F., M. Kathryn Brown, W. David Driver, David M. Glassman, Christopher J. Hartman, F. Kent Reilly III, and Lauren A. Sullivan 2004 Archaeological Investigations at Blackman Eddy, Belize. In Archaeology in the Belize Valley: Half a Century of Maya Settlement Studies edited by James F. Garber, pp. 48-69. University Press of Florida, Gainesville. Gill, Richard 1994 The Great Maya Droughts. Unpublished Ph.D. dissertation, Department of Anthropology, The University of Texas at Austin. Graham, Elizabeth 1985 Facets of Termin al to Post-Classic Activity in the Stann Creek District, Belize. In The Lowland Maya Postclassic A. F. Chase and P. M. Rice, eds., pp 215-229. University of Texas Press: Austin. 2004 Lamanai Reloaded: Alive and Well in the Early Postclassic. In Research Reports in Belizean Archaeology, Vol. 1 edited by Jaime Awe, John Morris, and Sherilyne Jones, pp. 223-241. Institute of Archaeology, Belmopan, Belize. Guderjan, Thomas H. 1991 Chan Chich. In Maya Settlement in Northwestern Belize: The 1988 and 1990 Seasons of the Rio Bravo Archaeological Project edited by T. H. Guderjan, pp. 3550. Maya Research Program and Labyrinthos, California. 2004 Public Architecture, Ritual, and Temporal Dynamics at the Maya Center of Blue Creek, Belize. Ancient Mesoamerica 15: 235-250. Hammond, Norman, and Matthew R. Bobo 1994 Pilgrimages Last Mile: Late Maya Monument Veneration at La Milpa, Belize. World Archaeology 26:19-34. Hammond, Norman, and Gair Tourtellot 2004 Out With a Whimper: La Milpa in the Terminal Classic. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 288301. University Press of Colorado, Boulder. Hageman, Jon 1992 The 1991 Test Pit program at Kinal, Guatemala. Unpublished M.A. report, Department of Anthropology, The University of Texas at Austin. Hanratty, Catherine C. 1998 Excavations in the Structure 37 Courtyard. Paper pres ented at the 63rd annual meeting of the Society for American Archaeology, Seattle, Washington. Helmke, Christophe G. B. 2006 A Summary of the 19992 Seasons of Archaeological Investigations at Pooks Hill, Cayo District, Belize. In Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2005 Belize Archaeology Symposium edited by John Morris, Sherilyne Jones, Jaime Awe, and Christophe Helmke, pp. 173192. Research Reports in Belizean Archaeology Volume 3. Institute of Archaeology, National Institute of Culture and History, Belmopan, Belize. Houk, Brett A. 1996 The Archaeology of Site Planning: An Example from the Maya Site of Dos Hombres, Belize. Unpublished Ph.D. dissertation, Department of Anthropology, The University of Texas at Austin. 2000 Life, the Universe, and Everything: Reevaluating Problematic Deposit 2 from Dos Hombres, Belize. In The 1998 and 1999 Seasons of the Chan Chich Archaeological Project edited by Brett A. Houk, pp. 141 150. Papers of the Chan Chich Archaeological Project, Number 4. Mesoamerican Archaeological Research Laboratory, The University of Texas, Austin, Texas. 2001 All Good Things: The Three Rivers Region ca. AD 850. Paper presented at the 66th annual meeting of the Society for 144

PAGE 147

Sullivan et al. American Archaeology, New Orleans, Louisiana. 2003 The Ties that Bind. In Heterarchy, Political Economy, and the Ancient Maya edited by Vernon L. Scarborough, Fred Valdez, Jr., and Nicholas Dunning, pp. 5263. The University of Arizona Press, Tucson. Houk, Brett A., Owen Ford, and Amy E. Rush 1999 Fear and Loathing in Northwest Belize. Paper presented at the 64th Annual Meeting of the Society for American Archaeology, Chicago, Illinois. King, Eleanor M. and Leslie C. Shaw 2003 A Hierarchical Approach to Site Variability. In Heterarchy, Political Economy, and the Ancient Maya, edited by Vernon L. Scarborough, Fred Valdez, Jr., and Nicholas Dunning, pp. 64-76. The University of Arizona Press, Tucson. Lewis, Brandon 2005 Research at Dos Barbaras: An Overview. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by F. Valdez, Jr., pp. 145-158. Occasional Papers, Number 4. Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin. Lincoln, Charles E. 1985 Ceramics and Ceramic Chronology. In A Consideration of the Early Classic Period in the Maya Lowlands edited by Gordon R. Willey and Peter Mathews, pp. 55-94.IMS Publication No. 10. Albany: Institute of Mesoamerican Studies, State University of New York. MeBar, Yoav and Brandon Lewis 2005 Dos Barbaras, Group B, Structure 6: 2003 to 2004 Excavations. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by F. Valdez, Jr. pp. 159-186. Occasional Papers, Number 4. Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin. Mock, Shirley 2004 Maya Traders on the North-Central Coast Belize Coast. In Research Reports in Belizean Archaeology, Vol. 1 edited by Jaime Awe, John Morris, and Sherilyne Jones, pp. 359-370. Institute of Archaeology, Belmopan, Belize. Padilla, Anthony, Molly Morgan, and Jon C. Lohse 2006 and 2005 Season Excavations at Akab Muclil, Northwestern Belize. In Final Season Summaries of the Blue Creek Regional Political Ecology Project: 2005 Investigations in Upper Northwestern Belize, edited by Jon C. Lohse, pp. 12-39. Report submitted to the Institute of Archaeology, National Institute of Culture and History, Belmopan, Belize. Pendergast, David 1992 Noblesse Oblige: The Elites of Altun Ha and Lamanai, Belize. In Mesoamerican Elites: An Archaeological Assessment edited by D. Z. Chase and A. F. Chase, pp.61-79. University of Oklahoma Press, Norman and London. Rice, Prudence M., Arthur A. Demarest, and Don S. Rice 2004 The Terminal Classic and the Classic Maya Collapse in Perspective. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice, pp. 1-11. University Press of Colorado, Boulder. Robichaux, Hubert 1995 Ancient Maya Community Patterns in Northwestern Belize: Peripheral Zone Survey at La Milpa and Dos Hombres. Unpublished Ph.D. dissertation, Department of Anthropology, The University of Texas at Austin. Sabloff, Jeremy A. 1975 Excavations at Seibal: Ceramics. Memoirs of the Peabody Museum of Archaeology and Ethnology, Vol. 13, No. 2 Harvard University, Cambridge, Massachusetts. Sabloff, Jeremy A. and Gordon R. Willey 1967 The Collapse of Maya Civilization in the Southern Lowlands: A Consideration of History and Process. Southwestern Journal of Anthropology 23(4):311-336. Shimkin, Demitri B. 145

PAGE 148

Terminal Classic Investigations in the Three Rivers Region 146 1973 Models for the Downfall: Some Ecological and Culture-Historical Considerations. In The Classic Maya Collapse edited by T. Patrick Culbert, pp. 269-300. University of New Mexico Press, Albuquerque. Smith, Robert E. 1955 Ceramic Sequence at Uaxactn. Middle American Research Institute Publication 20. New Orleans, LA: Tulane University. Sullivan, Lauren A. 2002 Evidence for Changing Dynamics in the Regional Integration of Northwestern Belize. In Ancient Maya Political Economies pp. 197-222, edited by Marilyn Masson and David Freidel. Altamira Press, Walnut Creek, CA Sullivan, Lauren A. and Kerry L. Sagebiel 2003 Changing Political Alliances in the Three Rivers Region. In Heterarchy, Political Economy, and the Ancient Maya: The Three Rivers Region of the east-central Yucatan Peninsula edited by Vernon L. Scarborough, Fred Valdez, Jr. and Nicholas P. Dunning, pp. 25-36. University of Arizona Press, Tucson. Sullivan, Lauren A. and Fred Valdez, Jr. 2004 Northwest Belize: A Regional Perspective on Cultural History. In Research Reports in Belizean Archaeology, Vol. 1 edited by Jaime Awe, John Morris, and Sherilyne Jones, pp. 185-196. Institute of Archaeology, Belmopan, Belize. Tourtellot, Gair III, and John Rose 1993 More Light on La Milpa Mapping: Interim Report on the 1993 Season. Version 2.25. La Milpa Arch aeological Project. Valds, J. A. and F. Fahsen 2004 Disaster in Sight: The Terminal Classic at Tikal and Uaxactun. In The Terminal Classic in the Maya Lowlands: Collapse, Transition, and Transformation edited by A.A. Demarest, P. M. Rice and D. S. Rice, pp. 140-161. University of Colorado Press, Boulder. Valdez, Jr., Fred and Lauren A. Sullivan 2006 The Programme for Belize Archaeological Proj ect: Recent Regional Investigations. In Research Reports in Belizean Archaeology, Vol. 3, edited by John Morris, Sherilyne Jones, Jaime Awe, and Christophe Helmke, pp. 317-328. Institute of Archaeology, Belmopan, Belize. Valdez, Jr., Fred 1989 La Decadencia de la Civilizacin Clsica Maya: Evidencias de la Tierras Bajas. Yaxkin XII (2):45-63. Walling, Stanley L., Peter F. Davis, Sandra Dias, and Melissa DeVito 2005 Report of the 2004 Rio Bravo Archaeology Project: Site R.B. 47: Chawak Butoob. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by F. Valdez, Jr. pp. 115-143. Occasional Papers, Number 4. Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin. Webster, David L. 1977 Warfare and the Evolution of Maya Civilization. In The Origins of Maya Civilization edited by R.E.W. Adams. University of New Mexico Press, Albuquerque. Willey, Gordon R. 1973 Certain Aspects of the Late Classic to Postclassic Periods in the Belize Valley. In The Classic Maya Collapse edited by T. Patrick Culbert, pp. 93-106. University of New Mexico Press, Albuquerque. Willey, Gordon R., T. Patr ick Culbert, and Richard E.W. Adams 1967 Maya Lowland Ceramics: A Report from the 1965 Guatemala City Conference. American Antiquity 32(3):289-315.

PAGE 149

13 TERMINAL CLASSIC EVENTS: COLHA, BELIZE AND THE CENTRAL MAYA LOWLANDS Fred Valdez, Jr. and Palma Buttles Terminal Classic events at the Maya site of Colha in northern Belize include the placement of a skull pit in the monumental center and an abandonment of the site followi ng this deposit. The site of Colha is contextualized and the violent activity of the Terminal Cla ssic occupation is presented. Implications of the sk ull pit and events across the Central Maya Lowlands are discussed for modeling one possible scenario of the Terminal Classic collapse. Introduction The collapse or demise of Classic Maya civilization has long been a subject of intense study and interest. We must begin by stating that the use of terms such as collapse, demise, or fall are in many ways inappropriate. These are utilized here not in any judgment of Maya civilization, but rather to denote the significant changes that exist between the expressed Classic Period manifestations and the Postclassic Maya societies. The use of any of these terms is intended to emphasize the end of a certain tr adition, way of life, and perhaps social-pol itical structure. The reasons or elements often discussed when examining the Classic Period collapse include internal and external considerations; and these are then refined to natural, economic, or sociopolitical elements (cf. Adams 1973; Sabloff 1973; Webster 2002; Demarest, Rice, and Rice 2004). Some suggest that the Classic Maya demise resulted from overpopu lation and the taxing of resources (Culbert 1973, 1977; Harrison 1977; Webster 2002). The role of external force or invasion is also considered for the collapse (Sabloff and Willey 1967; Adams 1973; Webster 2002), as is the possibility of an environmental crisis (Sanders 1962, 1963; Gunn and Adams 1988; Gill 2000). Disease has also been seen as a prime component in the Maya decline (Saul 1973). However, it seems most likely that the demise of Classic Maya civilization resulted from a combination of these factors (Sabloff 1973; Culbert 1977; Valdez 1987, 1989). With the premise that varying elements may have been at work in the Classic Maya demise, is remains critical to understand (or at least expect) that the result would have been a decline in th e different geographic areas occurring at varying rates and expressed archaeologically in multiple ways. The Southern Lowlands It is important to briefly review relevant data of the Terminal Classic Period in the southern lowland region of the Maya area (Figure 1). The Pasion, Peten, and Rio Bec areas will serve as general points of reference before di scussing the northern Belize site of Colha (Hester et. al. 1980; Figure 2) where findings bear significantly upon reconstructions of the Classic Maya collapse. Archaeological work at the sites of Altar de Sacrificios and Seibal provide the best information concerning the Classic Period collapse in the Pasion area. Both Adams (1973) and Sabloff (1973) used their ceramic analyses and other data (e.g., architecture and sculpture) for determining an invasion hypothesis of the Pasion sites by Mexicanized Maya from the Gulf Coast. An interesting observation from the Altar de Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 147-153 Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 150

Fred Valdez, Jr. &Palma Buttles 148 Figure 1 Map of the Maya Area. Sacrificios excavations is an indication of a violent end to the Classic Period as evidenced by excavated remains of large amounts of fired adobe from burned wattle and daub architecture (Adams 1973: 143). Adams (ibid.) further states: The palaces at Altar were perishable and were set on stone-faced platforms. Such platforms surrounded the court on three sides. Since underlying the burned adobe layer were Boca ceramics, a likely interpretation is that the elite-class residential architecture burned at the end of the phase. We cannot be sure whether the firing was deliberate. If we accept the interpretation given below of a military intrusion by people of Jimba culture, then the fire may well have been deliberate. From the available data, it seems that the Pasion area saw the Classic Maya collapse through the efforts of a foreign group(s). Some evidence indicates that the collapse, at Altar de Sacrificios may have been a violent termination; this is an issue that will be revived in discussing the Colha data. Terminal Classic events in the Peten region appear less dramatic that those noted for the Pasion. It seem s that the Peten sites may have declined (or collapsed) without external force. This certainly seems to be the case as understood for Tikal where a delicate balance between ecological, social, and political situations must have been maintained at all times. In his analysis of the Tikal data, Culbert (1973: 91-92) concludes: I must, then, take a stand as a strong proponent of an internal mechanism for the Maya collapse. Tikal succumbs too early and too rapidly for the trigger to have been external, and foreign in cursions in parts of the Maya Lowlands seem to me to be the result rather than the cause of the Maya collapse. The Peten, using Tikal as an example, most likely fell as a result of internal events. This interpretation will be echoed in the discussion of Colhas Classic Period demise. Figure 2 Map of Northern Belize. The Rio Bec area also provides information from the Terminal Classic Period. Influences in architecture and ceramics are credited for both north and 148

PAGE 151

Terminal Classic Investigations at Colha, Belize 149 Figure 3 Colha Operation 2011 (from Eaton 1980). south indicating that th is region was actively trading in both directions In his analysis of Terminal Classic lithics, Rovner (1974) made a curious observation that relates directly to finds at Altar de Sacrificios and Colha. Specifically, Rovner states: Over 80 percent of more than 600 lithic artifacts and debitage of this phase from the interior room s of Structure IV at Becan show marked smoke blackening, heat discoloration, fire crack ing, and spalls. It would appear that the inhabitants of these rooms were literally bur ned out of Structure IV. Lithic evidence thus suggests that the Classic Maya met a violent end at Becan (ibid: 130). Becan may be placed with Altar de Sacrificios in terms of witnessing at least one structure burned in the Terminal Classic. Both sites provide evidence for violent activity, although the source of this action remains uncertain, i.e., at Altar de Sacrificios the destruction and burning of (elite) residences ma y have resulted from invasion or from in ternal strife. The Terminal Classic burning of Structure IV at Becan is also without direct evidence for external or internal instigation. At both sites, however, the burning is associated with an elite structure(s). Figure 4. Skull Pit Illustration (after Steele et. al. 1980). Colha, Belize The 1980 field season at Colha produced a remarkable find within the main plaza of the monumental center. A large linear mound on the south side of the plaza,

PAGE 152

Fred Valdez, Jr. &Palma Buttles 150 believed to have been an e lite residential structure, was partially excavated. The excavation (Operation 2011) uncovered a skull pit, dating to the Terminal Classic, in a corner area between the central staircase and structure wall (Eaton 19 80; Figure 3). The wall above the skull pit is constructed of limestone blocks that were shattered by fire. The destruction of the limestone wall probably resulted from the burning of the elite residential house; a perishable structure that may have fallen in front of the wall and over the skull pit. The skull pit (Figure 4) contained decapitated skulls of 30 individuals in two layers. The populations represented in the remains were ten children, six months to seven years of age, a nd 20 adults ranging in age from young adults to old adults (Massey 1986). The skull pit remains have been analyzed in detail for pathologies, cultural modifications, and post-mortem treatment (ibid.). A brief review of the analysis will be presented before discussing the skull pit and its position in un derstanding certain Terminal Classic events. All technical information provided below is derived from Masseys (ibid.) analysis. Both children and adult teeth are known to have a number of pathologies including: calculus deposits, caries, ante mortem tooth loss, enamel hypoplasia, alveolar abscesses, a nd periodontal disease. The most common dental pathology observed for adults (19 of the 20) is calculus deposits followed by severe caries that affected 15 individuals. Eleven had lost teeth prior to death and seven of the adults showed markings of linear enamel hypoplasia. Dental caries are the most common problem noticed for the childrens teeth. Other dental pathologies in the children included linear enamel hypoplasia and unusual notching of the incisors. Bone pathologies in the adult population included infection of the bone, joint deformation, cribra orbitalis, and an unusual roughness of the palate, while the children were found to be limited to cribra orbitalis (ibid.). Cultu re modifications to bone and teeth are presen t in the Colha skull pit sample. Both cranial shaping (at least eight individuals) and teeth filing (nine adults) are present. Perhaps the most intriguing feature of the skeletal material is the presence of numerous cut marks. Twenty of the 30 skulls showed clear-cut marks; the other ten skulls are too damaged or covered with encrusted residue to be evaluated adequately. The average number of cut marks (per skull) in frontal view is 17. Figure 5 shows the composite of cut marks for each view. Most of the cuts are presumed to be the result of flaying. However, some cuts ar e indicative of efforts required in removing mu scle or other soft tissue including the eyes and tongue. Deep cuts were also observed on two of the cervical vertebrae representing unsuccessful attempts at decapitation. Figure 5. Distribution of Cut Marks (composite) on Skulls (after Massey 1986). 150

PAGE 153

Terminal Classic Investigations at Colha, Belize Charring of the bone is present on three skulls. One skull is badly burned while two are s lightly charred. All three were excavated from the lower level of the skull pit, indicating that the burning occurred prior to interment. In sum, the Colha skull pit represents a population of 30 individuals including adult men, adult women, and young children. Children and teenagers between seven and eighteen years of age are not represented in the sample, therefore, negating the probability that the skull pit contained the remains of a nuclear family. While numerous dental and bone pathologies have been noted, the analysis does not indicate widespread serious disease or malnutrition (ibi d.). However, the presence of these diseases (especially enamel hypoplasia) indicates at least periodic nutritional stress of significant degree. The pit of decapitated skulls and the burned elite structure at Colha provide some of the strongest evidence for a violent end to the Lowland Classic Maya. The site of Colha is briefly (50 to 100 years), abandoned followed by the Postclassic reoccupation by a northern Maya group (Valdez 1987). Other Considerations Environmental research by Gill (2000) has found a direct correlation between northern European climatic fluctuations and rainfall on the Yucatan Peninsula. Available data suggests that the Terminal Classic Period saw a serious drought that was quite devastating to the region (Hodell, Curtis, and Brenner 1995; Gill 2000). In a system where resources are taxed and population is high, a significant drought would add tremendous stress, encourage disease, and spark dissention. Other studies (Valdez and Mock 1991) have found marginal areas to be occupied during the Terminal Classic. For example, the Northern River Lagoon Site (NRL) is occupied primarily in the Terminal Classic Period. A primary purpose for its founding was probably related to salt making. The salt was then used in preparing fish and other marine resources for transport into the interior lowlands. With the demise of Classic Maya society, the NRL is abandoned. Summary and Conclusions In this review of the Terminal Classic Period, selected sites have been studied for relevant findings concerning the Classic Maya collapse. The osteological data (Massey 1986), the climatic information (Gill 2000), the interpretation of the NRL sites function (Valdez and Mock 1991), all indicate hard times in the lowland area. The archaeological data of burned structures at Altar de Sacrificios, Becan, and Colha, in addition to the skull pit indicate a violent end to the Classic Period. It is interesting that the burned structure(s) at each site are elite-asso ciated residence(s). Other sites not discussed here, but containing evidence for burned elite structures include Palenque, Tonina, Dos Pilas, and Copan. The Colha skull pit seems to be representative of an elite group given the cultural modifications observed on the skulls and the location of the pit in front of the residential structure. If the skull pit does represent an elite gr oup, one can postulate that the mass of the Colha population was probably in poorer physical condition. Disease among the commoner population must have been widespread and more serious as compared to the elite population who likely had better access to food, attendance, etc. Although much of the provided data serve little more that as circumstantial evidence for understanding the Classic Maya demise, we will propose a scenario 151

PAGE 154

Fred Valdez, Jr. &Palma Buttles 152 that seem s most consistent with the available information. First, there is a large Classic Period population that is pushing productivity to its limits. Second, a significant drought cut productivity through crop failures, etc. The drop in available water and food lead to malnutrition and disease; as may be evidenced in the Colha skull pit and as has be en reported for other sites (Saul 1973). Third, the decline in food production forces colonies into marginal areas for procuring supplemental foodstuffs as seems to be the case for the NRL site. Fourth, if the skull pi t represents an elite group, the general Maya population must have been plagued with disease and hunger. Such a situation would become intolerable and probably lead to an internal uprising. Any social and political problems would simply be added to the implied dire physical conditions (cf. Hamblin and Pitcher 1980). This scenario also incorporates why violent Terminal Classic activity occurs at elite-residential structures, i.e., this is an act of literally removing an ineffectual authority. At Tikal there is presently no evidence for this kind of violent end. However, the various conditions that seem to surround Tikal undoubtedly contributed to its demise as well. We are certainly in line with Culberts (1973) analysis that the Maya decline, at least at Tikal, were from internal mechanisms and varied from site to site. Where external forces appear, they seem to be an after-the-fact occurrence. These are Mexicanized Maya or some other opportunistic group that moves in on an already self-defeated Classic Lowland Maya. A final note concerns those sites or areas that are not abandoned after the Terminal Classic. The areas that continue in occupation after the demise are in close proximity to fresh water. For example, the Peten Lakes sites in Guatemala, Lamanai in Belize (Pendergast 1981), and Copan in Honduras (Webster and Gonlin 1988). It appears that the drought (Gill 2000) brought a significant blow to an already fragile society. Occupation at water safe sites most likely continued with a reduced population where the requirements of society were less demanding, at least for a while. The Postclassic Maya societies that followed these dramatic Terminal Classic events are clearly politically an d perhaps socially structured in a different order. As Webster recently noted, as it always does in the long run, the natural order prevailed, and the moral order changed. References Cited Adams, R. E. W. 1973 Maya Collapse: Transformation and Termination in the Ceramic Sequence at Altar de Sacrificios. In The Classic Maya Collapse edited by T.P. Culbert, pp.133163. University of New Mexico Press, Albuquerque. Culbert, T. P. 1974 1973 The Maya Downfall at Tikal. In The Classic Maya Collapse edited by T.P. Culbert, pp.63-93. University of New Mexico Press, Albuquerque. 1977 Maya Development and Collapse: An Economic Perspective. In Social Process in Maya Prehistory, edited by N. Hammond, pp.510-530. Academic Press. New York. Demarst, A. A., P. M. Rice, and D. S. Rice 2004 Terminal Classic in the Maya Lowlands University Press of Colorado. Boulder. Eaton, J. D. 1980 Operation 2011: Investigations within the Main Plaza of the Mo numental Center at Colha. In The Colha Project Second Season, 1980 Interim Report edited by Thomas R. Hester, Jack D. Eaton, and Harry J. Shafer., pp. 145-161. Center for Archaeological Research. The University of Texas at San Antonio and Centro Studi e Ricerche Ligabue, Venezia. 152

PAGE 155

Terminal Classic Investigations at Colha, Belize 153 Gill, R. B. 2002 The Great Maya Droughts University of New Mexico Press. Albuquerque. Gunn, J. and R. E. W. Adams 2003 Climatic Change, Culture, and Civilization in North America. World Archaeology 13(1): 87-100 Hodell, D. A., J. H. Curtis, M. Brenner 1995 Possible Role of Climate in the Collapse of Classic Maya Civilization. Nature 375:391-394. Hamblin, R. L. and B. L. Picher 1980 The Classic Maya Collapse: Testing Class Conflict Hypotheses. American Antiquity 45(2): 246-267. Harrison, P. D. 1977 The Rise of the Bajos and Fall of the Maya. In Social Process in Maya Prehistory, edited by N. Hammond. Academic Press, New York Hester, T.R., G. Ligabue, H.J. Shafer, J.D. Eaton, and R.E.W. Adams 1980 The 1980 Season at Colha, Belize: An Overview. In The Colha Project Second Season, 1980 Interim Report, edited by T.R. Hester, J.D. Eaton, and H.J. Shafer, pp. 114. Center for Archaeological Research. The University of Texas at San Antonio and Centro Studi e Ricerche Ligabue, Venice. Massey, V. K. 1986 The Human Osteology of a Terminal Classic Maya Skull Cache at Colha, Belize Masters Thesis, Department of Anthropology, Texas A & M University. Pendergast, D. M. 1981 Lamanai, Belize: Summary on Excavation Results, 1974 1980. Journal of Field Archaeology 8:29-53. Sabloff, J. A. 1973 Continuity and Disruption during Terminal Late Classic, In The Classic Maya Collapse edited by T.P. Culbert, pp.133163. University of New Mexico Press, Albuquerque. Sabloff, J. A. and G. R. Willey 1967 Collapse of Maya Civilization in the Southern Lowlands: A Consideration of History and Process. Southwest Journal of Anthropology 23(4):311-336. Sanders, W. T. 1962 Cultural Ecology of the Maya Lowlands, Part I. Estudios de Cultura Maya 2:79-121. 1963 Cultural Ecology of the Maya Lowlands, Part II. Estudios de Cultura Maya 3:203-241. Saul, F. P. 1 973 Disease in the Maya Area: The PreColumbian Evidence. In The Classic Maya Collapse, edited by T.P. Culbert, pp.301324. University of New Mexico Press, Albuquerque. Valdez, F., Jr. 1987 The Prehistoric Ceramics of Colha, Northern Belize Ph.D. Dissertation, Department of Anthropology, Harvard University, Cambridge, Massachusetts. 1988 La Decadencia de la Civilizacion Maya Classica: Evidencies de las Tierras Bajas Centrales. Yaxkin XII (2): 45-63. Valdez, F. Jr. and S. B. Mock 1991 Additional Consideration For Prehistoric Salt-making in Belize. American Antiquity 56(3):520-525. Webster, D. 2002 The Fall of the Ancient Maya Thames and Hudson. New York. Webster, D. and N. Gonlin 1987 Household Remains of the Humblest Maya Journal of Field Archaeology 15(2):169-190.

PAGE 156

14 FACTORS OF STRESS AND PROCESSES LEADING TO THE END OF THE CLASSIC PERIOD AT BLUE CREEK Thomas H. Guderjan and C. Colleen Hanratty Archaeological investigations at Blue Creek, Belize has produced a rich database regarding the events and processes leading to its abandonment and offers special opportunities for the investigation of the processes leading to the end of the Classic Maya period. In this paper, we discuss the stressful dynamics of Late Classic period, including rapidly increasing the processes leading to the end of the Classic Maya period. In this paper, we discuss the stressful dynamics of Late Classic period, including rapidly increasing population, declining soil quality, and expansion of agricultural systems to their maximum. We then discuss the specific responses that are documented for the Terminal Classic period at Blue Creek and how the remnant populations of the Early Postclassic made use of the Classic period agricultural infrastructure. Introduction There has never be en a lack of theory and speculation about the demise of complex Maya society at the end of the Classic period in the southern lowlands (i. e., Culbert 1973; Demarest, Rice, and Rice 2005). However, powered by the very popular volumes Guns, Germs, and Steel and Collapse! By Jared Diamond (1997, 2005), the overly simplistic environmental view has resurfaced with new strength. Partially in response to this, archaeologists are responding by assembling new and robust data regarding the end of the Maya Classic period, such as the 2006 Belize symposium. Along with our many colleagues, we have been investigating the site of Blue Creek in northwestern Belize for 15 years. This has resulted in a large database relevant to understanding the natu re of a Maya city (Guderjan 2007) as well as other issues. The work at Blue Creek has also produced a rich database regarding the events and processes leading to its abandonment and offers special opportunities for the investigation of the processes leading to the end of the Classic Maya period. In this paper, we discuss the stressful dynamics of Late Classic period, including rapidly increasing population, declining soil quality, and expansion of agricultural systems to their maximum. We then discuss the specific responses that are documented for the Terminal Classic period at Blue Creek and how the remnant populations of the Early Postclassic made use of the Classic period agricultural infrastructure. Blue Creek in the Early Classic. Before examining the Late and Terminal Classic, it is useful to review what is known of Blue Creek before then. In the Early Classic period, Bl ue Creek consisted of an inter-connected set of residential components that were separated by vast tracks of agricultura l lands (Guderjan 2005, 2007; Guderjan, Baker and Lichtenstein 2003). Each of these residential components had their own internal structures of power, legitimacy and authority that seem to have consisted of multi-generational lineages (Guderjan 2007; Guderjan, Diel, Giacometti, and Andrews in press; Guderjan and Hanratty 2006). These lineages interacted on social, economic and po litical levels with each other and the royal lineage that resided in the sites central precinct. Their power was apparently derived from control over critical resources. In th e case of Blue Creek, Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 155-165. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 157

Thomas H. Guderjan & C. Colleen Hanratty these resources were ag ricultural lands that covered at least 50 % of the landscape (Figure 1). In each of their residential complexes there were also central places, shrines, or terminus groups (Audet and Awe 2004; Cheetham 2004; Giacometti 2002). However, there were other residential components and multi-generational lineages that never achieved high status in the community because they did not control the agricultural lands adjacent to their homes. They were, in some way, in the employ of the land-holding elite. Figure 1 Map of central 20 square kilometers of Blue Creek The central precinct sits atop the 100 meter tall, north-south running Bravo Escarpment. Above and west of the escarpment is a hill and bajo upland zone. Residences are found on the tops and flanks of these hills and the small bajos between them exhibit virtually no residential activity as they were used for agriculture. Below and east of the escarpment are large areas that were used for intensive agriculture with water managed by a large system of ditches covering many square kilometers? Augmenting the presence of these high quality agricultural lands was Blue Creeks setting at the headwaters of the Ro Hondo (Barrett and Guderjan 2006). While the Ro Hondo today forms the border between Belize and Mexico, in the past it was a conduit rather than a border. The Ro Hondo is the northernmost river draining into the Caribbean Sea and provided canoe access from the coastal trade routes to the interior of the Petn region. For Blue Creek, the Ro Hondo provided access to distant markets and a route to export its considerable agricultural produce. As a consequence, Blue Creek became a wealthy city. Large quantities of imported artifacts such as jade are found throughout Blue Creek from the most humble of perishable st ructures at the base of the escarpment to the most sacred structures within the site core (Guderjan 1998, 2005, 2007). In addition to prestige items, approximately 40% of the utilitarian lithics were also imported most likely from Colha despite the local availability of resources (Barrett 2004, 2006). From the Guatemala highlands, Blue Creek also imported large quantities of obsidian (Haines 2000). Material goods were not the only items to make their way down river innovative forms of architecture and ideologies based upon divine kingship found their way from the banks of the Rio Hondo to the rulers of Blue Creek. The rulers of Blue Creek were legitimized by the presence of innovative and sy mbolically charged forms of architecture. The plan of the central precinct itself is an ex pression of a regional orderseveral sites with similar plans exist along the Bravo Escarpment including Dos Hombres (Houk 1996), Quam Hill (Guderjan 1991), Punta de Cacao (Robichaux 2002), and San Jose (Thompson 1939). Also, components of the central precinct incorporated concepts of ideology and power. Plaza A was surrounded by ritual 156

PAGE 158

Investigations at Blue Creek, NW Belize and m ultifunctional structures on three sides leaving an open view of the agricultural fields and settlements a hundred meters lower to the east. This plaza is dominated by Structure 1, the earliest documented columnated superstructure in the Maya lowlands (Driver 2002, Guderjan 2004; 2007). On the opposite side of the plaza, Structure 4 incorporated an axis mundi tethering the underworld to the cosmos (Guderjan 2004, 2005, 2007). Immediately north of Plaza A was a large platform on which an Early Classic ballcourt one of the earliest in northwestern Belize (Guderjan 2004, 2007). Plaza B is dominated by the royal palace and is flanked to the north and south by temple pyramids. The most notable is Structure 9, a typical Pe tn style temple with a single room superstr ucture positioned atop a steeply inclined substr ucture with a central staircase and stair-side outsets. The facade of Structure 9s outset near its summit was adorned with a five paneled, medium relief stucco frieze depicting a set of ahaw images. (Grube, Guderjan, Haines 1995). The end of the Early Classic was marked by a major caching event in which nearly 1000 jade artifact s were buried in the axis mundi shaft of Structure 4 (Guderjan 1998, 2004, 2007) and the razing of the columned superstructure of Structure 1 to accommodate a royal tomb (Driver 2002; Guderjan 2004, 2007). These may be coincidental with other transformations or they may mark the end of power for the ruling lineage. Regardless which is the case, the presence of this suite of architectural, ritual and burial elements reinforce the idea that Blue Creek was an autonomous city with its own local ruling lineage. Further, Blue Creek was participating in the forefront of regional ideology while still conforming to regional patterning of major centers of power. Blue Creeks wealth was further expressed by its access to large quantities of exotic goods such as jade. In short, Blue Creek was a small, wealthy city whose underlying economic base was its agricultural productivity and its relationship to the larger system of inter-polity trade. The Late Classic Blue Creeks fortunes changed significantly during the sixth century with the fall of Tikal and the subsequent restructuring of trade networks. Jade and other imports ceased to flow into the community. There was also a sharp decline in imported lithic materials during the Late Classic. Colha chert for example, accounted for 40% of the utilitarian lithics during the Early Classic, but only 20% during the Late Classic (Barrett 2006). Similar trends are reflected in the amount of basalt, granite, obsidian, and marine shell imported into Blue Creek during this time period. The Late Classic political reorganization is also reflected in the construction activities within the site core. For instance, during the early part of the Late Classic, Structures 2 and 3 were constructed in the form of a pseudo-E-Group that enclosed the east side of Plaza A (Guderjan 2006). Also, several buildings were re-oriented and buried under new, flat-topped structures more in similar to those seen in the northern coastal plain of Belize (Guderjan 2007). Nonetheless, Blue Creek remained an active and vibrant community during the Late Classic as construction and modification of residences for the elite members of the community grew at an unparalleled rate in both the central precinct and other elite residential zones (Guderjan, Lichtenstein and Hanr atty 2003; Guderjan and Hanratty 2006; Hanratty 2002). Kn Tan is a large residential complex consisting of 9 elite residences (Structures 37, 41, 42, 43, 45, 46, 81, 82, and 83 complexes) on the tops of prominent hills. During the Early Classic, there was 157

PAGE 159

Thomas H. Guderjan & C. Colleen Hanratty only m odest construction but most complexes were ritually charged. For instance, several ancestral shrines have been identified. One, Structure 34, included an important tomb from which a royal jewel in the form of a jade acrobat pendant was recovered (Guderjan and Hanratty 2006). This individual was the founder of a lineage that would reside around his tomb for 600 years. As time went on, they built larger and larger residences and left evidence of increasing wealth and authority. Like the residents of Rosita, their initial authority was probably based upon control of the critical resource, agricultural lands that surrounded their residences (Guderjan, Lichtenstein and Hanr atty 2003; Hanratty 2002). Relevant Processes during the Late Classic Period Several major processes can be documented during the Late Classic and in each case; the situation at Blue Creek is a microcosm of larger regional processes. The first three are all in the realm of cultural ecology or man-land relationships, or more specifically the interactions between the residents of Blue Creek and their environment. However, we are not building an argument of environmental causation for the Maya collapse. The environmental factors must be contra sted with those of human agency. There were also clearly factors beyond the scope of environmental determinism at play. The first process of interest is the significant increase in population during that Late Classic. This has been broadly documented and discussed for the southern lowlands but is still difficult to quantify. In some studies, this has probably been exaggerated due to methodological issues with relevant ceramic analyses. Specifically, ceramic attributes of the Late Preclassic and Early Classic periods have been mistakenly believed to cease at the end of the Late Preclassic. Since the temporal aspect of population studies has been based upon ceramic chronologies, this has the effect of overestimating the Late Preclassic population sizes while underestimating the size of Early Classic populations. This methodological fault lends false credence to a population decline during the hiatus and gives the appearance that Late Classic populations were increasi ng at an unrealistic rate. Nonetheless, Late Classic populations did increase considerably in size over the Early Classic scale. At Blue Creek, already established residential areas such as Kin Tan grew to their maximum size. In all cases these were located in proximity to high quality agricultural lands and had clearly defined internal hierarchies with shrines and other central places within each residential component. However, in the Late Classic, new residential components such as U Xulil Beh (Figures 1 & 2) were founded in marginal agricultural settings. Figure 2. Map of U Xulil Beh U Xulil Beh consisted of at least 20 residences located on a flat ridge surrounded on three sides by deeply down cut drainages. There was no central place, shrine, or residential structure that was more prominent than the others. Test excavations showed either only Late Classic or more 158

PAGE 160

Investigations at Blue Creek, NW Belize am biguous Classic dates and no evidence of significant material wealth. U Xulil Beh then, was most likely a Late Classic expansion community. Further, U Xulil Beh did not have access or a pparent control over the rich agricultural lands than underlay the power and authority of other residential areas. Instead, they built several broad terraces on the western margin of the groupalmost certainly for agricultural purposes. These were likely communally held and could not have produced a significant surplus of products to go into the larger market economy. Figure 3 Structure 50 deposit During the Late Classic period there was also maximization of intensive agriculture. It is easy to monitor the scale of the large ditched agricultural fields that had probably already been in place for several hundred years (Dunning, Beach, and Luzzadder-Beach 2007). However, it is possible, if not likely, that they were expanded laterally in the Late Classic to their maximum extent. Jeff Bakers early excavations of these ditches may have located a setting where significant fill had been carried in to fill low-lying areas. This then may have been di tched to connect to the extant system and to control soil moisture (Baker 2001, 2002). Further, growing platforms were built at the base of the escarpment near the ditched fields to expand the amount of productive land (Baker 2001, 2002; Lichtenstein 2000). While these are relatively small, only 48 square meters in one case, they represent a significant amount of construction effort and energetic and yielded only modest increased production. We can safely assume that they would not have been built if the highly productive soils immediately adjacent to them were not already under cultivation. West and above the escarpment, the bajos separating residential components had also probably been cultivated for hundreds of years. In the Late Classic, a situation similar to the construction of growing platforms adjacent to the ditched fields can be seen. Terracing and cross-drainage features were built with considerable labor to expand the amount of arable land by only very small amounts. It could be argued that these expansions were necessary to feed an expanding population. However, the amount of agricultural land within Blue Creek at least 40% of the areawas already more than adequate to grow food One factor in the need for expansion was the likelihood that significant productivity was being lost to soil erosion and the declining quality of soils. Another factor was most likely the desire to maximize production for commercial purposes. Mitigating the impact of this expansion of agricultural lands was the third process of significant so il erosion. At Blue Creek we see rapid eros ion as a consequence of every new clearing. Further, rapid prehistoric alluvial events have been documented at the base of the Bravo Escarpment dating to approximately A. D. 1000 (Dunning, Beach, and LuzzadderBeach 2007). Such events were certainly the result of equally rapid erosional events at higher elevations. Binford, Rice and their colleagues have documented rapid hillside erosion into the Petn lakes at the end of the Classic period that not only removed the topsoil but subsoils as well (Binford, et al, 159

PAGE 161

Thomas H. Guderjan & C. Colleen Hanratty 1987). So, despite the m aximization of agricultural lands, it is very likely that the end of the Classic was also marked by declining agricultural productivity. Figure 4 Photo of Structure RS-21, Terminal Classic Yucatecan shrine In general, then, we see evidence of increasing population coupled with increasing demands being placed upon the agricultural production systems that clearly point to increasing stress during this time. However, before our argument is misconstrued as an environmental deterministic approach, we must point out that other processes were also ongoing. There may have been a decline in the power of central authority at Blue Creek. More construction occurred in the elite residences outside of the central precinct than within it during the Late Classic. A small Early Classic plaza was converted to residential use. The ballcourt may have been terminated and in disuse. The graceful columned superstructure of Structure 1 was razed and a last royal tomb built. Elsewhere, one of us has argued that these events were the consequence of a political takeover of Blue Creek by another polity (Guderjan 1998). While this is far from certain, it is clear that significant restructuring occurred at the end of the Early Classic. Our point here is that we view the responses of human agencies to environmental challenges as being differential and important to our understanding of the effects of these challenges. Further, we do not fully understand the structure of decision making, response and organization of the Late Classic. The Abandonment of Blue Creek. At Blue Creek, we have been fortunate in that several terminal ceramic deposits have been found that date the final actions in several locations. As a result, we are now being able to track the progressive sequence of abandonment of the site. This is a sequence marked by th e breakdown of old power and authority structures, rapid depopulation and re-organization of the surviving people. The first locations to be abandoned were the central preci nct and nearly elite residences. Large quantities of ceramics, mostly Achote Black, were deposited in front of Structure 3 on Plaza A. This event has been variously interpreted as the result of a termination ritual (Guderjan 2004) or the redeposition of ceramics from a feasting event (Clayton, Driver and Kosakowsky 2005). However, the most important point is that the deposit is certainly was the final activity at that location. At about the same time, large amounts of portable goods were deposited against the baselines of buildings at Kn Tan. These deposits were placed against the groups most sacred ancestor shrines (Structures 34 and 37) in the Structure 37 Plazuela and we continue to interpret these deposits as the results of termination rituals. The artifact assemblage included more than 14,000 sherds, 428 pieces of lithic debitage and 203 Special Finds were recovered. Of the Special Finds, 36 % were chert biface fragments, 35% were obsidian blades, 8% were chert cores, and 7% were metate fragments. The remaining 32% consisted of items such as chert eccentrics, ceramic 160

PAGE 162

Investigations at Blue Creek, NW Belize appliqus, figurines, beads, stamps, and whistles ; mano fragments, armatures, spindle whorls, hammerstones, projectile point knives, scrapers and counterweights. These events are similar to the pattern of ritual residential termination noted at Yaxunah (Suhler and Freidel 1994), Blackman Eddy (Garber et al. 1998), Dos Hombres (Houk 1996), and Floral Park (Glassman et al. 1995). No further construction activities or evidence of occupation are found within the central precinct or Kn Tan both were abandoned (Guderjan, Lichtenstein and Hanratty 2003). Figure 5 Perspective drawing of Rosita Structures R6 and R21. We have recently excavated another such deposit (Figure 3), at the base of Structure 50. While this deposit has not yet been analyzed, its co mposition superficially compares more closely to those in the Structure 37 Plazuela than the one at Structure 3 and certainl y dates to the same time period as both. Structure 50 is another central shrine at a residential complex in Kn Tan that is located midway between the Structure 37 Plazuela and the Central Precinct. Given the te mporal similarities among the deposits at Structures 3, 34, 37 and 50) it appears that abandonment of the central precinct occurred rapidly and before the onset of the Terminal Classic period. Some communities continued to exist, however. The Rosita community is approximately 3.5 kms northwest of the central precinct. Rosita consists of at least twenty structures built from Late Preclassic through o the Classic periods (Preston 2007). In one group, a residential room was razed and replaced with a Yucatecan style shrine (Figures 4 and 5: also see HarrisonBuck and McAnany 2005). However, this was a makeshift construction that also involved stripping the masonry faade from the major building in the group (Preston 2007). At approximately the same time, a set of rooms was added to the summit of Structure 6, the other existing ritual building in the group. Two caches were Daylight Orange: Darknight Variety vessels were cached beneath the floor of this superstructure. (Fi gure 6). So, while population sizes were rapidly falling and central authority was crumbling at Blue Creek, the activity at Rosita indicates that residual populations were aligning themselves to external forces. Figure 6. Terminal Classic cache locations on summit of Rosita Structure 5 The Early Postclassic. Only recently have we found any evidence of Postclassic occupation. At the Rempel Group (or Acab Muclil), previously known to have been constructed during the Early Classic, new, small platforms were constructed and numerous burials were 161

PAGE 163

Thomas H. Guderjan & C. Colleen Hanratty placed in an d around the small monumental architecture (Padilla and Morgan 2006). Importantly, Aclab Muclil is located below the Bravo Escarpment, adjacent to the Ro Bravo and was virtually surrounded by Classic period ditched agricultural fields. This pattern follows the known reorientation of residual populations to defensive such as islands and other settings that could utilize remaining riverine trade networks. The Rempel Group or Acab Muclil was an outlying central place during the Early Classic period. However, it was the central place for the residual Early Postclassic population. Figure 7. Daylight Orange, Darknight Variety Vessels from Structure RS-5 caches These people may have exploited remnant orchards, perhaps of cacao or other tree crops that would have continued with minimal human intervention, in the nearby ditched field systems. We have yet made a strong determination of what was being grown in these fields, but Postclassic Maya behavior gives us some indications. By the Early Postclassic, approximately A.D. 1000, the ditches themselves were largely infilled. Rather than being access routes for small canoes, the ditches would have become muddy impediments for access to the fields. We now know, though, that the fields were still used during the Early Postclassic and that at least one pole and thatch field house was built adjacent to an intersection of two ditches (Guderjan, Beach, Luzzader-Beach and Preston 2007). This strongly indicates that plants such as maize, that require human intervention for reproduction, were not being grown in the fields as there would be no purpose in such remote field houses. Instead, it is likely that some sort of tree crop was being grown. Such a crop could continue to produce economically useful products 100 or 150 years after the general abandonment of the area. Such a crop could be cacao which was known historically to have come in large quantity from the Ro Hondo area through the city of Chetumal into larger trade netw orks during colonial times. Summary Population size and agricultural activ ity at Blue Creek peaked during the Late Classic period. Further, agriculture almost certainly produ ced export goods that continued economically reinforce the authority of Blue Creeks elite. However, the use of marginal lands and the investment of great energy into minimal expansion of agricultural lands indicate stress on the system. This stress may have been caused by increasing population size, declining soil productivity or a desire to maximize production for commercial purposes. Certainly by the Terminal Classic, the structures of authority and power were rapidly disintegrating and residual 162

PAGE 164

Investigations at Blue Creek, NW Belize populations were looking elsew here for validation of authority and perhaps to support their very exis tence. By the Early Postclassic, the last residual populations retained earlier values and probably survived by harvesting agriculture goods that still grew due to the efforts of their ancestors. But, they were under such pressure that defensive structures were needed. At sometime around A D 100, they too were gone. Acknowledgements We offer our many thanks and appreciation to our colleagues at Blue Creek and elsewhere who have contributed so much to this work and to the people of todays Blue Creek who have made our efforts possible. We also thank the staff of the Institute of Archaeology, especially Dr. John Mo rris, for organizing the Belize symposia and providing such a positive forum for interaction among archaeologists in Belize. References Cited Audet, Claudia and Jaime Awe 2004 The Functions of Terminus Groups at Baking Pot. Paper pres ented at the annual meeting of the Society for American Archaeology. Baker, Jeffrey 2001 Maya Wetlands: Ecology and Prehispanic Utilization of Wetlands in Northwestern Belize. Ph.D. dissertation, Department of Anthropology, University of Arizona, Tucson. 2003 Maya Wetlands: Ecology and Prehispanic Utilization of Wetlands in Northwestern Belize Ph.D Dissertation, Anthropology Department, University of Arizona. Tucson, Arizona. Barrett, Jason W. 2004 Creating Hierarchy through Entitlements: Inequality in Lithic Resource Access among the ancient Maya of Blue Creek, Belize. Ph.D. dissertation, Department of Anthropology, Texas A & M University, College Station. 2006 Rethinking Long-Distance Exchange and the Economic Interdependence of Maya Sites during the Late Preclassic Period: The View from Northwestern Belize. Archaeological investigations in the Eastern Maya lowlands: Edited by John Morris, Sherlyne Jones, Jaime Awe, and Christopher Helmke. NICH. Barrett, Jason W. and Thomas Guderjan 2006 An Ancient Maya River Dock and Dam at Blue Creek, Belize. Latin American Antiquity 17:2: 227-239 Cheetham, David 2004 The role of Terminus Groups in Lowland Maya Site Planning: An Example from Cahal Pech. The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber. University Press of Florida, Gainesville. Clayton, Sarah, David Driver, and Laura Kosakowsky 2005 "Rubbish or Ritual? Contextualizing a Terminal Classic Problematical Deposit at Blue Creek, Belize: A Response to "Public Architecture, Ritual and Temporal Dynamics at the Maya Center of Blue Creek" by Thomas Guderjan. Ancient Mesoamerica 16:1:119-131. Culbert, T. Patrick 1973 The Classic Maya Collapse University of New Mexico Press, Albuquerque. Demarest, Arthur, Prudence Rice and Don Rice 2004 The Terminal Classic in the Maya lowlands University Press of Colorado, Boulder. Diamond, Jared 1997 Guns, Germs and Steel W.W. Norton, New York. 2005 Collapse: How Societies Choose to Fail or Succeed W. W. Norton; New York. Driver, David 2002 An Early Classic Columned Temple at Blue Creek, Belize. Latin American Antiquity 13:63-85 163

PAGE 165

Thomas H. Guderjan & C. Colleen Hanratty Garber, James F., W. David Driver, Lauren Sullivan and David M. Glassman 1998 Bloody Bowls and Broken Pots: The Li fe, Death and Rebirth of a Maya House. in The Sowing and the Dawning: Dedication and Termination Ritual Events in the Archaeology and Ethnology of Mesoamerica edited by Shirley Mock, pages 125-134. University of New Mexico Press, Albuquerque. Giacometti, Antoine 2002 The Role of Plaza Complexes in Hinterland Settlements at the Maya Site of Blue Creek, Belize. M.A. Thesis, Institute of Archaeology, University College London. Grube, Nikolai, Thomas Guderjan, and Helen R. Haines 1995 Late Classic Architecture and Iconography at the Blue Creek Ruin, Belize. Mexicon 17:3:51-57. Guderjan, Thomas 1991 Maya Settlement in Northwestern Belize: The 1988 and 1990 Seasons of the Rio Bravo Archaeological Project. Labyrinthos, Culver City, CA. 1998 The Blue Creek Jade Cache: Early Classic Maya Ritual and Architecture. The Sowing and the Dawning: Dedication and Termination Ritual Events in the Archaeology and Ethnology of Mesoamerica pages 101-112 Edited by Shirley Mock. University of New Mexico Press, Albuquerque. 2004 Public Architecture, Ritual and Temporal Dynamics at the Maya Center of Blue Creek. Ancient Mesoamerica 15:2:117. 2005 The Early Classic at Blue Creek In Research Reports in Belizean Archaeology Vol III pages 131142. 2006 E-Groups, Pseudo E-Groups, and the Development of the Classic Maya Identity in the Eastern Petn. Ancient Mesoamerica 17: 1: 1-8. 2007 The Nature of an Ancient Maya City: Resources, Interaction and Power at Blue Creek, Belize. University of Alabama Press, Tuscaloosa. Guderjan, Thomas, Jeffery Baker, and Robert J. Lichtenstein 2003 Environmental and Cultural Diversity at Blue Creek. Heterarchy, Political Economy and the Ancient Maya pages 77-91 Edited by Vernon Scarborough, Fred Valdez, Jr. and Nicholas Dunning. University of New Mexico Press, Albuquerque. Guderjan, Thomas H., Timothy Beach, Sheryl Luzzader-Beach, and Tim Preston 2007 Blue Creek Archaeological Project, Report of 206 Fieldwork. Report submitted to the National Institute of Culture and History, Belize, by Maya Research Program, Fort Worth, Texas. Guderjan, Thomas H., Lori B. Diel, Antoine Giacometti, and Elizabeth Andrews In Press Maya Social Organization and the Implications of a non-elite burial at Blue Creek. Ancient Mesoamerica Guderjan, Thomas, Robert J. Lichtenstein and C. Colleen Hanratty 2003 Elite Residences at Blue Creek, Belize. Maya Palaces and Elite Residences pages 13-45. Edited by J. J. Christie. University of Texas Press, Austin. Gude rjan, Thomas and C. Colleen Hanratty, 2006 A Thriving Non-Royal Lineage at Blue Creek: Evidence from a Sequence of Burials, Caches and Architecture. Acta Mesoamerica 19. Haines, Helen R. 2000 Intra-Site Obsidian Distribution and Consumption Patterns in Northern Belize and the North-Eastern Petn. Unpublished Ph.D. Dissertation, Institute of Archaeology; University College, London. Hanratty, C. Colleen 2002 Kn Tan; Excavations in the Western Group at Blue Creek. Paper presented at the annual meeting of the Society for American Archaeology, Denver, 2002. Harrison-Buck, Eleanor and Patricia A. McAnany 2005 Terminal Classic circular shrines and Ceramic material in the Sibun Valley 164

PAGE 166

Investigations at Blue Creek, NW Belize 165 Belize: Evidence of Northern Yucatec Influence in the Eastern Maya Lowlands. In Research Reports in Belizean Archaeology Vol III. Houk, Brett 1996 The Archaeology of Site Planning: An Example from the Maya Site of Dos Hombres, Belize. Ph.D. Dissertation, Department of Anthropology, University of Texas, Austin. Preston, Tim 2007 Examining Power Heterarchies within an Ancient Maya Co mmunity. MA Thesis, Department of Anthropology, San Francisco State University. Robichaux, Hubert R., et al 2002 Preliminary report of the 2001 Season of the Punta de Cacao project. The University of the Incarnate Word, San Antonio, Texas.

PAGE 167

15 THE MIDDLE FORMATIVE IDEOLOGICAL FOUNDATIONS OF KINGSHIP: THE CASE FROM CAHAL PECH, BELIZE James F. Garber, Jennifer L. Cochran, and Jaime J. Awe Excavations at Cahal Pech revealed Middle Formative rit ual deposits whose symbolic meaning figure prominently in the ideological foundations of power and authority. One of these is a human skull placed in a bowl adjacent to which is a headless body. The other is a cosmogram-world creation cache, symbolizing a resurrected ancestor. The symbolic program of these taken as a set, mirror the iconographic depictions of the resurrected kings of the Classic Period and are elements of the Popol Vuh creation story. The Middle Formative deposits demonstrate the 2000-year history of symbolic power among the Lowland Maya. Introduction The 2005 field season marks the second year of excavations in Plaza B at Cahal Pech for the Texas State University Belize Valley Archaeological Project (BVAP). In the first field season (2004) a north-south trench was initiated in Plaza B (Garber et al. 2006). These excavations were continued in 2005. The focus of the work was to further document Early Formative and Middle Formative occupations at the site. Much of this work builds upon earlier rese arch at Cahal Pech conducted by Awe (1992) and later by Healy and Awe (1995, 1996). These efforts produced substantial data on a previously unrecognized occupation phase for the Maya Lowlands initiated at the end of the Terminal Formative called the Cunil Phase (1100-900 BC) (see Awe 1992; Garber et al. 2004 and Healy et al. 2004 for a detailed reporting of the Cunil phase and related Kanocha phase at Blackman Eddy). Additionally, the 2004 and 2005 excavations have provided detailed information on Middle Formative occupation and associated ritual deposits. Three of these deposits in particular demonstrate the symbolic foundations for power and authority that form the basis of kingship in the Classic period. Background The site of Cahal Pech is located on a hill overlooking the modern town of San Ignacio in the Cayo District and has been the subject of several investigations spanning over fifty years (Figures 1 and 2). The site was strategically placed on the western bank of the Macal River and covered an approximate area of 10 kilometers during the Classic period (Awe 1992; Healy et al. 2004). The site core consists of 34 large structures including, massive temple mounds, range structur es, two ballcourts, and formal courtyards and plazas containing several stelae and altars (Figure 2). A number of settlement clusters surrounding the site also contained monumental architecture. Investigations at Cahal Pech in the 2004 and 2005 field seasons revealed a series of Middle Formative structures with associated features and ritual deposits which directly overlaid the initial Terminal Early Formative occupation located on bedrock (Garber et al. 2006). Th e trench excavations of these early deposits revealed three low platforms and associated ritual deposits and Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 169-175. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 169

Middle Formative Investigations at Cahal Pech 169 Figure 1 Map of sites in the Belize River Valley. Figure 2 Cahal Pech site core (from Awe 1992). caches. Evidence of exotic items and a marine shell workshop was also encountered and has implications for emerging social stratification at this early date (Cochran 2005). Test units had been excavated in several of the Cahal Pech plazas. It was clear from these plaza tests that Plaza B was the original summit of the hill and that early materials in the other plazas were present, but deeply buried by plaza construction in an effort to enlarge the usable summit of the hill. Thus, Plaza B was the logical place to efficiently expose additional Terminal Early Formative and Middle Formative remains. As noted above, the test units in Plaza B had demonstrated extensive use of the hilltop during Terminal Early Formative and Middle Formative times but were of insufficient size to fully assess architectural

PAGE 170

Garber, et al features, activity areas, ritual activity, and community organization. The Plaza B testpitting program yielded tantalizing information on habitation and utilization of this area (Cheetham 1996). In an effort to help clarify this uti lization and get a more comprehensive view of the architectural variability, a north-sout h trench across Plaza B was initiated in 2004 and continued in 2005. Figure 3. Plaza B at Cahal Pech (after Healy et al. 2004). Excavations All BVAP Plaza B trench excavations are designated as Operation 1 (Op. 1). The main 2004 trench was 16 meters long and 1 meter wide. The 2005 excavations continued this trench an additional 12 meters (Figure 3). Extensions were added to the west of the main trench to investigate encountered features. The trench excavations revealed the presence of Classic and Late Preclassic plaza floors, Late Preclassic architecture and caches, a variety of Middle Formative caches, building platforms, ritual deposits, and a Terminal Early Formative platform. These are reported in detail elsewhere (Garber et al. 2006). Of particular in terest here is Middle Formative Platform B and three of its associated ritual deposits. Platform B was initially constructed in the early facet Kanluk Phase (900-700 BC). Platform B rests directly on bedrock and cuts through an earlier building, Platform C, constructed at the end of the Cunil Phase (1100-900 BC). Platform B is a low broad building in the northwest portion of Plaza B. Its edge consisted of roughly trimmed coursed limestone blocks. Its surface is composed of tamped marl and its core composed of marl and rubble fill. Three deposits associated with Platform B are the focus of this chapter and viewed as a ritual set, form the basis for symbolic power and authority that eventually takes the fo rm of kingship in the Classic Period. Figure 4 Jade artifacts top row of beads are from Cache 7 (skull in bowl). Cache 7 This cache was located in the southwest portion of Op.1g within the northsouth trench and consisted of a bowl, a poorly preserved human skull facing northwest, and six jadeite beads (Figure 4). The vessel was placed beneath a large capstone found at what appears to be the southeast corner of Platform B (Figures 5 and 6). The vessel is a Middle Preclassic, Kanluk phase, Sampope rro Red: Variety Unspecified bowl. It has a waxy red Chicanel-style slip that almost appears streaky (like Society Hall) on a few of the sherds. The vessel is slipped on the interior 170

PAGE 171

Middle Formative Investigations at Cahal Pech and exterior but not on the slightly concave base. The slip fits the description for Giffords Sampoperro Red: Sampoperro Variety (Gifford 1976:77). Paste is not typical for Sierra Red and is finer, buff in color, more Cunil in style, and oxidized throughout. Angular volcanic glass fragments were noted in the petrographic analysis. While Gifford associates this type with late facet Jenney Creek this paste may represent an earlier version. Gifford (1976:78) does describe an early version of Sampoperro Red: Rough-exterior Variety, which he associates with early facet Jenney Creek. BVAP project ceramicist Lauren Sullivan provided the above ceramic assessment. unexcavated Platform B Death (South) Skull in bowl Headless body in crypt symbolically resurrected in sky (north ) New life resurrection 13 greenstones headless figurin e 3 slate bars N Figure 5 Ritual resurrection of Burial 1 Burial 1 A crypt burial was encountered in the southeast corner of Platform B (Middle Formative) (Figure 6). Platform B and this crypt appear to be intrusive into earlier Platform C (Cunil). The crypt and headless body was directly on bedrock. The preservation was poor and highly fragmented, however it is clear that the headless body was fully articulated at the time of burial. Five shell tinklers and a complete obsidian blade were found immediately adjacent to the right humerus and appear to have been components of a band worn on the upper arm (Figure 7). Interestingly, stone celts as upper arm ornaments are a common feature of Chac Mool figures of the Terminal and Post Classic. Immediately to the south, the bowl and skull noted above was found in a separate crypt (Cache 7). It is assumed that this skull belongs to Burial 1, but it should be noted that a row of crypt stones separated the body from the skull (Figure 6). Crypt 1 Skull in bowl Crypt 2 Headless body Figure 6. Plan view of Crypt 1 (Cache 7) and Crypt 2 (Burial 1). We believe that this burial and cache are ritually related to Cache 2 found in 2004 (Garber et al. 2006). Cache 2 consists of a cluster of 13 greenstones immediately above a headless ceramic figurine. Three slate bars were found below the figurine. The 13 greenstones represent the upper world, the 3 slate bars the three-stone-place of creation, and the headless figurine a resurrected 171

PAGE 172

Garber, et al ancesto r. This cache was found in the northeast corner of Platform B (see Figure 5). Figure 7 Shell tinklers from Burial 1. Discussion and conclusions The head-in-bowl motif is pervasive in Classic Period Maya art and iconography. It is shown on numerous Classic Period polychrome vases. Th e head is shown in fleshy form, skeletal form, and at times a combination of fleshy and skeletal (ReentzBudet 1994:14-15). This motif however, is not Maya in origin, and has been traced back to Olmec times (Reilly 1994). A Preclassic Maya example on a mask faade comes from Blackman Eddy (Figure 8). On this example, a head is shown emerging from an outwardly flaring bowl shown in profile, adorned with three large dots. Recent hieroglyphic and iconogr aphic decipherment have underscored the importance of the three-dot symbolism in Maya ideology. When functioning as a symbolic motif, the three dots represent the three stone places, which the gods established as the cosmic center at the beginn ing of the present creation (Freidel et al. 1993:64-71). Additionally, the head and bowl motif is found in many regions of Mesoamerica (Figure 9). Figure 8 Upper east mask on Structure B1-2nd at Blackman Eddy (drawing by Jason Wiersema). In the Popol Vuh creation story, the hero twins obtain the se vered head of their father and resurrect him as the maize god. At the three-stone place of creation the maize god raises the world tree and creates the world. This severe d head within a bowl and its resurrection is a common Classic Period Maya iconographic theme. Figure 9. Formative and Classic examples of bowls with supernatural entities: (a) headdress detail from La Mojarra (redrawn from Winfield Capitaine 1988:figure 1); (b) Tzutzuculi (redrawn from McDonnald 1983:figure 12); (c) Humboldt Celt (redrawn from Reilly 1994:figu re 3.52); (d) Uaxactun (redrawn from Freidel et al. 1993:figure 3:9); (e) Blackman Eddy facade mask (drawn by J. Wiersema); (f) Lowland Maya vessel (redrawn from Freidel et al. 1993:figur e 1:5); (g) Teotihuacan (redrawn from Taube 1983:figure 3); (h) Teotihuacan 172

PAGE 173

Middle Formative Investigations at Cahal Pech We believe that the individual in Burial 1 was decapitated (probab ly postmortem), its head placed in a bowl in the south (Cache 7) and symbolically resurrected as an ancestor in the sky (north) with the deposition of Cache 2 (Figure 5). Placing a ruler or ancestor in the sky accomplishes the same symbolic objective as the placement of ruler portrait stelae in the Twin Pyramid Complexes at Tikal (Ashmore 1991; Garber et al. 2004). The ritual program of these Middle Formative deposits at Cahal Pech demonstrates the symbolic program of death and resurrection. These became critical components of the symbol system of Classic Period kingship and demonstrate the enduring ritual and sym bolic system of the Maya. References Cited Ashmore, Wendy 1991 Site-Planning Principles and Concepts of Directionality among the Ancient Maya. American Antiquity 2:199-226. Awe, Jaime J. 1992 Dawn in the Land between the Rivers: Formative Occupation at Cahal Pech, Belize and its Implications for Preclassic Development in the Maya Lowlands. PhD dissertation, Institute of Archaeology, University of London. Cheetham, David. T. 1996 Reconstruction of the Formative Period Site Core of Cahal Pech, Belize. In Belize Valley Preclassic Maya Project: Report on the 19947 Field Season edited by P. F. Healy and J. J. Awe, 1-33. Occasional Papers in Anthropology no. 12. Trent University, Peterborough, Ontario. Cochran, Jennifer 2005 The Role of Marine Shell Production during the Early and Middle Preclassic at the Site of Blackman Eddy, Belize. Paper presented at the Annual Meeting of the Society for American Archaeology, Salt Lake, April 2005. Freidel, David A., Linda Schele, and Joy Parker 1993 Maya Cosmos: Three Thousand Years on the Shamanss Path William Morrow, New York. Garber, James F., M. Kathryn Brown, Jaime J. Awe, and Christopher J. Hartman 2004 Middle Formative Prehistory of the Central Belize Valley: Examination of Architecture, Material Culture, and Sociopolitical Change at Blackman Eddy. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research. Edited by James F. Garber, University Press of Florida, Gainesville. Garber, James F., Jennifer L. Cochran, and Jaime J. Awe 2006 Excavations in Plaza B at Cahal Pech: the 2004 Field Season. In Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2005 Belize Archaeology Symposium pp.31-40, edited by John Morris, Sherilyne Jones, Jaime Awe, and Christophe Helmke. Institute of Archaeology, Belize. Gifford, James. C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley. Memoirs of the Peabody Museum of Archaeology and Ethnology, vol. 18. Harvard University, Cambridge. Healy, P. F. and J. J. Awe, editors 1995 Belize Valley Preclassic Maya Project: Report on the 1994 Field Season. Occasional Papers in Anthropology no. 10. Trent University, Peterborough, Ontario. 1996 Belize Valley Preclassic Maya Project: Report on the 1995 Field Season, Occasional Papers in Anthropology no. 12. Trent University, Peterborough, Ontario. Healy, Paul F., David Cheetham, Terry G. Powis, and Jaime J. Awe 2004 Cahal Pech: The Middle Formative Period. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research Edited by James F. Garber, University Press of Florida, Gainesville. 173

PAGE 174

Garber, et al 174 Reents-Budet, Dorie 1994 Painting the Maya Universe: Royal Ceramics of the Classic Period Duke University Press, Durham and London. Reilly, F. Kent III 1994 Visions to Another World: Art, Shamanism, and Political Power in Middle Formative Mesoamerica Ph.D dissertation, University of Texas, Austin.

PAGE 175

16 FROM HOUSE TO HOLY: FORMATIVE DEVELOPMENT OF CIVIC-CEREMONIAL ARCHITECTURE IN THE MAYA LOWLANDS Terry G. Powis and David Cheetham Investigations in the Belize Valley have yielded Middle and Late Formative pe riod (ca. 950 B.C.A.D. 100) civicceremonial architecture. During this period the focus of communal ritual life shifted from elite residences to modest temple platforms to temple-pyramids. We discuss this de velopment in a social and political-religious framework. Of particular interest is the shift (ca. 800 BC) from elite houses to temple platformsin some cases built directly abovea reverent treatment suggesting that the descendant s of incipient Maya leaders enjoyed an increased and perhaps uninterrupted role in political and religious affairs. The Belize Valley data are compared and contrasted with other Lowland Maya areas and distant Mesoamerican regions. Introduction Temples are a constant feature of ancient Mesoamerican societies, yet their origins remain somewhat obscure. By this we do not mean which society made them first, although that is also a question worth investigating. Rather, we are puzzled by their location within ancient communities why spot X as opposed to spot Y which remains an enigma in many cases, as do the social, political and ev en practical factors which surely dictated such choices. Without doubt, some of the earliest temple-building societies incorporated these structures into grand civic plans right from the outset. They were not haphazard additions; location was precise and predetermined. The sites of La Blanca (Love 2002) and La Venta (D rucker et al. 1959) are examples, with large earthen temples forming the very heart of the respective civic-ceremonial blue prints, in both cases most likely brought to fruition between 900800 BC in un-calibrated radiocarbon years. As the practice of temple construction and temple worship quickly spread to other regions, these buildings were often incorporated unto the unfolding civicceremonial plan of pre-existing communities. The central Maya Lowlands is one such region (Figure 1). Between about 1000800 BC, permanent village life was established here; quite late compared to neighboring regions. These villagers made use of basic household objects common elsewhere in Mesoamerica pottery, figurines, and so forth but in styles indicating an indige nous spin on common themes. Immigration cannot, on the basis of the present material record, account for the origins of these earliest pre-Mamom or "Cunil horizon" villa gers (Cheetham 2005). It was only at the end of this very formative stage of lowland Maya development that, along with the emergence of social ranking, modest temple buildings began to be constructed. Evidence is sketchy at present, but it appears that in some cases the focus of ritual life shifted from the households of emerging elites to modest temple platforms on the same spot. The observations we make here should be considered preliminary and, in some respects, speculative. Yet, there is an emerging pattern in the lowland Maya that is very intriguing and promises to tell us much about the foundation, nature, and development of the region's social and political structure. Fi gure 1 locates some of the sites we mention. While they are few Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 177-186. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 176

Terry Powis & David Cheetham and far between, it is important to note that lowland Maya temples are large; early constructions within them are deeply buried, relatively small, and costly to get at. We suspect that such data will increase in the years to come, allowing the model we present here to be evaluated more thoroughly, perhaps even discarded. Figure 1. Map of the Southern Maya Lowlands locating sites mentioned in text. Two Early Mesoamerican Cases Ancient Mesoamerican societies did not, of course, lack specif ic ritual loci before the advent of temple-life in the early Middle Formative period. Such activities took place in open-air settings, in small special-purpose buildings, or in and around residencesand quite likely all of them. There is ample evidence to back this up, but we need not rehearse it here. It is instructive, however, to look at a few cases th at contrast with the early lowland Maya. One of the better-known pre-temple archaeological cultures thrived along the Pacific coast of Chiapas between 1600-1150 BC. Dubbed the Mokaya people by John Clark (1991, 1994), these folks built large sprawling communities with standing architecture. The region's largest site of the era, Paso de la Amada, boasted a ballcourt and numerous raised house platforms, including large chiefly residences central to the community's social, political, and ritual life (Blake 1991; Blake et al. 2006; Clark 1994; Hill 1999; Lesure 1999; Lesure and Blake 2002). One building (Figure 2a) was rebuilt many times, creating an ever-larger earthen platform in the middle part of the site. But it was never a temple, and Paso de la Amada was abandoned by 1150 BC, well before temple-life spread across Mesoamerica. What is instructive about Paso de la Amada is the focal point that the elite residences maintained in the emergence of complex society within the society. The chiefly aspirants that lived there used a host of social maneuvers perhaps unwittingly which resulted in inequality. The built environment and the la bor coordination in making it was a key element in the shift away from egalitarianism (Clark 2004). Figure 2 Artists reconstruction of Early Formative period pre-temple Mesoamerican perishable superstructures: (a) Mound 6, a chiefly residence at Paso de la Amada, Chiapas (illustration by Ayax Moreno, courtesy of the New World Archaeological Foundation); (b) two San Jose phase buildings at San Jos Mogote, Oaxaca (after Flannery and Marcus 1996:Fig. 4.6, 2005:Fig. 18.2). 178

PAGE 177

Formative Development in the Maya Lowlands Another pre-temple archaeological culture worth mentioning is the Oaxaca Valley site of San Jose Mogote. Unlike Paso de la Amada, this community was not abandoned at the close of the Early Formative period, and thus the location and nature of early buildin gs is of particular interest in relation to what came after. Figure 2b shows reconstruction drawings of two early buildings from the site. Kent Flannery and Joyce Marcus (1976, 2005; Marcus and Flannery 1996) interpret the building on the left as an initiates or Mens House; a special-purpose structure with more resource investment in terms of upright posts and plaster. It was not a residential building, as suggested by the lack of household debris within and around it. The building on the right is interpreted as an elite residence, its occupants enjoying access to goods and raw materials not accessible to lower ranking families within the society. For our purposes, wh at is in structive about this case is that later San Jose Mogote temples those post-dating 850 BC were not built over any elite residential or special structures, at least not those documented thus far. There is no unbroken, in-situ sequence of residential-to-public architecture even though the peoples of this community fully participated in the temple life during the early Middle Formative period. As Joyce Marcus recently informed us (personal communication 2005), the peoples of San Jose Mogote rarely mixed sacred and secular contexts by converting a non-sacred context into a sacred context. This, we submit, was what was happening between about 850-800 BC in many lowland Maya communities. The Lowland Maya One of the best cases for an elite residential-cum-public building in the lowland Maya area occurs at Cahal Pech, Belize. A roughly 2,000 year, uninterrupted Figure 3. Cahal Pech, Belize: (a) The main plaza; (b) Structure B4 profile; (c) artifacts from Structure B4 excavations and artists reconstruction of Structure B4 at the end of the Cunil phase. (after Cheetham 1995:Fig. 1, 1996:Fig. 1, 1998:Fig. 3a) 179

PAGE 178

Terry Powis & David Cheetham sequence of public build ings was exposed in the southeast corner of the site's main plaza within Structure B4 (Figure 3a). The stratigraphic sequence of this building (Figure 3b) includes nine superimposed temples, the latest of which is Late Classic, the earliest of which is early Middle Formative approximately 800 BC. Below this are a series of residential platforms dated to pre-Mamom Cunil times on the basis of artifactual content and associated radiocarbon dates (Awe 1994; Cheetham 1998; Healy et al. 2004; Powis et al. 2001). There is little doubt that these are residen tial buildings. Unlike the Oaxaca initiates buildings noted above, the architectural fill, immediate patio area, and floors themselves are strewn with trash of an unmistakably domestic type. Yet these Cunil residences particularly the final ones exhibit traits which clearly set them apart from contemporaneous domestic structures within the rest of the early village, portions of which were uncovered elsewhere below the main plaza (Cheetham 1996). In essence, the families residing at B4 maintained a higher rank than their neighbors, just a stones-throw away. Several classes of exotic items are associated with the B4 buildings (Figure 3c), including potsherds in cised with abstract supernatural creatures and other mythicreligious concepts. Such depictions are rare to absent elsewhere below Plaza B the central and oldest part of the site. Indeed, the Cunil occupants of B4 enjoyed access to a wealth of exotics, which are rare or lacking beyond their immediate living space. Many of the designs on the pots occur during or just before Cunil times in regions beyond the Maya lowlands, including Chiapas (e.g., Clark and Cheetham 2005), Oaxaca (e.g., Flannery and Marcus 1994), and the Mexican Highlands (e.g., Niederberger 1976). Long-distance contact with foreign peoples seems to have been part of the process that led to the emergence of the apparent, if not rudimentary, social ranking at Cahal Pech (Cheetham 1998). This interregional c ontact clearly had a heavy ritual and religious component, to judge by the themes etched on the pots. Figure 4. Looking down at Structure B4 9-sub, the first public temple building in the B4 architectural sequence. (photograph by D. Cheetham) Architecturally, the B4 buildings particularly the final Cunil examples are more elaborate than contemporaneous structures below the plaza. Plaster was liberally used, interior floors are recessed into the platform on which the buildings stood, interior benches were constructed and, on one building, the exterior was painted red-and-white. But more important for the present discussi on is the fact that, like later temples in the B4 sequence, the Cunil residences were oriented slightly west of magnetic north. Th ere exists clear continuity in terms of building orientation. Moreover, at the close of the Cunil phase the superstructure of the final residential building was burnt, its ashes neatly piled onto the interior floor along with several exotic objects and incised potsherds. Presumably, this was done to celebrate the long-standing residential service of this location, and quite possibly to inaugurate the next building the first temple the front face of which is shown in Figure 4. This was a smooth transition; we think it was a 180

PAGE 179

Formative Development in the Maya Lowlands deliberate conversion of the location from secular to sacred and that it was driven by the same emerging elites that occupied the final B4 Cunil residence. It may, however, be incorrect to characterize the early residences as entirely secular given the persistence, regularity, and exclusiveness of the Cunil era iconograp hy in this location. Until re cently, these Cahal Pech data seemed very unusual. This is no longer the case. Several lowland Maya sites have yielded data consistent with this interpretation. One of these is Blackman Eddy, located nine kilo meters east of Cahal Pech. In the north part of Blackman Eddy (Figure 5a), a stratigraphic sequence of building platforms has been recorded from a partially bulldozed temple (Garber et al. 2004a, 2004b). The investigators classified each platform as one of three types: domestic, public/integrative, and monumental/restrictive (Brown and Garber 2003). For our purposes, the latter two classes can be considered temples. The early domestic or residential buildings (Figure 5b) at the bottom of the architectural sequence were built on or close to bedrock, are apsidal and circular in shape, and yielded Cunil horizon markersmost notably, ceramics and figurines. Like Cahal Pech, there is evidence of imported exotics and incised pan-Mesoamerican designs on pots. Unlike Cahal Pech, intrasite data are lacking, preventing an assessment of the exclusivity of these items to this location, although we suspect that this was the case. The main point is that Blackman Eddy provides another case of clear continuity from residential to public architecture, in one area. Due to the broad exposures of Structure B1, the buildings above the early residential structures ha ve been almost fully exposed and are unmistakably public in nature (Figure 5c). The other Belize Valley site of note is Xunantunich (Figure 6a), where tunneling Figure 5. Blackman Eddy, Belize: (a) Plaza B, showing Mound 1 location and bulldozer cut; (b) plan of pre-Mamom B1 domestic structure outlines/postholes; (c) artists reconstruction of Middle Formative public platforms built directly over Cunil phase residences. (after Garber et al. 2004a:Figs. 3.4, 3.6, 2004b:Fig. 4.1) 181

PAGE 180

Terry Powis & David Cheetham along bedrock below the site's largest Classic period temple yielded an undisturbed deposit of Cunil material (Strelow and LeCount 2001). No architecture was enc ountered, but potsherds and other items were found along a bedrock slope leading up to the summit of the original hill, deep below the massive Classic period temple. In other words, a Cunil sheet midden behind a residential structure that likely stood at the t op of this natural promontory the highest spot on the site. The collection of Cunil potsherds from this location, like Cahal Pech and Blackman Eddy, includes examples with panMesoamerican motifs. This pattern em erges again as we move west into the Peten district of Guatemala. The University of Pennsylvania excavations of Tikals North Acropolis (Coe 1990) uncovered a chultun and other domestic features, wh ich, on the basis of pottery and other artifacts, can be securely assigned to the Cunil horizon. This location (Figure 6b) appears to have been a natural high point in the original topography, although with so much modification of Tikal's landscape it is difficult to be sure. In any case, the pottery has incised panMesoamerican designs and there is, of course, a clear case of later temple construction in this location. Despite considerable excavati on, no similar deposits were found in the immediate area of the North Acropolis and Great Plaza. Two other collapsed chultuns containing abundant Cunil horizon material were, however, found in Tikal's Mundo Perdido Group during excavations by Juan Pedro Laporte and Vilma Fialko (1993; Cheetham et al. 2003). The approximately 500 meter distance between the North acropolis and Mundo Perdido suggests that pre-Mamom Tikal may have consisted of two small villages, or perhaps one very dispersed village. Figure 6. Sites with Cunil Horizon remains: (a) Xunantunich, Belize, loca ting excavation tunnel at the rear of the principal Mound, El Castillo (after Keller 1993:Fig. 1); (b) Tikal, Guatemala, locating deposit at the center of the North Acropolis (after Harrison 1999:Fig. 36). This also appears to be the case at Uaxactun. The Carnegie Institution's excavations in Plaza E (Ricketson and Ricketson 1937) recovered mixed deposits of Mamom and Cunil horizon pottery, the latter including pan-Mesoamerican motifs. An undisturbed sample of identical pottery has recently been identified from a midden deposit discovered during the Carnegie excavations below a sequence of small temples within the South Court of Group A (Smith 1937), located a kilometer to the west of Plaza E and on much higher ground 182

PAGE 181

Formative Development in the Maya Lowlands Figure 7. Sites with Cunil Horizon remains: (a) Uaxactun, Guatemala, locating deposit below Structure A-1 (after Ri cketson and Ricketson 1937:Fig. 198); (b) Holmul, Guatemala, locating deposit below Group II, Structure B (map courtesy of Francisco Estrada-Belli). (Figure 7a). This represents domestic building fill or an associated midden. Recently, Francisco Estrada-Belli, Nina Neivens, and Michael Callaghan discovered a large samp le of Cunil pottery deep within Group II Building B, at Holmul (Figure 7b). Although some early Mamom sherds (post Cunil) were mixed in this deposit, it comes from architectural fill nearly 4 meters above the surrounding plaza and directly below a series of later temples. As their team digs further into this building a Cunil building may come to light. Thus far, no other deposit like this has been encountered at the site, although preMamom potsherds are have been found at Cival, a large center to the north of Holmul. Conclusions At the outset of this paper we posed a question: Why in early Mesoamerican societies were the first temple buildings constructed in spot X as opposed to spot Y? In the lowland Maya case, there is an emerging pattern, whic h may provide an answer. At sites like Cahal Pech, Blackman Eddy, and probably many, many others, preMamom elite residences were the focal point of community ritual based on the incorporation of es oteric knowledge, concepts, and quite likely exotic finished goods and raw materials. The symbols themselves were not passive decorative designs, but rather, meaningful representations that, through time, became associated with internal competition, information exchange, prestige, and ultimately differential social status. They signify powerful supernatural and cosmological themes potent in the emergence of ranked society in the Maya lowlands, as they did elsewhere in Mesoamerica several centuries earlier. Around 800 BC, for reasons yet to be determined, at least some emerging Maya elites deliberately transformed their residential space into the region's first temple platforms. Within 500 years those buildings became temple-pyramids. This pattern is different from well-documented archaeological cultures in other regions, where elite residences did not become public buildings, at least not public temple buildings. What we are suggesting is that in at least some instances Middle Preclassic 183

PAGE 182

Terry Powis & David Cheetham tem ples represented the "house" of the founding lineage. They remained such a "house" even though, physically and publicly, a radical transformation had taken place. With one archit ectural alteration they went from private to public, from secular to sacred. But the terms secular and private are probably not completely appropriate terms for all practices undertaken at the early elite residences. The occupants likely maintained communal functions as well, shepherding those not in direct possession of esoteric knowledge and long-distance contacts. That may have been the glue binding these early communities. Whatever the case, their legacy was secured for centuries to come through the reverent trea tment of their living space, ensuring their descendants an increased and perhaps uninterrupted role in political and religious affairs. References Cited Awe, Jaime 1994 Ritual, Religion, and Cultural Complexity in the Middle Formative Belize River Valley. Paper presented at the 12th Maya Weekend, University Museum, University of Pennsylvania, Philadelphia. Blake, Michael 1991 An Emerging Early Formative Chiefdom at Paso de la Amada, Chiapas, Mexico. In The Formation of Complex Society in Southeastern Mesoamerica edited by William R. Fowler, Jr., pp. 27-46. CRC Press, Boca Raton. Blake, Michael, Richard G. Lesure, Warren D. Hill, Luis Barba, and John E. Clark 2006 The Residence of Power at Paso de la Amada, Mexico. In P alaces and Power in the Americas: From Peru to the Northwest C oast edited by Jessica Joyce Christie and Patricia Jo an Sarro. University of Texas Press, Austin. Brown, M. Kathryn, and James F. Garber 2003 Excavations on Structure B1 at Blackman Eddy: The 2002 Season. In The Belize Valley Archaeology Project: Results of the 2002 Field Season, pp. 5-20. Report submitted to the Department of Archaeology, Belmopan, Belize. Southwest Texas State University, San Marcos. Cheetham, David 1995 Excavations at Structure B-4, Cahal Pech, Belize: 1994 Operations. In Belize Valley Preclassic Maya Project: Report on the 1994 Field Season, edited by Paul F. Healy and Jaime J. Awe, pp. 18-44. Trent University Occasional Papers in Anthropology No. 10. Peterborough. 1996 Reconstruction of the Formative Period Site Core of Cahal Pech, Belize. In Belize Valley Preclassic Maya Project: Report on the 1995 Field Season, edited by Paul F. Healy and Jaime J. Awe, pp. 1-33. Trent University Occasional Papers in Anthropology No. 12. Peterborough. 1998 Interregional Interaction, Symbol Emulation, and the Emergence of SocioPolitical Inequality in the Central Maya Lowlands. Masters Thesis, Department of Anthropology and Sociology, University of British Columbia. 2005 Cunil: A Pre-Mamom Horizon in the Southern Maya Lowlands. In New Perspectives on Formative Mesoamerican Cultures pp. 27-38, edited by Terry G. Powis. BAR International Series 1377, Oxford. Cheetham, David, Donald W. Forsyth, and John E. Clark 2003 La cermica Pre-Mamom de la cuenca del Ro Belice y del Centro de Peten: Las correspondencias y su s implicaciones. In XVI Simposio de Investigaciones Arqueolgicas en Guatemala edited by Juan Pedro Laporte, Brbara Arroyo, Hctor Escobedo and Hector Meja, pp. 615-634. Museo Nacional de Arqueologa y Etnologa, Asociacin Ti kal, Guatemala. Clark, John E. 1991 The Beginnings of Mesoamerica: Apologia for the Soconusco Early Formative. In The Formation of Complex Society in Southeastern Mesoamerica edited by William R. Fowler, Jr., pp. 13-26. CRC Press, Boca Raton. 184

PAGE 183

Formative Development in the Maya Lowlands 1994 The Development of Early Formative Rank Societies in the Soconusco, Chiapas, Mexico. Ph.D. dissertation, Department of Anthropology, University of Michigan, Ann Arbor. 2004 Mesoamerica Goes Public: Early Ceremonial Centres, Leaders, and Communities. In Mesoamerican Archaeology: Theory and Practice edited by Julia A. Hendon and Rosemary R. Joyce, pp. 43-72. Blackwell, Oxford. Clark, John E., and David Cheetham 2005 Cermica del Formativo de Chiapas. In La produccin alfarero en el Mxico antiguo, Volumen I, edited by Beatriz Leonor Merino Carrin and ngel Garcia Cook, pp. 285-433. Instituto Nacional de Antropologa e Historia, Mexico City. Coe, William R. 1990 Excavations in the Great Plaza, North Terrace, and North Acropolis of Tikal. University of Pennsylvania, University Museum Monograph No. 61. Tikal Report 14. Philadelphia. Drucker, Philip, Robert F. Heizer, and Robert J. Squier 1959 Excavations at La Venta, Tabasco, 1955. Smithsonian Institution, Bureau of American Ethnology, Bulletin 170. U.S. Government Printing Office, Washington, D.C. Flannery, Kent V., and Joyce Marcus 1976 Evolution of the Public Building in Formative Oaxaca. In Culture Change and Continuity: Essays in Honor of James Bennett Griffin edited by Charles E. Cleland, pp. 205-221. Academic Press, New York. 1994 Early Formative Pottery of the Valley of Oaxaca, Mexico. Memoirs of the Museum of Anthropology University of Michigan, No. 27. Ann Arbor. 2005 Excavations at San Jose Mogote 1: The Household Archaeology. Memoirs of the Museum of Anthropology, University of Michigan No. 40. Ann Arbor. Garber, James F., M. Kathryn Brown, Jaime J. Awe, and Christopher J. Hartman 2004a Middle Formative Prehistory of the Central Belize Valley: An Examination of Architecture, Material Culture, and Sociopolitical Change at Blackman Eddy. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp 25-47. University Press of Florida, Gainesville. Garber, James F., M. Kathryn Brown, W. David Driver, David M. Glassman, Christopher J. Hartman, F. Kent Reilly III, and Lauren A. Sullivan 2004b Archaeological Investigations at Blackman Eddy. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edited by James F. Garber, pp. 48-69. University Press of Florida, Gainesville. Harrison, Peter D. 1999 The Lords of Tikal: Rulers of an Ancient Maya City Thames and Hudson, London. Healy, Paul F., David Cheetham, Terry G. Powis, and Jaime J. Awe 2004 Cahal Pech, Belize: The Middle Formative Period. In The Ancient Maya of the Belize Valley: Half a Century of Archaeological Research edi ted by James F. Garber, pp. 103-124. University Press of Florida, Gainesville. Hill, Warren D. 1999 Ballcourts, Competitive Games, and the Emergence of Comple x Society. Ph.D. dissertation, Department of Anthropology and Sociology, University of British Columbia. Keller, Angela H. 1993 Vision and Revision: The Remapping of Xunantunich. In Xunantunich Archaeological Project: 1993 Season, pp 86-99. Report Submitted to the Department of Archaeology, Belmopan, Belize. University of CaliforniaLos Angeles. Laporte, Juan P., and Vilma Fialko 1993 Anlisis Cermico de Tres Depsitos Problemticos de Fase Eb, Mundo Perdido, Tikal. In Tikal y Uaxactn en el Preclsico, edited by Juan P. Laporte and Juan A. 185

PAGE 184

Terry Powis & David Cheetham 186 Valds, pp.53-69. Universidad Nacional Autnoma de Mxico, Mxico. Lesure, Richard G. 1999 Platform Architecture and Activity Pattern s in an Early Mesoamerican Village in Chiapas, Mexico Journal of Field Archaeology 26(4):391-406. Lesure, Richard G., and Michael Blake 2002 Interpretive Challenges in the Study of Early Complexity: Economy, Ritual, and Architecture at Paso de la Amada, Mexico. Journal of Anthropological Archaeology 21(1):1-24. Love, Michael W. 2002 Early Complex Society in Pacific Coastal Guatemala: Settlements and Chronology of the Ro Naranjo, Guatemala. Papers of the New World Archaeological Foundation No. 66. Brigham Young University, Provo. Marcus, Joyce, and Kent V. Flannery 1996 Zapotec Civilization: How Urban Society Evolved in Mexicos Oaxaca Valley Thames and Hudson, London. Niederberger, Christine B. 1976 Zohapilco, Cinco Milenios de Ocupacin Humana en un Sitio Lacustre de la Cuenca de Mxico. Coleccin Cientfica 30. Instituto Nacional de Antropologia e Historia, Mxico City. Powis, Terry, David Cheetham, Bobbi Hohmann, Jaime Awe, and Paul Healy 2001 Un Estudio Preliminar de la Arquitectura del Preclsico en el Valle de Belice. In XIV Simposio de Investigaciones Arqueolgicas en Guatemala edited by Juan P. Laporte, Hector Escobido, and Monzn de Suasnavor, pp. 81-101. Ministerio de Cultura y Deportes, Instituto de Antropologa e Historia, Asociacin Tikal. Guatemala City. Ricketson, Oliver G., Jr., and Edith B. Ricketson 1937 Uaxactun Guatemala Gr oup E, 19261931. Carnegie Institution of Washington Publication No. 477. Washington, D.C. Smith, Robert E. 1937 A Study of Structure A-1 Complex at Uaxactun, Guatemala. Carnegie Institution of Washington, Publication No. 456. Washington, D.C. Strelow, Duane, and Lisa LeCount 2001 Regional Interaction in the Formative Southern Maya Lowlands: Evidence of Olmecoid Stylistic Motifs in a Cunil Ceramic Assemblage from Xunantunich, Belize. Poster presented at the 66th annual meeting of the Society for American Archaeology, New Orleans.

PAGE 185

17 ROCKSHELTER EXCAVATIONS IN THE CAVES BRANCH RIVER VALLEY Gabriel Wrobel, James Tyler, and Jessica Hardy Research by the Belize Valley Archaeological Reconnaissance project during the summers of 2005 and 2006 focused on the excavation of th e Caves Branch Rockshelter, as well as th e initial surveying and test pitting of other sites in the surrounding river valley. The data from the Caves Branch Rockshelter suggest that a small farming community used it as a cemetery sometime during the Protoclassic period, after which other local groups occasionally visited it. The skeletal population from the cemetery includes both sexes and all age groups, including a high ratio of infants. These demographic proportions are typical of pre-industrial societies, though are rarely found at Maya sites because of cultural bias in mortuary patterning at complex urban centers. The age and sex ratios of the skeletal population, as well as the lack of signs of social complexity in this early, rural community, suggest that this context may contain a relatively complete population that could serve as a skeletal reference population, thus aiding in future bioarchaeological studies of the ancient Maya. Introduction The Caves Branch River Valley is located just east of Belmopan along the Hummingbird Highway near the modern village of Armenia and the ancient Maya site of Deep Valley. The karst mountains lining the valley are riddled with caves evidencing human activity, and these sites have been a focus of investigations by the Belize Valley Archaeological Reconnaissance project during the 2005 and 2006 field seasons. So far, the most extensive excavations have been conducted at Caves Branch Rock shelter (CBR), first investigated by Juan Luis Bonor in 1994 and 1995 (Bonor 1995a and 1995b, Bonor et al 1999). During 2006, the Caves Branch project was expanded by systematically identifying other rockshelters in the area showing evidence of ancient use. Test pits were placed in one of these, Deep Valley Rockshelter 1 (DVR-1), to help determine the dates and nature of its use. The project was initiated with two basic research agendas. First, we are focusing on these sites in an effort to further identify variation in ancient cave use, which was a central theme of the Western Belize Regional Cave Project (Awe 1998: 1). The WBRCP specifically investigated light, penumbral and dark zone contexts at three sites, Actun Nak Beh, Actun Uayazba Kab, and Actun Halal to determine if the ancient Maya used specific areas differentially. With a few notable exceptions, rockshelters largely have been overlooked in Maya cave studies despite the ritual significance these sites had for ancient, as well as for modern, Maya. The preliminary data from Caves Branch clearly show that the use of rockshelters by the Maya changed over time. Furthermore, the mate rial culture found in rockshelters is easily distinguishable from that found in dark zone caves, suggesting that the geomorphological differences were important to the ancien t Maya in defining the ritual functions of sites. The basic research goals related to caves are (1) to further characterize the ritual use of rockshelters, noting changes in their use over time, and (2) to explore other rockshelters in the area to determine intersite variability. A second important research focus of the Caves Branch projec t is bio-archaeology. The density of burials found by Bonor Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 187-196. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 186

Wrobel, et al during his prelim inary excavations led him to estimate that the site held approximately 150 individuals, which appears now to be an underestimate (Bonor et al 1999). Glassman's analysis, following Bonor's original excavations, identified individuals of both sexes and a range of age groups, including numerous infants (Glassman and Bonor 2005). Based on the sheer numbers of individuals, on the age/sex profile of the initial skeletal population, and on Bonors hypothesis that this was the funerary site of a small rural agricultural community, we decided to reopen excavations in the hopes that the skeletons were representative of an entire community. If this were the case, then this skeletal population would be relatively unique in the Maya region. Without the complex rules governing burials at ceremonial centers, its possible that this skeletal population may be demographically accurate and thus could be used as a biological reference population to create standards for analysis and comparison with other sites, much the way data from the colonial site of Tipu is used currently (Wrobel et al. 2002, Danforth et al. in press). To determine whether the skeletal series is such an appropriate reference sample, the bio-archaeological research goals are focused on (1) determining whether the mortuary use of the site was limited in time and (2) seeking evidence for the presence of mortuary differentiation within this community, which might indicate that some individuals were buried elsewhere Caves Branch Rockshelter The Caves Branch Rockshelter is approximately 35 meters long and 15 meters high, with a maximum depth of 10 meters (Figure 1). Nearby are other cave sites, a Late Classic plazuela group and the bulldozed remains of a number of housemounds, and a fresh water source. Bonor and colleagues (1999) hypothesized that ritual activity at the site was limited to a single function, that of a domestic cemetery. To test this hypothesis, we looked for the presence of unique interments, such as secondary or sacrificia l burials, or of other specialized deposits, which would suggest that the site had multiple ritual roles. Figure 1 Map of Caves Branch Rockshelter with excavation operations. Interpretation of burial features is often difficult since the site was constantly churned up by new burials, in the process disturbing older one. In many cases, this activity, as well as other taphonomic processes, usually will result in puzzling contexts. However, the bones uncovered thus far do not exhibit cut marks or any other indication of ritual postmortem treatment. In addition, none of the disarticulated remains found in the Caves Branch Rockshelter appear to be formal 188

PAGE 187

Investigations at Cave Branch Rock Shelter bundled rem ains of s econdary burials. Instead, much in the same way that the bones of individuals placed in tombs were often pushed to the side to make way for a new body, the bones of i ndividuals disturbed by a newly dug grave were included in the grave fill. Often, some of the larger or more distinctive elements were piled on or near the new interment and excavation of nearby contexts generally revealed the undisturbed portion of the body left intact. Dates for the burials were derived from several vessels, which were clearly interred as grave goods (Figures 2 and 3). The styles found at CBR are utilitarian, and are best described by Reents (1980: 168186, figure 21a-24d) and Gifford (1976). The vessels found in grav es all date to the Hermitage phase, and include types such as Succotz Striated and Cocay Appliqued, which are found throughout caves in Western Belize, including Petroglyph, St. Hermans, Actun Balam, and others in the Caves Branch, Roaring Creek, and Sibun River Valleys (Graham et al. 1980). None of these vessels have kill holes, which suggest an important distinguishing characteristic between grave goods and of offerings in the form of ceramic vessels often placed in caves. Thus far, no artifacts have been found within the vessels. Most show charring on the bottom, suggesting that they were cooking vessels, and may have held food when they were buried. Potsherds in the rockshelter not directly associated with burials show a much wider variety of fo rms spanning a longer time period, ranging from the Middle Preclassic through the Terminal Classic, including Floral Park, Hermitage, Spanish Lookout and New Town Complexes. Again, most seem to come from very functional cooking vessels, though there are some other decorative examples as well. This range of dates is typical of other ceramic assemblages found in other rockshelters and caves in the area. Figure 2. Protoclassic vessel included as grave good in Caves Branch Rockshelter. Figure 3. Detail of Protoclassic vessel included as grave good in Caves Branch Rockshelter In general, the ar tifacts found in the rockshelter are consistent with those expected from a small rural farming community, including net weights, local fauna, chert flakes, and jute shell. An arrow point found near the surface suggests hunters stopped by the rockshelter in the Terminal Classic or Postclassic period (Figure 4). While a fe w exotic or imported items were found at the site, including several small pieces of jadeite, obsidian, and carved marine shell, th ese are items that are consistently present in low levels in most settlement contexts. In addition, these 189

PAGE 188

Wrobel, et al Figure 4. Arrow point surface collected from Caves Branch Rockshelter exotic items were found in the general matrix rather than with burials, perhaps suggesting a later date than the burials. The only possible indi cator of status hierarchies in the population was an elaborately carved bon e (Figure 5) found near the head of a small child. This individual was discovered in the south wall of unit 13G and was represented solely by cranial elements. The skull was either placed face down or became situated in that manner as a result of taphonomic processes following interment. While it has been stated that the population interred within the Caves Branch Rockshelter appears to represent a small, agrarian, non-elite community (Glassman and Bonor 2005, Wrobel and Tyler 2006), the presence of this type of artifact raises questions as to this assumption. It has been suggested that this type of woven mat motif is, in many cases, a symbol intimately tied with royalty and council houses. Citing the Dictionario San Francisco (Michelon 1976), Grube et al. (1995) illustrates that these mat houses were referred to as a popol nah. Friedel et al. (1993) observed this mat pattern on the walls of a building at the site of Uaxactun. Their interpretation of this motif suggests that it represents, a community council house and a place where the king interacted with his peopleespecially through the performance and teaching of sacred dance (Friedel et al. 1993:143-143). Whether or not this specific artifact represents a direct link to a royal council house is, at this point, debatable given the utilitarian nature of all the other artifacts recovered thus far, and the lack of differentiation in the burials, which might suggest the presence of a social hierarchy. Figure 5. Carved faunal bone showing weaved mat motif Finally, a Late Archaic Lowe point was discovered in association with the lower torso of a probable female adult, which was placed in a tightly flexed position (Figure 6). It was evident that this burial had been disturbed, probably as a result of numerous burials interred directly above this individual. The Lowe point was located approximately 9cm. from the knee area of the individual and wa s oriented at 39 degrees NNE with the point facing towards 190

PAGE 189

Investigations at Cave Branch Rock Shelter the rockshelter. As discussed by Lohse et al. (2006), radiocarbon dates for this type of Archaic point have been based on three sam ples taken from northern Belize, two of which have been described by the author as loose associations ( p. 217), a context that is all too common with regard to these types of discoveries. However, these samples have provided a date between 2500-1900 B.C. for Lowe points, a range that can be corroborated by examples Pohl et al. (1996) discovered at Pulltrouser Swamp. This date raises a number of interesting points concerning this specific interment at the rockshelter. While its close association to the Protoclassic burials would suggest the individual is Maya, it is from the deepest cultural levels of the rockshelter. If the burial does date to the archaic period, then it would represent the oldest dated human remains in Belize, and among the oldest in Mesoamerica. If, however, the burial is contemporaneous with the Protoclassic interments, then this represents a case of curation of discarded objects, perhaps for use as divination/ divi ning tools or personal sacra, as discussed by Brown (2000) for both ancient and modern Maya ritual. Figure 6. Archaic Lowe point found with a burial in the Caves Branch Rockshelter Deep Valley 1 Rockshelter During the 2006 field season, the Caves Branch project initiated excavations at the site of Deep Valley Rockshelter 1 (DVR-1) in an attempt to begin to characterize the diversity of material culture of rockshelter sites in the region (Figure 7). This site was targeted for investigation based on its similarity to CBR. Both rockshelters are approximately the same sizeheight, depth, leng th, with a flat floor, and both have small caves within them. Like at CBR, the surface of DVR-1 was littered were numerous ceramic sherds, faunal remains, two pieces of human bone, and jute shells. However, the initial excavations have clea rly shown that the deposits at DVR-1 are very distinct from those at Caves Branch. The matrix at DRV1 is incredibly dense with jute snails--a 1x1 test pit yielded over 20,000 shells in the first 60 cm. These snails were a source of food for the ancient Maya, are often associated with ritual activity in caves, and even today are abundant in the Caves Branch River, which flows next to DR V-1. Interestingly, the DVR-1 jute are generally very small, and with the exception of the larger shells, the majority does not have their ends spirelopped, which is the typical way to extract the meat when eating them (Halperin et al 2003). At CBR, the jute was generally larger and less plentiful. One explanation for these differences may be related to the riverine source of the jute shell, which may have been different for the two sites. Another possibility is that the jute deposits were temporally distinct at the two sites. A recent study by Cook and Salter-Pedersen (2006) of pomacea shells at Chau Hiix found that they were smaller in the Early Classic compared to the Terminal Classic, which they tentatively interpret as overharvesting when the p opulation of the site was larger. The dense deposits of jute shell at DVR-1 are similar to that of the 191

PAGE 190

Wrobel, et al 192 rockshelters in the Maya mountains, where they were used as grave fill (Prufer 2002). Thus far, no burials have been found at DVR-1, and the relative paucity of early ceramics there suggests th at it did not have the importance of CBR to the Protoclassic population. Figure 7 Map of Deep Valley Rockshelter 1 with excavation operations Skeletal Remains fr om Caves Branch Rockshelter Skeletal remains excavated from the Caves Branch Rockshelter in the last two seasons have undergone preliminary analysis. Bonors ex cavations identified a total of 32 primary burials, and in the 12 weeks of excavations conducted during the 2005 and 2006 seasons, 55 more. As discussed above, part of the Caves Branch project's research agenda focuses on defining the demographic character of the skeletal population in these rockshelters. The age distribution of the CBR skeletal series (Table 1) shows a mortality pattern that heavily favors infants and children (30.4%), when compared to other Classic Maya sites. Such a high infant mortality is very typical of pre-industrial societies. These statistics are im portant to physical anthropologists in demonstrating differences in health between groups separated by time, space, or social class. Unfortunately for bio-anthropologists wo rking in the Maya region, cemetery samples like this are rarely identified. One reason for this is that archaeologists often tend to target architecture and bio-an thropologists in this region do not generally have their own archaeological research designs. There are few sites with large Maya skeletal collections in general, and for none of these can we can be certain that the sex and age distributions are representative of their source living populations, since their composition can be affected by cultural and taphonomic factors. In other words, our samples are not samples, since they are not a random group of i ndividuals selected from the population, and thus are not good models for ancient Maya populations. Instead, the skeletons represent those individuals that archaeologists are most likely to find and collect. So far, Caves Branch Rockshelter has produced a very promising collection. The population shows biological stresses, such as linear enamel hypoplasias and porotic hyperostosis, typical of Maya populations. Again, though, the most exciting aspect is the possibility of actually being able to quantify these rates within the population. Conclusions The following conclusion summarizes the current data from CBR and DVR-1 as they relate to the general hypotheses of the Caves Branch project outlined above. 1. To further characterize the ritual use of rockshelters, noting changes in their function over time.

PAGE 191

Investigations at Cave Branch Rock Shelter Table 1. Sex and age distribution of Caves Branch skeletal population through 2005 fieldseason. The mortuary component of Caves Branch Rockshelter is unusual for the ancient Maya, who in general are not thought to have used cemeteries until the Colonial era, instead burying their dead within residential structures and ceremonial locations (Welsh 1988, Whittington 1991: 172). While human remains are often found in caves, mortuary use of most caves, like ceremonial architecture, generally is limited to a few individuals based on their special social roles in life or their sacrificial roles in cave rituals (Minja res 2003, Gibbs 2000, Brady 1997, Brady et a. 1997). Despite these distinctions, however, the Caves Branch Rockshelter also is similar to the more traditional caves thus far documented by the WBRCP, since it is a ritual site incorporated into a natural rock feature located peripherally to a settlement, rather than a wholly constr ucted feature located within a settlement thus there are similarities in patterns of use. So, the changes in the use of rockshelters over time parallel those noted in dark zone caves (See Moyes and Prufer, this volume). The diagnostic vessels associ ated with graves at CBR all date to a similar time frame, likely sometime between 100 and 400 AD. However, the site was obviously visited long after the mortuary use of the site ceased. Like other cave sites, we find smashed pottery well into the Late Classic and even the Terminal Classic. 2. To explore other rockshelters in the area to determine inter-site variability. In addition to CBR, several other rockshelters have been located in the Caves Branch River Valley. The excavations at Deep Valley Rockshelter 1 showed that activities associated with rockshelters might have been as variable as is caves. Research planned for the 2007 field season will continue to document variability in rockshelter use by placing test pits in a series of other rock shelters to try to determine whether any factors such as time period, location, or morphology is related to material culture indicatin g particular uses of the sites. 3. To determine if the mortuary use of Caves Branch Rockshelter was limited in time. Thus far, the data seem to indicate that the CBR skeletons were relatively cotemporaneous. And all of the diagnostic vessels found in association with burials date to the Protoclassic period. Ceramic sherds from other time periods spanning the Middle Preclassic through the Terminal Classic are not associated with burials, but instead appear to have been broken and scattered on the surface. Some of these were found in deeper levels because of bioturbation and looting activity. The overlapping nature of the burials suggests that the site was used for burial over the B-2 2-6 6-18 30-40 40+ Adult Indet. Total % Male N/A N/A --1 3 1 --5 10.2 Female N/A N/A --2 4 5 --11 22.4 Subadult 7 7 1 --------15 30.6 Indeterminate ------1 3 14 --18 36.7 Total 7 7 1 4 10 20 0 49 100 Percent 14.2 14.2 2.0 8.1 20.4 40.8 0 100 193

PAGE 192

Wrobel, et al space of several generatio ns. Thus, relatively cotemporaneous, this could still mean that the population spanned several hundred years, which would not be ideal for a reference population. 4. To seek evidence for the presence of mortuary differentiation within this community, which would indicate that some individuals were buried elsewhere. The burials found in the last two seasons have revealed much more variation than reported by Bonor (1995a, 1995b, 1999). However, these variations are basically in the form of body position and orientation, rather than in elaboration. The possible exception to th is is the individual buried with the incised bone, carved with a woven mat motif. Again, though, based on the lack of other eviden ce of vertical social tiers, we agree with Bo nors assessment of the skeletal population as being a simple agricultural community. While its still certainly possible that some individuals may have been distinguished in death by their social role within the community, we see no evidence for it within the rockshelter. The nearby plazuela group is large enough to suggest an elite presen ce in this valley, but an initial probe into the eastern structure in 2005 revealed only Late Classic ceramics, suggesting that the mortuary use of the rockshelter predates at least this piece of evidence for social verticality. Research at the Caves Branch Rockshelter represents a dynamic research agenda. Future research initiatives will provide continued focus on the skeletal remains while embracing newfound research themes such as the potential archaic presence at the site and the changing role of the social and political organization in the Protoclassic/Early Classic transition. Information gained through the continued investigation of the CBR and surrounding areas will provide a wealth of information on topics that will be nefit all those engaged in Maya research. Acknowledgements : We would like to thank the following individuals and institutions for their support and assi stance: Dr. Jaime Awe, Dr. John Morris, Rafael Guerra, Sherilyne Jones, and the rest of the Belize Institute of Arch aeology; Christophe Helmke, Gwendolen Raley, Cameron Griffith, Dr. Lisa LeCount, Myka Schwanke, Jill Jordan, and Danielle Tanguis; Ian Anderson and all the staff at the Caves Branch Adventure Lodge. The University of Mississippi Office of Research and Sponsored Programs, and the UM Department of Sociology and Anthropology. References Cited Awe, Jaime J. 1998 The Western Belize Regional Cave Project: Objectives, Co ntext, and Problem Orientation In The Western Belize Regional Cave Project: A Report of the 1997 Field Season edited by J. J. Awe. Department of Anthropology Occasional Paper No. 1. University of New Hampshire, Durham. Bonor, Juan Luis 1995a Excavaciones de Salvamento en "Caves Branch Rock Shelter", Cayo District, Belize. Investigadores de la Cultura Maya 1:46-70. 1995b Trabajos Recientes en la Regin de Caves Branco, Distrito del Cayo, Belice. Investigadores de la Cultura Maya 4: 250267. Bonor Villarejo, Juan Luis, and David M. Glassman 1999 La arqueologa y el estudio de los restos humanos en Caves Branch Rock Shelter, Distrito de El Cayo, Belice. Mayab No. 12: 18-30. Brady, James E. 1997 Settlement Configuration and Cosmology: The Role of Caves at Dos Pilas. American Anthropologist 99 (3): 602-618. 194

PAGE 193

Investigations at Cave Branch Rock Shelter Brady, J. E., A. Scott A. Cobb, I. Rodas, J. Fogarty, and M. Urquiz 1997 Glimpses of the Dark Side of the Petexbatun Regional Archaeological Project: The Petexbatun Regional Cave Survey. Ancient Mesoamerica 8 (2):353364. Brown, L. 2000 From Discarded to Divination: Demarcating the Sacred through the Collection and Curation of Discarded Objects. Latin American Antiquity 11(4): 319-334. Cook, Della C, and Ellen Salter-Pedersen 2006a Pomacea snails and Classic Maya: archaeogastronomy? Paper presented at the annual meeting of the Society for American Archaeology, San Juan, Puerto Rico. Danforth, Marie Elaine, Gabriel Wrobel, David Swanson, and Carl Armstrong In press A Model Growth Curve for Juvenile Age Estimation Using Diaphyseal Long Bone Lengths among Ancient Maya Populations. To appear in Latin American Antiquity Friedel, David, Schele, Linda, and Joy Parker 1993 Maya Cosmos: Three Thousand Years on the Shamans Path. William Morrow: New York. Gibbs, S. A. 2000 An Interpretation of the Significance of Human Remains from the Caves of the Southern Maya Lowlands MA Thesis, Trent University, Peterborough. Gifford, J. C. 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Valley Memoirs of the Peabody Museum of Archaeology and Ethnology 18, Harvard University, Cambridge. Glassman, David M., and Juan Luis Bonor Villarejo 2005 Mortuary practices of the prehistoric Maya from Caves Branch Rock Shelter, Belize. In Stone Houses and Earth Lords: Maya Religion in the Cave Context edited by Keith M. Prufer and James E. Brady, pp. 285-296. University of Colorado Press: Boulder. Graham, Elizabeth, Logan McNatt, and Mark A. Gutchen 1980 Excavations at Footprint Cave, Caves Branch, Belize. Journal of Field Archaeology Vol. 7(2): 153-172. Grube, Nikolai, Thomas H. Guderjan, and Helen R. Haines 1995 Late Classic Architecture and Iconography at the Blue Creek Ruin, Belize Mexicon 17(3): 51-56. Halperin, Christina T., Sergio Garcia, Keith M. Prufer, and James E. Brady 2003 Caves and ancient Maya Ritual Use of Jute Latin American Antiquity 14(2): 207219. Lohse, J. C., J. Awe, C. Griffith, R. M. Rosenswig, and F. Valdez, Jr. 2006 Preceramic Occupations in Belize: Updating the Paleo-Indian and Archaic Record. Latin American Antiquity 17(2): 209-226. Michelon, Oscar (ed.) 1976 Diccionario de San Francisco (manuscript written before 1690). Bibliotheca Linguista Am ericana, Vol. II. Ada demische Druck-u. Verlagsanstalt, Graz. Minjares, A., Jr. 2003 Human Skeletal Remains of the Ancient Maya in the Caves of Dos Pilas, Guatemala MA Thesis, Texas A&M University, College Station. Pohl, Mary D., Pope, Kevin O., Jones, John G., Jacob, John S., Piperno, Dolores R., Defrance, Susan D., Lentz, David L., Gifford, John A., Danforth, Marie E., and J. Kathryn Josserand. 1996 Early Agriculture in the Maya Lowlands. Latin American Antiquity 7:355372. Prufer, Keith M. 2002 Communities, Caves, and Ritual Specialists: A Study of Sacred Space in the Maya Mountains of Southern Belize. Ph.D. Dissertation, Anthropology Program, Southern Illinois University, Carbondale. Reents, Dorie 1980 The Prehistoric Pottery from Petroglyph Cave, Caves Branch Valley, El Cayo 195

PAGE 194

Wrobel, et al 196 District, Belize, Central America M.A. Thesis, University of Texas, Austin. Welsh, W. B. M. 1988 An Analysis of Classic Lowland Maya Burials BAR International Series 409. British Archaeological Reports, Oxford. Whittington, S. L. 1991 Detection of Significant Demographic Differences between Subpopulations of Prehispanic Maya from Copan, Honduras, by Survival Analysis. American Journal of Physical Anthropology, Vol. 85: 167-184. Wrobel, Gabriel D., Marie E. Danforth, and Carl Armstrong 2002 Estimating Sex of Maya Skeletons by Discriminant Function Analysis of Longbone Measurements from the Protohistoric Maya Site of Tipu, Belize. Ancient Mesoamerica 13: 255-263. Wrobel, G. D. and J. Tyler 2006 Revisiting Caves Branch Rockshelter: Results of the 2005 Excavations. In The Belize Valley Archaeological Reconnaissance Project: A Report of the 2005 Field Season Edited by Christophe Helmke and Jaime Awe. Belize Institute of Archaeology, National Institute of Culture and History, Belmopan.

PAGE 195

18 FIFTEEN YEARS OF SE TTLEMENT PATTERN SURVEYS IN THE THREE RIVERS REGION OF NORTHWESTERN BELIZE: WHAT HAVE WE LEARNED? Hubert R. Robichaux This paper reviews the settlement survey work accomplishe d over the last 15 years in the Belizean portion of the Three Rivers Region. Settlements units are characterized and discussed in terms of cities, towns, smaller nucleations, and rural zones. Conclusions are reached concerning the size of urban zones, their population density, total population, and spatial organization. Although Maya peoples were present in the region during the Middle Preclassic, significant construction of communities generally did not begin until the Late Preclassic. Population and settlement reached its maximum during Late Classic times. The entire region was largely abandoned and uninhabited by the end of the Terminal Classic period. The accumulated data are informative about the ancient culture present in the region, suggesting among other things that a hierarchical, political and economic structure was present, and that militarism was a major element of Late Classic political strategies. Introduction In 1930, in British Honduras, as Belize was then called, virtually nothing was known about ancient Maya ruins in the northwestern portion of the country in the area now known as the Three Rivers Region. That situation started to change in 1931 with the arrival of a young Br itish archaeologist, J. Eric S. Thompson, who commenced his archaeological research in the region. This paper will attempt to summarize the results of the large amount of archaeological survey that has been conducted in that region since 1931, and will also outline some conclusions that can be drawn about ancient Maya settlement and culture, based upon this, and other data. The Three Rivers Region: The Three Rivers Region, as it is presently defined, includes contiguous areas of northeastern Guatemala and northwestern Belize (Figure 1). Almost certainly some sites located in the southeastern corner of the Mexican state of Campeche (Sprajc 2001) were politically associated with the large sites of Rio Azul and La Milpa in the Three Rivers Region, but the Mexican sites have thus far not been considered in the analysis of the region. In this paper, I will focus principally on the Belizean half of the region. The Rio Azul Archaeological Project (e.g., Adams 1999), and the Ixcanrio Regional Archaeological Project have studied the other areas in Guatemala (e.g., Adams and Valdez 2003). The Belizean portion of the region to day largely consists of three parcels of land (Figure 2). They are the Mennonite land to the north, the Programme-for-Belize (PfB) land in the center, and the Gallon Jug property to the south. The Three Rive rs Region derives its name from the presence of three small rivers, the Rio Azul, the Rio Bravo, and Booths River that traverse the area from southwest to northeast. Today the vast majority of the region is covered by dense jungle, and is largely devoid of human occupation. The Three Rivers Region contains a vast treasure of ancient Maya ruins that has only in recent years become fully apparent. The region has experienced a curious pattern of archaeological study wherein an Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 197-212. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 196

Hubert R. Robichaux Figure 1 Map showing the location of the Three Rivers Region of northwestern Belize and northeastern Guatemala Figure 2 Map of the Belizean portion of the Three Rivers Region showing the location of the Mennonite land, the Programme-for-Belize property, and the Gallon Jug property extended time gap exists between Thompsons initial forays into the region in the 1930s, and the follow-on research to be described here. In 1931 Thompson (1963:228-230) intended to study a small site he referred to as Kaxil Uinic (now thought to be either the site of Chan Chich (Guderjan et al 1991:59) or Ekenha (Houk 1998:5), but was dissuaded from doing that by the lack of a suitabl e labor force in that area. He settled instead upon the site of San Jose which is today located on the large 198

PAGE 197

Investigations in the Three Rivers Region, NW Belize Yalbac prop erty, at the southeast corner of the Three Rivers Region, just to the south of the Gallon Jug property. In 1931, 1934, and 1936 Thompson, with a small crew, mapped the San Jose site center and conducted excavations. In working at San Jose, Thompson made a noteworthy break from the then prevalent pattern in Maya archaeology of studying only the largest, most spectacular sites, which practice he rightfully thought gave a distorted view of ancient Maya life and culture. Later, in 1938, Thompson conducted a reconnaissance of the very large Three Rivers Region site of La Milpa (Thompson 1963:270). He made a sketch map of part of the La Milpa site center, discovered a number of stelae, and interpreted the discernable calendrics on them. For almost 50 years following Thompsons 1930s incursion into the Three Rivers Region, there were only incidental short visits into the region by archaeologists, including those of a number of government officials (Hammond 1991). One exception was Mary Neivens somewhat ill fated mapping efforts at the Blue Creek ruins on the Mennonite land in 1976. It was really not until 1988 that archaeological research resumed in earnest in the Three Rivers Region with reconnaissance and survey work conducted by the Maya Research Program under the direction of Thomas Guderjan (Guderjan 1991), and also by a contracted archaeological assessment of the La Milpa site by Anabel Ford and Scott Fedick (1988). Guderjans work, a little over 15 years ago represents the opening of the modern period of archaeological investigation in the Belizean portion of the Three Rivers Region. His program of reconnaissance, in which informants aided him, was important because it located, mapped, or otherwise described the site centers of 44 small to large sites whose locations stretched from the Mennonite land in the north to the Gallon Jug property in the south. Guderjans survey results provided a strong hint as to the archaeological richness present under the jungl es of the region. These discoveries stimulated the initiation of other projects that, with the approval of the Belize government, began to work in the Three Rivers Region. By way of summary, it can be said that most of the subsequent work in the region has been accomplished by the following Belize government-approved projects: work on the Mennonite land under permits issued to the Maya Research Program directed initially by Guderjan, and later by Jon Lohse; work on the Programme-for-Belize land under permits issued to the PfB Archaeological Project directed ini tially by Richard E.W. Adams and Fred Valdez, Jr., and subsequently by Valdez alone, and also by permits issued to the La Milpa Archaeological Project directed by Norman Hammond and Gair Tourtellot for the specific study of the 6 km radius area surrounding the La Milpa site center; work on the Gallon Jug property has been accomplished under permits issued to the Center for Maya Studi es directed by Brett Houk for study of the Chan Chich site, and, finally, under permits issued to the Punta de Cacao Archaeological Pr oject, directed by the author, for the st udy of the Punta de Cacao site. These all are, or were, multi-year projects, with the PfB Archaeological Project being the longe st lived, extending 15 years, from 1992 until the present. The PfB project also, perhaps, has had the most diversity in the conduct of its research. The work of many archaeologists and specialists from many universities has been conducted on the PfB land through the medium of subprojects subsumed under the main PfB Projects permit, resulting in the accumulation of a very rich data base dealing with virtually all areas of the vast 199

PAGE 198

Hubert R. Robichaux PfB property (for exam ple, see Valdez 2005). The PfB project has discovered a plethora of sites previously unknown to archaeologists, including large ones such as Dos Hombres, Gran Cacao, Maax Na, and Great Savannah. The research results from these projects with regard to settlement patterns in the Three Rivers will now be considered. From this data, and the data from elsewhere, we can say that the ancient lowland Maya population was not evenly distributed across the landscape. Instead, their population had a definite tendency to cluster at certain places. Some of these clusters, or nucleations of populatio n, were larger than others. There was, in fact, something of a continuum in the size of these population nucleations, extending downward from the largest to the smallest (see, for example, Adams et al 2004). Archaeologists have persistently used vague or neutral terms, such as site, minor center, major center, ceremonial center, district, etc., to label these ancient Maya population nucleation. This has been so due to the difficulty of placing any given cluster of settlement into its larger settlement context because of the low visibility in the jungl e environment widely present in the Maya lowlands. However, due to the increasing availability of survey data over extended areas in the Three Rivers Region and elsewhere, it has by now become evident that what we really have are cities, towns, villages, hamlets, i.e., the usual sizeand function-based typology used cross-culturally, across the globe, to describe human settlement. The settlement zones between the various nucleated areas are rural in nature. Maya urban areas, while not as densely packed as those in many other cultures, nonetheless contain settlement at a much greater density than that in the surrounding rural zones. In the following consideration of the settlement studies that have occurred in the Belizean portion of the Three Rivers Region in the last 15 years, I will frame my discussion in terms of cities, towns, smaller nucleations, and rural areas. Cities There was only one city in the Belizean half of the Three Rivers Region, and that was La Milpa on the PfB property. The center of this large city has been mapped successively by Thompson (1939), Ford and Fedick (1988), Guderjan (1991), and Tourtellot (e.g., Tourtellot et al 2003). Mapping of various segments of the outer portions of the city have also been accomplished independently by Tourtellot (e.g. Tourtellot et al 2003), Rose (2000), Robichaux (1995), Scarborough (1994), Kunen (Kunen and Hughbanks 2003), and others. At least nine sq. kilometers of this ancient city have been mapped. What emerges from this resear ch is that at the time of its maximum size, during Late Classic 2, and Terminal Classic times, (roughly A.D. 730-900), La Milpa was a large urban area, a city, having a radius of ca. 5 km, and it had an estimated population density in the range of 730-900 persons per sq. km living within inhabitable areas (Robichaux 1995:285, 290; Hammond et al 1996:86; Tourtellot et al 2003). Using this data and subtracting out areas of uninhabitable te rrain within the five km radius area such as that of seasonally inundated bajos one is led to the conclusion that La Milpa had a total population of over 45,000 inhabitants. The city center of La Milpa, hereinafter referred to as the central precinct had an area of ca. 0.45 sq. km. The central precinct contained the citys largest buildings, housed many of its most important people, and was the scene of important political, religious, and economic activities. However, it is clear that the vast majority of the population of La Milpa lived 200

PAGE 199

Investigations in the Three Rivers Region, NW Belize not in the ce ntral precin ct, but in the balance of the community that encompassed out to ca. five km away. I refer to this larger area as the outer-city (or outer-town as the case may be). It is furthermore clear that imbedded at different locations within that five km radius area of La Milpa (Figure 3) were various nuclei of elite settlement such as Thompsons Group (Robichaux 1995), and Say Ka (Guderjan 1991; Houk and Lyndon 2005; Houk et al 2006). Without the benefit of the extensive mapping that has occurred at La Milpa, these nuclei would have been misinterpreted as minor centers, or satellites of La Milpa, and considered separate settlements, distin ct from La Milpa. These scattered elite nuclei within La Milpa seem, in Late Classic 2 and Terminal Classic times at least, to have functioned as overseers of particularly productive environmental resource zones present within the city. Undoubtedly these elite in the outer-city linked in some way to the citys rulership, and helped integrate the large population politically and economically. Beyond the citys ca. five km radius lay an area of reduced aver age population density, in the range of 175-200 persons per sq. km. (Robichaux 1995:259-260; Hammond et al 1996:86), which is considered rural in nature. Excavations at various locations within the city of La Milpa indicate the first substantial human settlement in its immediate area occurred in the Late Preclassic period. Stel ae citing the deeds of local kings began appearing at La Milpa during the Early Classic period as the community grew (Tourtellot et al 1994:122), and the city expanded outward to reach its maximum size, population, and complexity during the Late Cla ssic period. Like virtually all sites in the Three Rivers Region, La Milpa was abandoned and largely uninhabited by A.D. 900. Below La Milpa in the settlement hierarchy in the Belizean portion of the Three Rivers Region are at least nine sites that seem to be town-sized places, some larger, some smaller. Figure 3 Map of the Belizean portion of the Three Rivers Region showing the locations of La Milpa, and the ancient towns discussed in the text Towns These are sizable communities that would have been home to several thousand persons during Late Classic 2 and Terminal Classic times. They include Blue Creek on the Mennonite land, Gran Cacao, Great Savannah, Maax Na, and Dos Hombres on the PfB land, Chan Chich, Punta de Cacao, and Quam Hill on the Gallon Jug property and San Jose on property to the south of the Gallon Jug property. All of these towns have central precincts containing a significant amount of monumental architecture, incl uding public-function plazas, and sizable elite residential enclaves. Each has one or more ballcourts, and each has at least one stela present, with the exception of Quam Hill, where one or two stelae appear to have been looted (Guderjan 1991). These town-sized places played a significant role in the settlement hierarchy of the region, serving as economic, political, and administrative intermediaries between 201

PAGE 200

Hubert R. Robichaux cities above them and smaller settlement units below. Wari Camp (Guderjan 1991; Levi 1997, 2003), on the PfB land, may also be another town-sized site, but that determination awaits the publication of more data on it. Also, El Infierno (Guderjan 1991), on the Gallon Jug property near the Guatemala border, is possibly another town but there is too little data to confirm that now. The town-sized sites will now be discussed, beginning with Blue Creek. Blue Creek Mapped in various efforts by Nievens (1991), Guderjan (Guderjan et al 2003), and Lohse (1995), Blue Creek is perhaps the most thorou ghly studied site in the region. Approximately 20 sq. km. surrounding the central precinct have been mapped over a 10-year period (Guderjan et al 2003:77), making it the most completely mapped site in the entire Three Rivers Region. A large amount of excavation in various parts of the town has occurred. This ancient town was mostly situated on high ground at the upper edge of an escarpment that overlooks the coastal plain of Belize to it east. Blue Creek is located near the point where the Rio Bravo and Booths River join to form the Rio Hondo. Apparently a trading center dealing both with sites below it on the coastal plain, and other sites atop the escarpment, Blue Creek may have been wealthier than most other towns in the region. The site map (Guderjan et al 2003: 79) of Blue Creek shows that the town had component parts both on top of the escarpment and just below it on the coastal plain. In apparent contrast to Dos Hombres, which is discussed below, the Blue Creek map does not reveal much settlement on the intervening, steeply sloping escarpment itself. The settlement within Blue Creek tended to be in a pa tchy pattern, having various residential barrios (Guderjan 2003:82), with lightly inhabited zones being present between some of the barrios. This may be somewhat illusionary, however, because a part of the forest has been cleared at Blue Creek, making it possible that some structures within these lightly inhabited buffer zones were destroyed during the land clearing process. Examination of the site map indicates these barrios, or nuclei, are all located within a ca. four km radius of the central precinct, and I take that to be the approximate radius of the ancient town. These barrios appear, in some cases, to represent different social status groups, or mixtures of different social status groups, located in differing environmental zones. Guderjan (2003:79) i ndicates that some agricultural fields were present within the town limits. I have not found a published estimated population density for Blue Creek, and therefore cannot estimate its population. Nonetheless, Blue Creek seems to have been one of the largest towns in the Three Rivers Region. It should be noted that construction at Blue Creek, in contra st with that at most sites in the region, seems to have peaked out during the Early Classic rather than the Late/Terminal Classic (Guderjan et al 2003:83), although the town continued to be occupied during those later periods. Great Savannah Very little data is available on this site, but it has been estimated to be similar in size to the better studied Gran Cacao (Adams 1995:5). Gran Cacao The only mapping done at this site has been in its central precinct. The size and complexity of the central precinct (Lohse 1995) indicates that Gran Cacao was a town-sized place, but we have no information on its radius, or population density. Ceramics recovered by Levi 202

PAGE 201

Investigations in the Three Rivers Region, NW Belize (1994), Lohse (1995) and others in the centr al precinct indicate the site was occupied from Late Preclassic times until some point in the Term inal Classic. Lohse has recently conducted additional excavations at Grand Cacao in the ball court area (Lohse personal communication 2005). Msax Na This appears to be a town-sized place, possibly a little smaller than most of the other towns in the region. The presently available map of the site (King and Shaw 2003, 2006; Shaw et al 2005) covers mainly the central precinct, and there is little knowledge of the towns outer areas, its spatial size, or its population density. Dos Hombres An approximately 6 sq. km. area of Dos Hombres has been mapped through the efforts of various archaeologists. The central precinct of this town was mapped by Houk (Houk 1996), and others. Various parcels of the outer portion of th e ancient town have also been mapped independently by various archaeologists includ ing Robichaux (1995), Lohse (2001), Hageman (Hageman and Lohse 2003), Walling (e.g., 2004, 2006) and Glabb (Glabb and Taylor 2005). Based upon my surveys on top of the escarpment, to the west of the Dos Hombres central precinct, I estimated that Dos Hombres had a radius of approximately 3.5 km (Robichaux 1995:290). The steep Rio Bravo escarpment runs roughly north south through the town, ca. 1.5 km west of the central precinct. Studies by Stan Walling (2004) on the steeply inclined escarpment face show that it was densely settled during the Late Classic period, when dense overall settlement at Dos Hombres may have provoked people to occupy more difficult terrain zones. The settlement present on the escarpment continued outward to the west, on top of the escarpment in the area I mapped, resulting in my estimate of a 3.5 km radius for the town. The estimated average population density in my surveys at Dos Hombres was ca. 500 persons per sq. km. Walling (2004:6) estimates the population of the ca. 0.7 sq. km. area he mapped and excavated on the escarpment as being ca. 850 inhabitants. This would appear to generate a density of about 1,200 persons per sq. km for this location, which he calls Chawak Butoob. I have not found any other estimates of population density for Dos Hombres, but Lohses map (Lohse 2001) of a 2.5 km long transect to the east of the central precinct shows heavy settlement in the outer 1.75 km of that transect. Settlement, overall, at Dos Hombres does have a somewhat patchy pattern, with, for exampl e, little settlement being present within the broad, north-south trending flood plain of the Rio Bravo river which passes below the escarpment, within a kilometer west of the central precinct. A 3.5 km radius generates a total area of ca. 38 sq. km. for the town of Dos Hombres. If one estimates that 40% of that area was uninhabitable then that leaves (river flood plain/bajos/marsh/etc) about 23 sq. km of inhabitable land. If we use an average population density for that area of 500 persons per sq. km, we arrive at an estimated total population of 11,400 inhabitants for Dos Hombres during Late Classic 2 times. Excavations show the Maya first settled the Dos Hombres area during the Middle Preclassic, but significant construction did not begin there until the Late Preclassic times. Construction and population appear to have reached their maximum at Dos Hombres during Late Classic 2 and Terminal Classic times. It was abandoned by the end of the Terminal Classic period. Chan Chich 203

PAGE 202

Hubert R. Robichaux A ca. 2sq. km area of Chan Chich has been mapped, most of it being in the towns central precinct (Houk and Robichaux 1996; Houk 1998, 2000). Thus we have no knowledge of the towns radius, or its average population density, and overall population. Construction began at Chan Chich in the Middle Preclassic at the location that later became the center of the town. The Late Preclassic and the Late Classic appear to have been the periods when most construction occurred at Chan Chich. It was abandoned by the end of the Terminal Classic period Punta de Cacao A continuous 3.9 sq. km area extending outward from the central precinct of Punta de Cacao has been mapped (Guderjan 1991; Robichaux et al 2002; Robichaux and Miller 2003; Robichaux 2005a, 2005b). The mapped zone extending to the west of the central precinct suggests this ancient town may have had a radius of ca. 2.2 kilometers. An estimated average population density of 401 persons per sq. km. for the town has been computed, indicating that Punta de Cacao had a total population in the vicinity of 6,000 persons during the Late Classic 2 period, the apparent time of its maximum size and population. Maya peoples were present near the center of the later town during Middle Preclassic times. Most of the construction in the central precinct of the site took place during the Late Preclassic and Late/Terminal Classic periods. Preliminary indications are that the outertown was largely constructed and occupied in Late Classic 2 and Terminal Classic times. Punta de Cacao was abandoned by the end of the Terminal Classic. Quam Hill This site was located near the property line separati ng the PfB and Gallon Jug properties. Based upon Guderjans map and comments concerning its central precinct (Guderjan 1991), Quam Hill may be a sizable town. Other than the mapping of part of the central precinct, no research has been done at Quam Hill, and we know nothing of its radius, or average population density. San Jose Thompson (1939) limited his mapping and excavations at this site solely to to the central precinct, as was the custom then. Based upon his map, San Jose appears to have been a small town-sized site. We have no information on its radius, population density, etc. Some of the construction of large st ructures at San Jose occurred during Late Classic 2 and Terminal Classic times. While varying in size, towns-sized places had several thousand inhabitants living in them during Late Classic 2 and Terminal Classic time, and they helped provide an organizational framework for populations living in the region. Also present in the Three Rivers Region were nucleation of population that were smaller than towns. These will be considered next. Smaller Nucleations These were separate communities of smaller size, below the status of towns, where the total population probably amounted to less than 1000 persons. Based upon their apparent size we can label them villages, however, their occupants were less egalitarian than those of villages in many other cultures. Among their number are Las Abejas (Sullivan 1997) and Dos Barbaras (Lewis 2003) on the PfB property, and the Ekenha, Gallon Jug, and Laguna Seca sites on the Gallon Jug property (Guderjan 1991). There are probably many more of these 204

PAGE 203

Investigations in the Three Rivers Region, NW Belize sm aller nucleations remaining to be discovered. They seem to consist of a few larger, upper social class type structures, with surrounding smaller residences that, to this point, have not been well mapped. These smaller nucleations are located beyond the boundaries of larger cities and towns, and appear to be distinct from rural areas. Unfortunately, the general lack of extensive mapping around any of these sites impedes further description. Note that Thompsons Group and Say Ka, cited earlier, are not included here as smaller nucleations since they are embedded within the La Milpa outer-city zone and were therefore components of the city of La Milpa, at least during Late Classic 2 and Terminal Classic times. Rural Areas As has already been noted, rural areas have a lower pop ulation density, and are present beyond and away from population nucleations, or urban zones. Data from various regions, including the Three Rivers Region, suggest that the rural ancient Maya populations had a density in the approximate range of 150-200 persons per sq. km during Classic times (Rice and Culbert 1990:26; Culbert et al 1990:116; Robichaux 1995:186, 222, 259-260; Hammond et al 1996:86). The inhabitants of rural areas can probably be rightly considered as country folk, having a somewhat different culture than urban Maya, being more independent, and less accountable to the pr evailing political authority than people within the urban areas. Rural settlement has been mapped on the immediate outskirts of La Milpa and Dos Hombres, and at other locations within the PfB land. In particular, Sunahara and Meadows (2005) have mapped and excavated in largely rural areas beginning directly to the west of Dos Hombres and extending all the way to the Guatemala border. Reese-Taylor has mapped and excavated in rural areas to the west of Maax Na (Reese-Taylor 2003), and Hughbanks (Kunen and Hughbanks 2003) has mapped an apparently rural ar ea he calls Guijarral located to the east of La Milpa. El Intruso (Brandon Lewis, personal communication 1992; Rene Munoz, personal communication 1994), Cabeza Verde (Lewis 2003), and the Medicinal Trail Site (Hyde 2005; Hyde et al 2006), in the same general area, also appear to be in a rural context. Hageman has mapped rural areas along the survey transect that was mapped between Dos Hombres and La Milpa (Hageman and Lohse 2003), and Glabb (Glabb and Taylor 2005) has commenced mapping of rural areas along a survey transect between Dos Hombres and Gran Cacao. The data from these efforts, as it is currently published, show that settlement in rural areas is not uniform, nor evenly distributed, with hamlets being present here and there, and isolated farmsteads elsewhere. Some environmentally difficult areas are largely uninhabited. A variety of house types, house sizes, spatial arrangements, and social classes are found in rural areas, but the overall structural and population density is characteristically low. Observations Surveys in the modern period of archaeological research in the Three Rivers Region have produced a large amount of data that illuminate many aspects of ancient Maya settlement and culture. The data indicate Maya peoples first migrated into the region during Middle Preclassic times. These pioneering farming peoples, having little competition, settled where they perceived to be the best places. They succeeded, grew in population, and spread out spatially. Gradually, with time, they adapted to many different environmental 205

PAGE 204

Hubert R. Robichaux niches and came to fill much of the landscape, including some less than optimum locations. The appearance of communities of greater than village size seems to have widely begun during the Late Preclassic period. Many larger places got their start then. Most communities in the region seem to have had a lower level of growth during the Early Classic period, and also at the beginning of the Late Classic period, but this may be partly an aberration due to confusion rela ting to the dating of certain ceramic types (Fred Valdez, Jr., Lauren Sullivan, Kerr y Sagebiel, personal communications 2005). The Three Rivers Region reached its maximum population and settlement size, as well as it highest political and economic complexity during Late Classic 2 and Terminal Classic times. This apogee lasted less than 150 years. With almost no exceptions, the cities, towns, villages, hamlets, and rural areas in the Three Rivers Region became uninhabited by ca. A.D. 900, and the region has largely remained so until today. Our best knowledge of what towns and cities were like comes from the Late Classic 2, and Terminal Classic periods because most of the ruins visible on the surface at most of the sites in the region date to that time. Thus, the spatial sizes and populations densities we compute for La Milpa, Dos Hombres, Punta de Cacao, and other urban areas are mostly applicable to that time. We can state with confidence that ancient Maya cities and towns existed, and that individually they extended out over rather large areas. In the Three Rivers Region of Belize, La Milpa was the largest urban place, with a radius of ca. 5 kms. Blue Creek and Dos Hombres may have been the largest towns. By way of comparison with other regions, the site of Caracol in west, central Belize seems to have extended outward to ca. 7 km from its central precinct (Chase and Chase 1994:2), and Tikal in the Peten of Guatemala, appears to have had a radius of ca. 6 km (Culbert et al 1990). Figure 4 is a map showing the city of La Milpa and the towns present in the Belizean portion of the Three Rivers Region. For each, a circle has been drawn around its central precin ct with the radius based either on survey data (La Milpa, Blue Creek, Dos Hombres, and Punta de Cacao), or with an assumed radius of 2.2 km, for all other towns, based upon the apparent approximate radius of Punta de Cacao, a smaller town. This map shows that a sizable portion of the Three Rivers Region was urban in character. Although an idealized circular form has been used to depict urban areas, it is recognized that the shape of a city or town was frequently irregular, and not circular as shown, due to environmental or other reasons. Figure 4. Schematic map of the ancient city of La Milpa showing the central precinct and outer-city potions of the site, as well as the locations of two elite nuclei (Thompsons Group and Say Ka) situated within the outer city. The La Milpa city extends outward up to the 5 km radius circle shown. The surrounding rural zone is also shown. 206

PAGE 205

Investigations in the Three Rivers Region, NW Belize The population density within urban areas in the Three Rivers Regions during Late Classic 2 times varied within the approximate range of 400-800 persons per sq. km, with the most densely settled place evidently being La Milpa. Cities and towns can be characterized as having had two principal components, the central precinct, and the balance of the community, i.e., the outercity, or outer-town, where most of the population resided. Th ere seems to have been a rough correlation between the size of a citys central precinct and the size of its outer-city. Based upon the data from within the Three Rivers Region, and from elsewhere in the Maya lowlands (e.g., Tikal and Caracol), we know that the outer-city had within it, at various distances from the central precinct, elite settlement nucleations which likely oversaw activities in their section of the city on behalf of the polity rulership. I interpret this pattern of ancient Maya urban settlement to be most consistent with what geographers have labeled the multi-nuclei model of urban settlement (Harris and Ullman 194 5), wherein the city is divided into a number of functional areas, or nuclei, that are integrated in some way to promote the citys functioning and survival. These various nuclei within the community urban area may serve political, economic, social, religious, military (e.g. defensive), water-management, administrative, or other functions. These various nuclei located throughout the city were physically integrated through mediums such as elevated causeways, customary paths or walkways, natural waterways, artificial waterways, etc. This model can be especially helpful in sorting out the dynamics and functioning of an ancient Maya community. Political relationships and linkages between cities and towns within the Three Rivers Region has been discussed elsewhere (e.g., Adams 1999; Robichaux 2000, 2002, 2004; Adams et al 2004), and will only be briefly considered here. Notwithstanding frequent arguments to the contrary, the totality of data from the region, and elsewhere, including settlement analysis, epigraphy, iconography, etc., strongly indicates that a ne sted, hierarchical relationship between cities, towns and smaller nucleations prevailed. Clearly, in the Belize half of the Three Rivers Region, La Milpa was the dominant place during Late Classic 2 times, and it must have dominated most of th e nearer towns, and lower levels of settlement. The people with the greatest freedom from the hierarchical framework were probably rural farmers located away from the urban areas. Epigraphic data from various Maya regions indicate that a political ordering of the lords who ruled the various Maya cities, towns, and small places existed. In the Three Rivers region, a plate fragment found at Dos Hombres has the u-yahaw glyph as part of its text (Robichaux and Houk 2005). It indicates the subordina tion of a lord at one location to the lord at another location. The texts and/or scenes depicted on a large number of stelae throughout the Maya lowlands, including La Milpa, indicate that militarism was a strong component of the Classic period Maya culture. Kings armed with spear and shield is a common motif. Kings frequently are shown standing on prisoners from other sites. Texts proclaim victories over other loca tions. Painted pots have court scenes depicting scowling rulers glaring down at tied captives who are guarded by armed persons. Some sculptural and painted vessel scenes show lines of warriors armed with spears, all identically dressed, i.e., wearing uni forms. Other pots and murals show battles fully engaged. Other media, including common graffiti, show prisoners undergoing various forms of 207

PAGE 206

Hubert R. Robichaux torture and physical a buse. It is, thus, beyond doubt that fo rce was a frequent instrument of ancient Maya political policy. Such force, controlled by the ruler of a city, or larger polity, was certainly applied externally against other polities, and it was also available as well to be used internally within a city or larger polity. Force intimidates those who have less. It is, in general, very unlikely that smaller places had more military power than larger places these settlement units. The collapse of Maya civilization in the Three Rivers Region was complete, resulting in the region becoming virtually uninhabited by humans. It seems likely that the data already in hand must shed some direct light on the process by which this dramatic collapse occurred, but the task of extract ing and distilling that information largely remains to be done. Although the Three Rivers Region has become one of the more intensely studied regions in th e Maya lowlands, the data accumulated there has yet to be adequately disseminated to, and absorbed by the larger community of scholars studying the ancient Maya. For instance, a recent publication summarizing what is known about the ancient Maya (Demarest 2004) appears not to mention the Three Rivers Region explicitly, nor even the large city of La Milpa. Nonetheless, the intense level of archaeological research that has taken place in the Belizean half of the Three Rivers Region during the last 15 years has truly produced, and continues to produce, a treasure of data, which is now becoming available to scholars. Acknowledgements I would like to express my sincere thanks to Drs. Jaime Awe and John Morris, and their staff at the Belize Governments Institute of Archaeology, for their support of our work at Punta de Cacao and for the opportunity to participate in the Belize Archaeology Symposium 2005. Barry Bowen, the Chan Chich Lodge, and many financial supporters in the United States have made our research at Punta de Cacao a practical reality, and I am very appreciative of their generosity. I would like to thank archaeologists Drs. Richard E.W. Adams, Fred Valdez, Jr., Brett Houk, William Folan, Joel Gunn, and Thomas Guderjan for their generous support over many years. Fred Valdez was particularly generous in providing data for this paper concerning survey work on the PfB land. Gair Tourtellot has also been helpful in sharing information about the ancient city of La Milpa. Kay Sunahara and Richard Meadows provided me a copy of their paper describing their surveys and excavations to the west of Dos Hombres. I owe a great deal to the University of the Incarnate Word for their overwhelming support of the Punta de Cacao project in Belize. Lastly, I thank my wife, Cathy, and Jenny, and my daughters, Marlene and Jackie, and their families, for their support. Reference Cited Adams, Richard E.W. 1995 Introduction. In The Programme for Belize Archaeological Project: 1994 Interim Report. The 1994 Interim Report edited by R.E.W. Adams and Fred Valdez, Jr., pp1-14, The Center for Archaeology and Tropical Studies and the University of Texas at San Antonio 1999 Rio Azul, An Ancient Maya City University of Oklahoma Press, Norman Adams, Richard E.W., and Fr ed Valdez, Jr. (editors) 2003 Ixcanrio Research Reports 1 & 2: The 1990 and 1991 Seasons The University of Texas at San Antonio Adams, Richard E.W., Hubert R. Robichaux, Fred Valdez, Jr., Brett A. Houk, and Ruth Mathews 2004 Transformations, Periodicity, and Urban Development in the Three Rivers Region. In The Terminal Classic in the Maya 208

PAGE 207

Investigations in the Three Rivers Region, NW Belize Lowlands: Collapse, Transition, and Transformation edited by Arthur A. Demarest, Prudence M. Rice, and Don S. Rice. University Press of Colorado, Boulder Chase, Diane Z., and Arlen F. Chase 1994 Details in the Archaeology of Caracol, Belize: An Introduction. In Studies in the Archaeology of Caracol, Belize edited by D. Chase and A.Chase, pp 1-11. PreColumbian Art Research Institute, Monograph 7, San Francisco, California Culbert, T. Patrick, Laura J. Kosakowsky, Robert E. Fry, and William A. Haviland 1990 The Population of Tikal, Guatemala. In Pre-Columbian Population History in the Maya Lowlands edited by T. Patrick Culbert and Don Rice, pp.103-122. University of New Mexico Press, Albuquerque Demarest, Arthur 2004 Ancient Maya: The Rise and Fall of a Rainforest Civilization. Cambridge University Press Ford, Anabel, and Scott Fedick 1988 Draft Report on the Archaeological Resource Potential and Management of the Programme for Belize Lands, Orange Walk, Belize. Mesoamerican Research Center. Social Process Research Institute. University of California, Santa Barbara Glaab, Rigden A., and Chrissy Taylor 2005 Dos Hombres-Gran Cacao Intersite Research: The 2004 Mapping Season. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp 35-44, Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Guderjan, Thomas H. (editor) 1991 Maya Settlement in Northwestern Belize: The 1988 and 1990 Seasons of the Ro Bravo Archaeological Project Maya Research Program, San Antonio, Texas, and Labyrinthos, Culver City, California 2004 The Spatial Arrangement of a Maya City Document submitted to Geographic Review. Department of Sociology, CJ, and Anthropology, TCU, Ft. Worth, Texas Guderjan, Thomas H., Jeffrey Baker, and Robert Lichtenstein 2003 Environmental and Cultural Diversity at Blue Creek. In Heterarchy, Political Economy, and the Ancient Maya edited by V. Scarborough, and F. Valdez, Jr., pp 7791, The University of Arizona Press, Tucson Hageman, Jon B., and Jon C. Lohse 2003 Heterarchy, Corporate Groups, and Late Classic Resource Management in Northwestern Belize. In Heterarchy, Political Economy Economy, and the Ancient Maya edited by V. Scarborough and F. Valdez, Jr., pp 109-121, The University of Arizona, Tucson Hammond, Norman 1991 The Discovery of La Milpa Mexicon 13: 46-50 Hammond, Norman, Gair Tourtellot, Sara Donaghey, and Amanda Clarke 1996 Survey and Excavation at La Milpa, Belize, 1996. Mexicon 18(5): 86-90 Harris, Chauncy, and Edward Ullman 1945 The Nature of Cities. Annals of the American Academy of Political and Social Sciences 242 :7-17 Houk, Brett A. (editor) 1998 The 1997 Season of the Chan Chich Archaeological Project Papers of the Chan Chich Archeological Project, Number 3. Center for Maya Studies, San Antonio, Texas 2000 The 1998 and 1999 Seasons of the Chan Chich Archaeological Project Papers of the Chan Chich Archaeological Project, Number 4, Mesoamerican Archaeological Research Laboratory. The University of Texas at Austin Houk, Brett A., and Hubert R. Robichaux (editors) 1996 The 1996 Season of the Chan Chich Archaeological Project Papers of the Chan Chich Archaeological Pr oject, Number 1. Center for Maya Studies, San Antonio, Texas Houk, Brett A., and Michael Lyndon 209

PAGE 208

Hubert R. Robichaux 2005 The 2004 Investigations at Say Ka: A Pilot Project. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp 45-62, Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Houk, Brett A., Rebecca E. Bria, and Shelly Fischbeck 2006 The 2005 Investigations at Say Kah, Belize. In Programme for Belize Archaeological Project: Report of Activities from the 2005 Field Season edited by Fred Valdez, Jr., pp. 17-40. Occasional Papers, Number 6 Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Hyde, David M. 2005 Excavations at the Medicinal Trail Site: 2004. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp. 7-14. Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas Hyde, David M., Shelly Fishbeck, and Rissa Trachman 2006 Report of Excavations at the Medicinal Trail Site for the 2005 Season. In Programme for Belize Archaeological Project: Report of Activities from the 2005 Field Season edited by Fred Valdez, Jr., pp. 7-16. Occasional Papers, Number 6, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin King, Eleanor M., and Leslie C. Shaw 2003 A Heterarchical Approach to Site Variability: The Maax Na Archaeological Project. In Heterarchy, Political Economy, and the Ancient Maya edited by V. Scarborough and F. Valdez, Jr., pp 64-76, The University of Arizona Press, Tucson 2006 Preliminary Report of Fieldwork at Maax Na, Belize, 2005. In Programme for Belize Archaeological Project: Report of Activities from the 2005 Field Season edited by Fred Valdez, Jr., pp. 89-98. Occasional Papers, Number 6, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Kunen, Julie L, and Paul J. Hughbanks 2003 Bajo Communities as Resource Specialists: A Heterarchical Approach to Socioeconomic Organization. In Heterarchy, Political Economy, and the Ancient Maya edited by V. Scarborough and F. Valdez, Jr, pp 92-108, The University of Arizona Press, Tucson Lewis, Brandon 2003 Environmental Heterogeneity and Occupational Specialization: An Examination of Lithic Tool Production in the Three Rivers Region of the Northeastern Peten. In Heterarchy, Political Economy, and the Ancient Maya edited by Vernon L. Scarborough, Fred Valdez, Jr., and Nicholas Dunning, pp. 122-135. The University of Arizona Press, Tucson 2005 Research at Dos Barbaras: An Overview. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp145-158, Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Levi, Laura J. 1995 Site Map, Evaluation of Salvaged Ceramics, and Artifact Storage Data; Site RB-43, Programme for Belize Lands. In The Programme for Belize Archeological Project: 1993 Field Season edited by R.E.W. Adams. The University of Texas, San Antonio 1 997 Statement of Findings of the 1997 PfB Regional Residential Survey Unpublished manuscript in the possession of the author. University of Texas at San Antonio 2003 The PfBARP Regional Residential Survey: Project Summary and Report of Fieldwork in 2002 Unpublished manuscript in the possession of the author. University of Texas at San Antonio Lohse, Jon C. 1995 Results of Survey and Mapping during the 1994 PfB Season at Gran Cacao. In The Programme for Belize Archaeological 210

PAGE 209

Investigations in the Three Rivers Region, NW Belize Project: 1994 Interim Report edited by R.E.W. Adams and Fred Valdez, Jr., pp. 106-114. The Center for Archaeology and Tropical Studies and The University of Texas at San Antonio 2001 The Social Organization of A Late Classic Maya Commun ity: Dos Hombres, Northwestern Belize Ph.D. dissertation, The University of Texas at Austin Marcus, Joyce 1983 On the Nature of the Mesoamerican City. In Prehistoric Settlement Patterns edited by E. Vogt and R. Leventhal, pp. 195242, University of New Mexico Press, Albuquerque Montgomery, John 2002 How to Read Maya Hieroglyphs Hippo crme Books, Inc., New York, NY Reese-Taylor, Kathy 2003 Western Escarpment Survey and Archaeological Project Unpublished manuscript in the possession of the author. Rice, Don S., and T. Patrick Culbert 1990 Historical Contexts for Population Reconstruction in the Maya Lowlands University of New Mexico Press, Albuquerque Robichaux, Hubert R. 1995 Ancient Maya Community Patterns in Northwestern Belize: Peripheral Zone Survey at La Milpa and Dos Hombres Unpublished Doctoral Dissertation, The University of Texas at Austin 2000 The Stelae of Rio Azul, Guatemala. In Rio Azul Reports, Number 5, The 1987 Season edited by R.E.W. Adams, pp 35-53. The University of Texas at San Antonio 2002 Aspectos de la Organizacin de Asentamiento de los Mayas Antiguos en la Regin de los Tres Ros en Guatemala Noreste y Belice Noroeste: Una Vista de la Tercera Fila de la Jerarqua de Asentamiento. In Los Investigadores de la Cultura Maya 11 Tomo 1, pp. 168-179, Universidad Autnoma de Campeche, Campeche, Mxico 2004 The Ancient Community of Punta de Cacao, Belize, and Its Place in the Prehistory and History of the Three Rivers Region: The 2003 Field Season of Punta de Cacao Archaeology Project. Research Reports in Belizean Archaeology Vol. 1: pp.197-209 Robichaux, Hubert R. (editor) 2005a Punta de Cacao Archaeological Project: An Interim Report on the 2003 Field Season University of the Incarnate Word, San Antonio, Texas 2005b Punta de Cacao Archaeological Project: a Preliminary Report on the 2004 Field Season University of the Incarnate Word, San Antonio, Texas Robichaux, Hubert R., Alexandra Miller, Evans Roberts III, Candace Pruett, Benjamin Smyth, E. Thomas OBrien, Fred Valdez, Jr., and Robynne R. Valdez 2002 Preliminary Report: The 2001 Season of the Punta de Cacao Archaeological Project University of the Incarnate Word, San Antonio, Texas Robichaux, Hubert R., and Alexandra Miller (editors) 2003 Punta de Cacao Archaeological Project: Interim Report on the 2002 Field S eason University of the Incarnate Word, San Antonio, Texas Robichaux, Hubert R., and Brett A. Houk 2005 A Hieroglyphic Plate Fragment from Dos Hombres, Belize: Epigraphic and Archaeological Evidence Relating to Political Organization in the Three Rivers Region of Northwestern Belize and Northeastern Guatemala Mono y Conejo, Journal of the Mesoamer ican Archaeological Research Laboratory, Volume 3, pp. 4-12, University of Texas at Austin Rose, John J. 2000 A Study of Late Classic Population Growth at La Milpa, Belize Ph.D. Dissertation, University of Pittsburgh Scarborough, Vernon L. 1994 Maya Water Management. National Geographic Research and Exploration 10(2):184-199 211

PAGE 210

Hubert R. Robichaux 212 Shaw, Leslie, Eleanor King, and Beverly Chiarulli 2005 Research at Maax Na, Belize: Report on the 2004 Field Season. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp 97-114, Occasional Papers Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Sprajc, Ivan 2001 Archaeological Reconnaissance in Southeastern Campeche, Mexico. Scientific Research Center of the Slovenian Academy of Sciences and Arts, Ljubljana Sullivan, Lauren A. 1997 Classic Maya Social Organization: A Perspective from Las Abejas Ph.D. Dissertation, University of Texas at Austin Sunahara, Kay S., and Richard K. Meadows 2005 Modeling Ancient Maya Non-Urban Complexity in Northwest Belize: A Working Paper. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp 63-96. Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Thompson, J. Eric S. 1939 Excavations at San Jose, British Honduras Carnegie Institution of Washington, D.C. 1963 Maya Archaeologist University of Oklahoma Press, Norman Tourtellot, Gair III, John J. Rose, Nikolai Grube, Sara Donaghey, and Norman Hammond 1994 More Light on La Milpa: Maya settlement archaeology in northwestern Belize. Mexicon 16(6):119-124 Tourtellot, Gair, Francisco Estrada Belli, John J. Rose, and Norman Hammond 2003 Late Classic Maya Heterarchy, Hierarchy, and Landscape at La Milpa, Belize. In Heterarchy, Political Economy, and the Ancient Maya edited by V. Scarborough, and F. Valdez, Jr., pp 37-51, The University of Arizona Press, Tucson Valdez, Fred, Jr. (editor) 2005 Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin Walling, Stanley 2004 Archaeological Investigations at Chawak Butoob: Pre-Hispanic Settlement, Water Management, and Community Structure on the Rio Bravo Escarpment, Northern Belize. Draft manuscript in the possession of the author, Department of Classics and General Humanities, Center for Archaeological Studi es, Montclair State University, Upper Montclair, New Jersey Walling, Stanley, Peter Davis, Jonathan Hanna, Leah Matthews, Nahum Prasarn, and Christine Taylor 2006 Residential Terracing, Water Management, Matrix Analysis, and Ceremonialism at Chawak Butooob, Belize: Report of the 2005 Rio. Bravo Archaeological Survey. In Programme for Belize Archaeological Project: Report of Activities from the 2005 Field Season edited by Fred Valdez, Jr., pp. 41-88. Occasional Papers, Number 6, Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin

PAGE 211

19 HINTERLAND SETTLEMENT IN NORTHWESTERN BELIZE: EXCAVATIONS AT THE MEDICINAL TRAIL SITE David M. Hyde and Shelly L. Fischbeck The Medicinal Trail site is a hinterland settlement approximately 6 to 8 kilometers east of the major ceremonial center La Milpa. Located in the La Lucha uplands atop a ridge with numerous associated landscape modifications, the site consists of numerous courtyard and plazuela groups. Recent excavations have focused on an extended household group comprised of six mounds organized around three courtyards plus an additional mound slightly to the north. An elevated social status of the early inhabitants of the group is inferred by the recovery of exotic materials such as jade, obsidian, and granite, and by the construction of a Preclassic ceremonial round platform. These artifacts and features suggest that the occupants of this group were responsible for a variety of ritual activities in the hinterland. Results fr om surveying show extensive water an d land management features, including terraces and depressions, indicating an agriculturally based economy. The research at the Medicinal Trail site is important because of the insights into hinterland socioeconomic organization. Introduction This paper provides an introduction to an overview of excavations conducted at a small household group located 5 to 8 km east of La Milpa in th e Three Rivers region, an arbitrarily defined archaeological and geographic area that encompasses nearly 2000 km of northeastern Guatemala and northwestern Belize (Adams 1995; Robichaux et al. 2001). The group is part of a community referred to as the Medicinal Trail Site, which consists of a few formal courtyard groups, numerous informal clusters of mounds and multiple landscape modifications such as terraces, depressions, and linear features. The site is located in the La Lucha uplands atop a ridge, and can be characterized as a terraced community (Scarborough and Valdez 2003). The site extends from the Turtle Pond to the west, located in the Programme for Belizes La Milpa Research Station facility, and the escarpment to the east. The first excavations at the site were conducted in 2002 by Laura Ferries (2002) and Danica Farnand (2002) and consisted of two separate studies that investigated mounds (Op5) and terraces (Op6) respectively. In 2004 Jennifer Chmilar (2005a, 2005b) excavated the Turtle Pond (Op8), a seasonally inundated depression at the base of a slope west of the site. Since 2004 excavations have been conducted at the largest group so far identified at the Medicinal Trail site, referred to as Operation 7 (Op7). Op7 was located by members of the Programme for Belize Archaeological Project while on survey in January of 2004, with excavations beginning the next month and continuing in 2005 and 2006. Op7 consists of six mounds distributed around three contiguous cour tyards aligned on a north-south axis, and one additional mound to the north (Figure 1). At least four depressions have been identified immediately flanking Op7 on three sides, and there are many terraces, presumably related to agriculture, extending across the slopes moving away from the site center towards the Turtle Pond to the west, and towards the escarpment to the east. Ceramics recovered from the site indicate an occupation lasting from at least the Late Preclassic through the La te Classic (Lauren Sullivan, personal communication 2005). Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 213-220. Copyright 2007 by the Institute of Archaeology, NICH, Belize

PAGE 212

David M. Hyde and Shelly L. Fischbeck Figure 1 Map of Operation 7 Environmental Setting Escarpments, uplands, and bajos (seasonal swamps) characterize the karstic environment of the Three Rivers region (Brokaw and Mallory 1989). Three terraced uplands, faced by steep escarpments with increasing elevations from east to west, form the dominant topographic features in the area (Houk 2000). Formed by a series of southwest-to-northeast-trending fault lines; from east to west the three terraces are the Booths River Escarpment, the Rio Bravo Escarpment, and the La Lucha Escarpment (Houk 2000; Sullivan 2002; Figures 2 and 3). The upland terraces are generally undulating with broadly rounded hills and stretches of level ground (Houk 2000). The three rivers from which the region takes its name drain the study area. The Ro Azul genera lly only flows during the rainy season, while the Booths River and the Rio Bravo flow year round. The Ro Azul begins approximately 28 km northeast of the site of Tikal in the central Petn, Guatemala, and flows from southwest to northeast towards Mexico, eventually joining the Rio Hondo a nd flowing east into Chetumal Bay. Figure 2. Physiography of the Three Rivers region. (Courtesy of PfBAP, adapted from Houk [1996]) The Medicinal Trail Site Throughout the Medicinal Trail Site community numerous terraces, linear mounds, and depressions have been located. The terraces run both perpendicular and parallel to the slope, and many end at depressions suggesting both agricultural, erosion control, and water management functions. Extensive excavations to the west and east of the northern and central courtyards have revealed evidence for landscape modification. The bedrock in these areas slopes downward from north to south and the area was stripped of sediment down to bedrock. Next fill, or in some cases re-deposited midden ma terial, was laid on this exposed surface artificially leveling these areas. This was followed by the construction of an evenly laid, compact prepared surface that converged with the groups structures. Numerous terraces have been identified on the slopes leading away from Op7. Danica Farnand (2002) excavated four terrace features along the tourist trail portion of the site in 2002. Al l dated to the Late Classic period based on recovered ceramics. 214

PAGE 213

Medicinal Trial Investigations in Northwestern Belize Figure 3. Elevational transect from the northern Caribbean coast of Belize to the Mirador Basin in the northern Peten. (From Dunning et al. 2002). Figure 4. Profile of Terrace Excavation. Of the four, she identified two as agricultural in function and one as a water control feature, while the fourth appears to be structural. Additional terrace excavations took place .5 kilometers downslope, west of Op7 in the 2005 field seas on. The terrace wall consists of two rows of large stones with smaller cobbles between them (Figure 4). The preserved portion of the wall is approximately 80 cm tall and 80 cm thick. Additional large stones down slope from the terrace wall suggest that the original wall was taller but has since collapsed. A similar construction is seen in one of the agricultural terraces identified by Farnand (2002). The high percentage of lithic debitage behind, within, and in front of the terrace wall is of particular interest. Large cortical flakes were found throughout the excavation, increasing in number with depth. These flakes may have been intentionally placed in the terraces to enhance the drainage of the clayey terrace soils (Healy et al. 1983: 405). Additional excavations are needed to determine the terrace functions and their construction history. Numerous depressions are dotted throughout the Medicinal Trail site, including four in proxi mity to Op7. The two depressions immediatel y west of Op7 were extensively excavated in 2004 and will be reported on later by John Lowe as his Masters thesis. The largest of the depressions at Op7, located just southeast of structure seven, has been excavated and 215

PAGE 214

David M. Hyde and Shelly L. Fischbeck reported on by Yoav Me-Bar (2005) and Jeff Brewer (2005). Northern Courtyard The northern courtyard group consists of three m ounds, one each on the west and east sides and one to the north offset to the west slightly. Not visible on the surface are two additional mounds, both dating to the Late Preclassic. Structure 1 is located on the western side of the courtyard and dates to the Late Classic based on recovered ceramics (Figure 5). The outside of the building is about 8m long by 4m deep while the interior space is approximately 5m wide by 3m deep. The doorway is a little less than 2m across and the walls are on average 80 cm thick. There is a plaster bench in th e northwest corner of the structure against the back wall extending north south. Figure 5. Plan Map of Structures 1 and 2. There is a low-walled platform with ancillary structure ex tending north off of Structure 1, towards Structure 2, facing the courtyard (Figure 6). This ancillary structure likely had wattle and daub walls above the masonry walls, and topp ed with a thatched roof. The ancillary stru cture appears to have been added to the courtyard later. Initially a freestanding low wall extended north off of Structure 1, ending slightly west of Structure 2. Then at a later date it was attached to Structure 2 and filled in to the east to create a platform for the construction of the ancillary structure. In the center of the northern courtyard excavations revealed a Late Preclassic plaster floor and two additional platforms (Figure 6). Intruding into this floor were two nested Late Preclassic Sierra Red vessels. Excavation of the plaster floor revealed that this cache was positioned over the northwest corner of a square platform. Resting on another plaster floor the square platform is slightly less than 1 meters on a side with remnant red painted. A cache consisting of three sets of lip-to-lip Sierra Red vessels was discover ed to the west of the platform. The vessels were broken in place although it is difficult to say whether this was intentional or not. Since the cache underlies a Late Preclassic floor that covered the Northern Courtyard the plaster floor may have been cut through to allow placement of the cache, and then re-plastered. This plaster floor was clearly present in the southeastern portion of the unit, but only the most tenuous remains of plaster remained in the western portion ove rlying the cache. Intrusive tree roots may have played a role in this absence of a clear plaster floor. On the same plaster floor as the round structure and approximately 50 cm north is a round structure nearly 4m in diameter, and 40 cm tall resting on a plaster floor. There is some indication that on the west side of the platform there were two stairs. No plaster or postholes were found on top. Excavations insi de the platform revealed nearly 2m of construction fill down to bedrock. Under the fill resting on bedrock was a secondary burial with at least two 216

PAGE 215

Medicinal Trial Investigations in Northwestern Belize ceram ic vessels. The skeletal remains were covered with sascab cementing them to the bedrock, making recovery exceedingly difficult. Figure 6. Plan Map of the Pre-Classic Structures at Op7 The Late Preclassic plaster floor was removed prehistorically or never existed east of the round struct ure and rectangular platform (Figure 6). At the edge of the square platform the plaster floor abruptly ends and along the plaste r edge is a row of stones that extends north towards the round structure. To the east of the plaster floor is an alignment of large, flat-topped stones rectangular or oval in shape approximately 50 cm in length. The ar ea east of this stone alignment was excavated down to bedrock. Interestingly, the bedrock here is approximately 30 cm below the plaster floor whereas bedrock is nearly 1.5 m down under the round structure. Central Courtyard The Central Courtyard consists of two structures on an elevated plaza. One of the mounds is a presumed residential mound that is L-shaped and located in the northwest corner of the patio, the other a large templelike structure on the east side of the courtyard. Excavations in the middle of the courtyard area indicate that the plaza is artificially elevated. The construction sequence is ballast stone fill covered by clay, then a layer of cobble, smaller fill material, then sascab. Above the sascab the sequence repeats large cobble, small cobbles, then gravelly fill. Ceramics from the gravelly fill date the early Late Classic (Tepeu 1-2) while all ceramics below are Late Preclassic. To determine the nature of access to this courtyard from the south where there is a significant change in elevation, excavations were conducted at the edge of the courtyard. Instead of a staircase, a vertical wall was uncovered consisting of well cut limestone blocks and a thick retaining wall made from unshaped stones, between which was fill material. This would indicate restricted access to the Central Courtyard from the south and only a narrow passageway into the space from the north. Southern Courtyard The Southern Courtyard consists of a single mound on the west side of the patio below the elevated Central Courtyard. The patio area of the S outhern Courtyard had been artificially leveled, much like the areas to the east and west of the group. Having stripped the sediment to bedrock, fill and midden material was placed over it, creating a level area onto which was constructed an earthen floor. This ear then floor is found throughout the courtyard and abuts with the Structure 6 platform. Structure 6 is a masonry construction approximately 12m long on the long axis resting on a platform that extends between 30 cm (in the south) and 50 cm (in the north) in front of the walls (Figure 7). The doorway is almost 1.5m wide and rather than being situated in the center, is offset to the south. The walls of Structur e 6 are thick like 217

PAGE 216

David M. Hyde and Shelly L. Fischbeck Structu re 1, on average 80 cm. Excavations have indicated a single large interior space and as of yet there is no interpretation for the function of this structure or why the door is offset. Figure 7 Plan Map of Structure 6. Structure 7 Structure 7 is located about 25m north of the main group near a large depression (Figure 8). Structure 7 is a low platform of approximately 80 cm in height and 9 m x 4 m horizontally with the long axis running E-W. Excavations around the platform have revealed an artificial pit approximately 2m in diameter, likely the result of quarrying, with a midden inside. The midden contained large amounts of charcoal, lithic debitage, a charred macrobotanical specimen (possibly a squash seed) and large ceramic sherds. Ceramics from the midden date from the Late Preclassic, the Proto Classic, Early Classic, and some Late Classic (Lauren Sullivan, personal communication 2006). The western portion of the midden deposit extends about 15 cm beneath the eastern wall of Structure 7, near the platforms northeast corner indicating that Structure 7 was constructed after the midden and pit were created. In addition to the midden deposit, there are curvilinear scars on the exposed bedrock around and under the platform that are roughly the size and shape of a limestone block. Moreover, numerous cut stone blocks were stacked on the north side of the northeast corner of the platform with additional blocks slumped nearby suggesting quarrying. Figure 8 Plan map of Structure 7. Discussion A few preliminary interpretations can be made regarding the inhabitants of Op7 based on the excavations described above. The group dates to the early Late Preclassic with some evidence of a possible Middle Preclassic o ccupation, through the Terminal Classic. The group is situated in a favorable hilltop setting with the slopes leading away from it extensively modified with features related to agricultural intensification. From at least the Late Preclassic important ce remonial functions were fulfilled at this locality based on the presence of a Round Structure and associated platform. Numerous studies have demonstrated that good agricultural land is limited in the Maya Lowlands and were therefore occupied first in most areas (Fedick and Ford 1991; Furley and Newey 1979; Scarborough and Valdez 2003). As 218

PAGE 217

Medicinal Trial Investigations in Northwestern Belize populations increased, the founding households, like Op7, had an econom ic advantage over later arrivals due to their access and control over a non-replicable source of economic wealth, which lead to the development of inequality and the emergence of a hinterland elite with limited social power (Hendon 1991; McAnany 1993). Social status in Maya studies is often presented as either Elites, those in the large centers, and Commoners those outside the major centers. However, social status is much more complex and should be viewed as a continuum with a wide range of variability in each category. Findings at Op7 fit well with indicators of elite social status among the Prehistoric Maya such as the presence of public architecture and labor intensive structures, the preferential location of settlement, and presence of exotic items like jade, obsidian, and shell (Marcus 2004; Table 1). Conclusion Three years of excavation at Op7 of the Medicinal Trail Site in northwestern Belize has led to the identification of a long lasting hinterland elite settlement in the periphery east of La Milpa. All three courtyards and an isol ated platform to the north have all been sampled to varying degrees. Late Preclassic material has been recovered from all areas of the site and most of the exotic materials date to the Late Preclassic and were rec overed from standard contexts such as construction fill or midden. The economic wealth of the Op7 inhabitants was likely derived from the surrounding landscapes agricultural potential based on the extensive modifications identified on the slopes leading away from the group. Additional work remains to be done however. Additional survey is needed to more fully understand the settlement density at the Medicinal Trail Site and more testing is required, particularly of the terraces, to obtain a chronology in order to more fully understand the development of the community. Acknowledgements : We would like to thank the Institute of Archaeology, Belmopan, Belize for permission to conduct this research and for providing the opportunity to present it at the symposium they organized in 2006. Additionally we would like to thank Dr. Fred Valdez, Jr. of the University of Texas at Austin and Project Director of the Programme for Belize Archaeological Project under whom this research was funded and overseen. Thanks also to Lauren Sullivan who conducted the ceramic analysis for the Medicinal Trail site and to Kirsten Atwood, Maria Dawson, Sandra Dias, Micaela Obledo, Rissa Trachman and Marisol Cortes Rincon for their assistance in the field. References Cited Adams, R. E. W. 1995 The Programme for Belize Archaeological Proj ect: 1994 Interim Report, Introduction. In The Programme for Belize Archaeological Project 1994 Interim Report edited by R. E. W. Adams and F. Valdez, Jr., pp. 1-14. The Center for Archaeology and Tropical Studies and The University of Texas at San Antonio. Brokaw, N., and B. Mallory 1989 Natural History of the Rio Bravo Resource and Conservation Area, Belize City. MS on file at the Programme for Belize offices, Belize City, Belize. Chmilar, J. 2005a Ancient Maya Water Management: Archaeological Investigations at Turtle Pond, Northwestern Belize Masters thesis on file with the Department of Anthropology, University of Cincinnati. 2005b Water Management at the Turtle Pond: A Preliminary Report of Excavations at RB62, Op 8. In Programme for Belize 219

PAGE 218

David M. Hyde and Shelly L. Fischbeck 220 Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp. 27-34. Occasional Papers, Number 4, Mesoamerican Archaeological Research Laboratory. The University of Texas at Austin. Dunning, N. P., S. Luzzadder-Beach, T. Beach, J. G. Jones, V. Scarborough, and T. P. Culbert 2002 Arising from the Bajos: The Evolution of a Neotropical Landscape and the Rise of Maya Civilization. Annals of the Association of American Geographers 92:267-283. Farnand, D. M. 2002 Agricultural Formation Histories of Prehistoric Terraces of the Medicinal Trail Site, Northwestern Belize. M.A. Thesis, Department of Anthropology, Division of Graduate Studies and Research of The University of Cincinnati. Ferries, L. C. 2002 Site Formation and Occupation History of the Medicinal Tra il House Mound Group at the Programme for Belize Archaeological Project, Belize. M.A. thesis, Department of Anthropology, Division of Research and Advanced Studies of The University of Cincinnati. Furley, P. A. and W. W. Newey 1979 Variations in Plant Communities with Topography over Tropical Limestone Soils. Journal of Biogeography 6:1-15. Healy, P.F., J.D.H. Lambert, J.T. Arnason, and R.J. Hebda. 1983 Caracol, Belize: Evidence of Ancient Maya Agricultural Terraces. Journal of Field Archaeology 10: 397-410. Hendon, J. A. 1991 Status and Power in Classic Maya society: An Arch aeological Study. American Anthropologist 93:849-918. Houk, B. A. 1996 The Archaeology of Site Planning: An Example from the Maya Site of Dos Hombres, Belize. Ph.D. dissertation, Department of Anthropology, The University of Texas at Austin. 2000 An Introduction to the 1998 and 1999 Seasons. In The 1998 and 1999 Seasons of the Chan Chich Archaeological Project edited by B. Houk, pp. 1-14. Papers of the Chan Chich Archaeological Project, Number 4. Mesoamerican Archaeological Research Laboratory, The University of Texas at Austin. Marcus, J. 2004 Maya Commoners: The Stereotype and the Reality. In Ancient Maya Commoners, edited by J. C. Lohse and F. Valdez, Jr., pp. 255-283. University of Texas Press, Austin. McAnany, P. A. 1993 The Economics of Social Power and Wealth among Eighth-Century Maya Households. In Lowland Maya Civilization in the Eighth-Century A.D., edited by J. A. Sabloff and J. S. Henderson, pp. 65-90. Dumbarton Oaks, Washington D. C. Me-Bar, Y. 2005 Excavation in a Depression at the Medicinal Trail Site: 2004. In Programme for Belize Archaeological Project: Report of Activities from the 2004 Field Season edited by Fred Valdez, Jr., pp. 15-26. Occasional Papers, Number 4, Mesoamerican Arc haeological Research Laboratory. The University of Texas at Austin. Robichaux, H. R., A. Miller, E. Roberts III, C. Pruett, B. Smyth, T. O'Brien, F. Valdez, Jr., and R. Valdez 2001 Preliminary Report of the 2001 Season of the Punta de Cacao Project. Manuscript of file, University of the Incarnate Word, San Antonio, Texas. Scarborough, V., and F. Valdez, Jr. 2003 The Engineered Environment and Political Economy of the Three Rivers Region. In Heterarchy, Political Economy, and the Ancient Maya: The Three Rivers Region of the East-Central Yucatn Peninsula edited by V. L. Scarborough, F. Valdez Jr., and N. Dunning, pp. 3-13. The University of Arizona Press, Tucson.

PAGE 219

20 INVESTIGATIONS OF AKAB MUCLIL: A LATE HINTERLAND SETTLEMENT IN NORTHWESTERN BELIZE Antonio E. Padilla, Molly Morgan and Jon C. Lohse During the 2005 season, investigations were carried out at the site of Akab Muclil, a corporate group cluster located in an environmentally diverse area close to the Rio Bravo and adjacent wetlands. Following the premise that political economies are rooted in local traditions, research was focused at understanding social and political dynamics of a hinterland settlement that potentially maintained ties with nearby centers such as Blue Creek and Gran Cacao. Findings reveal patterns of household practices and growth beginning in the Early Classic and continuing through the Late to Terminal Classic and into the Early Postclassic. Introduction During the Terminal Classic many sites in the Southern lowlands experienced a reorganization or restructuring event. A decline in monumental construction, a depopulation of the area, an alteration of elite power, and changes in material assemblages characterized this change. (Chase and Rice 1985). Several scholars hypothesize that during this period there was a collapse or transition that led to the depopulation of large cere monial sites in the Southern lowlands and the growth of sites in the Northern lowlands (see Sharer 1996; Traxler 2006). This phenomenon is seen in the archaeological record with the decline of major powers in the Sout hern lowlands such as Tikal and Calakmul and the rise of powers to the north such as Chichen Itza (Martin and Grube 2000; Sharer and Traxler 2006). Such transitions mark the beginning of the Postclassic period in Belize and most of the Maya world, (Sharer and Traxler 2006). Many, though not all Postclassic sites located in the Northern lowlands are situated in and around coastlines, islands, rivers, and lakes or la goons (Chase and Rice 1985). This case also holds true for Postclassic sites found in Belize, such as communities at Progresso Lagoon and Laguna Seca (Masson and Rosenswig 1998), Santa Rita (Chase 1985), Laguna de On (Masson 1993), Lamanai (Pendergast 1981), and communities in the Stann Creek District (Graham 1985). According to Masson (1999:285), Postclassic period aquatic settlement in the Maya lowlands was intimately connected to a circum-Yucatecan canoe trade that connected the Gulf of Mexico to Honduras, and appears to have been well-developed since at least the Terminal Classic period, as sites in coastal Belize suggest. These patterns suggest that proximity to comestible aquatic resources, in addition to access to trade networks, was an important factor in locating Postclassic sites. So far, sites meeting these criteria are limited to the coastline and river systems of northern Belize. This is due to research oriented to the exploration of Postclassic sites, which traditionally are based in the northern region of Belize. There is not much known about the Postclassic in the Three Rivers Region of northwestern Belize. During the 2005 season investigations were carried out at the site of Akab Muclil, a small, nucleated settlement cluster probably representing an extended family or corporate gr oup residential cluster in northwestern Belize. The site is situated in an active environmental zone with a Research Reports in Belizean Archaeology Vol. 4, 2007, pp. 221-230. Copyright 2007 by the Institute of Archaeology, NICH, Belize.

PAGE 220

Antonio E. Padilla Figure 1 Topographic map (20 m contour interval) of upper northwestern Belize, showing major sites and some of the regions hinterland settlements. perennial wetland/swamp immediately to the east, the Rio Bravo escarpment 1 kilometer to the west, and the Rio Bravo 500 m to the south. The site lies almost equidistant between two larger ceremonial centers, Blue Creek approximately 4 km to the north and Gran Cacao to the south (Padilla et al. 2006a) (Figure 1) As our work progressed, evidence was revealed for an unexpectedly late date of occupation for much of the site. Like many sites in the regi on, some structures were established in the Early Classic and were maintained through the Late Classic and into the Terminal Classic. What was unexpected, though, was evidence for Early Postclassic occupations of some buildings, including both domestic structures and the temple that occupies the center of the cluster (Padilla et al. 2006a). This paper documents the chronological sequence of occupation of Akab Muclil, from the Early Classic through the Early Postclassic. Site Layout and Unit Placement Akab Muclil was built on two platforms, a low one to the south that abuts a higher main platform at the center of the site (Figure 2). We identified at least 13 structures of varying height and dimensions on these two platforms. The main platform supports the heart of the site, including Structure 1, measur ing roughly 4.5 m high; eight smaller buildings, Structures 2; and a small stela in the southeast corner. The southern platform supports two smaller structures (10 and 11) at the southeastern end, and what appear to be two ground-level platforms designated as Structures 12 and 13 at the southwestern corner. Excavations included a combination of vertical penetration to record construction histories and horizontal exposure to pursue architectural alignments and explore ceramic and faunal middens visible on the surface. Early Classic Occupation Earliest evidence of site occupation is from the Early Classic, with the construction of Structure 1. Additional Early Classic ceramics were found beneath 222 222

PAGE 221

Investigations at Akab Mucli, Northwestern Belize Figure 2 Map of Akab Muclil showing location of excavations. Structures 10, 11, and 12 on the lower southern platform, but these came from mixed construction fill contexts rather than discrete contexts. Considering the large size of many of the mounds at this site, particularly those located on the main platform, more Early Classic components than what we found are almost certainly located here. Structure 1 is the largest structure of the site and stands 4.5 m high at the center of the main platform. A looters trench penetrates the center of the northern face of the structure; excavations off to one side of this trench exposed various construction phases (Figure 3). The first of these phases consisted of a leveling event with a mixture of pebble fill and dark gray soil placed directly above modified bedrock, which was then covered with plaster, creating a platform standing 60cm high. A small burial crypt, unfortunately loot ed, is at the center of this platform. This phase was followed by further modification of bedrock into two short steps located 60cm and 1.10 m north of this low platform. The initial platform was covered in the Early Classic with three other layers of constr uction fill and plaster floors, raising the plat form a total of 30cm, 50cm, and 15cm, respectively. Lastly, the buildings northern wall was extended, moving the edge of the platform 60cm northward, to the edge of the first carved bedrock step. The addition contained two plaster floors, and an outer masonry wall composed of cut limestone blocks. By the time these modifications were finished, the platform stood nearly 2m tall, and was capped with a hard plaster floor. Late and Terminal Classic Occupation A number of constructions occurred in all parts of the site during the Late and Terminal Classic. These periods represent what might be considered the sites apogee or peak in terms of occupation and perhaps prominence in the region. Descriptions of these periods of the site history begin with Structure 3 on the northwest corner of the main platform and continue clockwise around the site, ending with Structure 8, immediately to the southwest of Structure 1. A huge tree was growing out of Structure 2 northwest of the main building, so no excavations took place there. Structure 3 was also badly damaged by trees and roots, hence only preliminary excavations were possible. The building appears to have consis ted of two separate rooms, at least one of which faced southeast directly to the central Structure 1. While no undisturbed contexts were recorded ceramics recovered from this unit indicate Late-to-Terminal Classic construction. Structure 4 is on the northeast corner of the main platform, and like Structure 3 was aligned toward the central, Structure 1. Excavations exposed substantial Late and Terminal Classic construction on the 223 223

PAGE 222

Antonio E. Padilla Figure 3. West Profile, showing Early Classic construction, Subop H, Structure 1 (site terrain map looking west). Photo facing North shows looted burial crypt above plastered platform Figure 4. East profile, Late/Terminal Classic construction, Subop L, structure 4 (site terrain map looking east). (Photo) Facing North with arrows in photo pointing to the different floors. southern end of this building additional excavations into the center of this large mound would probably locate earlier contexts. Excavations revealed four phases, with each phase adding more cobble fill and a higher floor (Figure 4). The first construction was a platform 50cm above the uneven bedrock surface. Above this, another layer of cobbl e fill about 20cm thick supported a plaster fl oor that showed evidence of burning. The third episode of cobble fill and refloorin g raised the platform another 30cm. The last floor and corresponding height of the wall was 1.40 m above bedrock; the final wall facing the outer part of this build ing was of finely cut limestone blocks three courses high. 224 224

PAGE 223

Investigations at Akab Mucli, Northwestern Belize A 1x2 m unit was excavated in the small plaza area south of Structure 1 over what appeared to be a stela fragment. The monument was smoot hed on all edges except the top and the northwest side, which were clearly broken. It s original height is unknown. When this stela fragment was found, it was oriented approximately 10 degrees W of N so that it perfectly faced the central Structure 1. It appears that this fragment was the stela butt that was placed in a 30cm deep hole in the platform. The hole was subsequently fill ed first with a soft silty soil, into which the monument was placed, and then packed with stones, pebbles, and tightly compacted soil. Based on ceramics from the stela cut, the dedication/placement occurred in the Lateto-Terminal Classic period. Figure 5 Non-residential wall, looking east. Far eastern end is marked by arrow; wall makes northward turn at arrow. Exposure measures 8 m, scale extended at 2 m. An interesting non-residential feature was discovered on the lower platform at the southeast corner of th e site. Excavations in Subop E revealed a thr ee course-high, eastwest running wall seated above an eroded plaster floor. During excavations we exposed eight meters of a wall to a corner where it turned north (Figure 5). We exposed another two mete rs of this section before running out of time at the end of the season. Although the full extent of the wall is not determined, it is believed that it extended the entire length of the southern platform and may have served a defensive purpose. Figure 6 North wall profile of Structure 8, Subop D (site terrain map looking north). What we ended up calling Structure 11 (excavated in Subop C) is a nearly ground level feature just west and slightly north of the long wall described above. The poor condition of the structure and badly preserved stone alignments made it difficult to interpret its function. However, our excavations did recover a large concentration of terrestrial and aquatic fauna including jute shells, tu rtle, fish, and brocket deer, suggesting the structure might have been related to food processing. The final structure that showed evidence of Late and/or Terminal Classic 225 225

PAGE 224

Antonio E. Padilla occupation was Structure 8, a large residential structure located at the southwest corner of Structure 1 that opens into the sm all plaza area with th e stela. This house was built of well-shaped limestone blocks and included a hard plaster interior floor. Excavations exposed the east-facing doorway and a complex sequence of constructions and modifications in the buildings interior (Figure 6). A plastercoated limestone bench had been built against the rear (west) wall, almost directly opposite the eastern doorway. At some point, this room (nearly 4m deep) was partitioned by an interior N-S running partition that was built on the plaster floor. Excavations between the partition and the bench revealed an elab orate ritual involving the burial of two adults (Burials 2 and 3), a young juvenile, and possibly a fourth individual. Figure 7. Plan of double burial in Structure 8 with grave offerings. At some point in the Late Classic, the sites inhabitants removed part of the bench, and placed Burial 2 inside the bench on the plaster floor. We recovered remains including unidentifiable bone fragments, as well as teeth, mandible, crania, and long bones. Burial 3 was placed on the plaster floor east of the bench and Burial 2, just inside the partition wall. Identifiable bones included ribs, long bones, crania, mandible, and teeth. Both interments were flexed, with their heads facing west. Burial 2, positioned slightly to the south of Burial 3, was aligned with its head to the south, while Burial 3 had its head to the north. Immediately south of Burial 3, a smaller bone cluster was identified that may indicate a third individual. Based on morphology of the teeth and bones, we believe that these represent the remains of a very young juvenile, though osteological analyses are necessary for a more secure assessment of age. Figure 8 Artifacts recovered from Subop J, including small broken vessel, perhaps dedicatory offering, in construction fill (A); green obsidian blade segment (B), and ground stone fragment (C). Grave objects located near Burial 3 included a limestone mano to the east; a rounded chert hammerstone to the west; and four shell beads to th e south, near what may possibly be an infant bu rial. Lastly, a black carbonized substance wa s identified between the two burials that appears to be burned 226 226

PAGE 225

Investigations at Akab Mucli, Northwestern Belize organic subs tance. Burial 3 had been placed over a cut in the plaster floor (Figure 7). Excavations into this cut recovered only a few very badly preserved long bone fragments. It is believed that the cut in the plaster floor immediately preceded the two interments, which occurred as a single event. We can only speculate at present about whether the individual s in this deposit represent a single family or whether these individuals were unre lated. Nevertheless, these burials clearly represented an important moment in th e history of Structure 8. Their interment in a single event, together with the careful positioning of the bodies, raises the possibility that they were sacrificial offerings marking an important event. Early Postclassic Occupation Not much is known of the Postclassic in northwestern Belize. Laura Levi (1994) described some looted vessel fragments recovered from Gran Cacao as possibly Postclassic, and minor visitations have been noted at Chan Chich (Guderjan 1991), and Dos Hombres (Houk 1996). Visitations to La Milpa appear to have been slightly more substantial, and are noted by the presence of possible squatters remains in the main plaza and the resetting of a number of the sites twenty stelae (Hammond and Bobo 1994). However, to date no significant residential settlements have been identified from this time period in the region. At Akab Muclil, the majority of the structures located on the main platform contain at least one construction phase dating to the Early Postclassic, while discrete contexts from this period were also recognized to the nort heast of the main platform and on the lower platform to the south. Structure 1, the main temple in the center of the site, has significant Postclassic additions. These additions included a thick (15-20cm) packed marl floor over bedrock in front of the structure, and a large frontal staircase leading to the structures summit, which supported a small room. Jambs to either side of the doorway into this chamber were partially visible at the margins of the looters trench. After the room had filled in with building collapse, suggesting a brief hiatus in occupation or use of the structure, the entire building was capped with very large roughly cut limestone boulders. Since the lowest plaster floor level contained diagnostic Early Postclassic ceramic sherds mixed with some Late/Terminal Classic types, this final episode is also attributed to the Early Postclassic. It remains for future excavations to determine when the summit structure was built, or what happened to this building during the Late and Terminal Classic. At Structure 4, at the northeast corner of the main plaza, excavations penetrated layers of collapsed cobbles in front of (to the south) the building. We exposed a one-course high step below this collapse that extended from the base of the southern wall out into the main plaza. The collapse contained a mixture of Late-toTerminal Classic and Early Postclassic period ceramics, suggesting continued occupation and use of th e building. The step abuts the wall and was clearly a later architectural addition. Ceramics in the fill of this step include Ea rly Postclassic types. Structure 6, in a small patio at the far northeastern corner of the site, appears to have been built entirely in the Early Postclassic. Excavations here revealed a front (western) wall facing a small courtyard, and a back (eastern) wall facing off of the side of the sites main platform. Both walls were built of a single course of crudely shaped limestone blocks. Some burned cobbles were id entified the west of the structure. A heavily utilized blade segment of green obsidian was recovered 227 227

PAGE 226

Antonio E. Padilla here; this is the only piece of green obsidian found at the site. Also, a ground stone fragm ent of granite from the Maya Mountains was recove red, indicating that some long distance trade routes remained active in the region. Excavations also recovered a small fragmented vessel included in the fill, perhaps placed as a dedicatory offering (Figure 8). Back at the far southeast corner of the site, our excavations in Subops A and E exposed what appeared to be a collapsed structure. Buried amid loose cobbles and humus, we found low southern (back) and northern (front) walls sitting on a compact dirt floor. Ceramics associated with these features were from the Late, Terminal, and Postclassic periods. This low building, which we interpret to have been a house, was built over the collapsed remains of the long non-residential wall we described earlier. The collapse separating the two constructions, plus the dramatic change in implied function, suggest that some hiatus separated the two constructions. About 20 meters to the west, excavations into what seemed from the surface to be a thin deposit of ceramics and faunal material exposed a dense midden of mixed ceramics from Late-to-Terminal Classic and Early Postclassic all the way down to bedrock, nearly a meter and a half. Dating the components of this part of the site is somewhat difficult. Most of the ceramics were Terminal Classic types, though Early Postclassic sherds were recovered from every lot, sometimes in low frequencies, all the way down to the bottom of our excavations. A ground-level (nonraised) platform with a low two-course retaining wall was also exposed in this unit. It is not known whether this structure was a house or served an ancillary purpose. Fauna from the midden included turtle, aquatic shells, peccary, and both white tail and brocket deer. Some bones exhibit cut marks. Just a few meters west of this unit, another possible grou nd-level structure, Structure 13, was exposed in Subop G. Excavations here removed the topsoil and humus, exposing what appears to be only faint stone alignments. Early Postclassic sherds covered this part of the site. Moving back to the main platform, Structures 8 and 9 also have Early Postclassic components. Excavations into Structure 9 consiste d only of stripping topsoil; we recorded an E-W running wall of the structure and a second, lower wall that looked like the edge of an underlying supporting platform. Fragments of a badly preserved skeleton were recovered from the topsoil of this unitthis was labeled Burial 1, though does not represent a formal interment. Its even possible that this individual is historic in age, though no plastic or metal objec ts were found here. Pottery from this limited excavation was a mixture of all time periods from the Preclassic through the Early Postclassic. Structure 8, with the elaborate Late Classic ritual burial deposit, also had an Early Postclassic house built at the surface. Front (east) and rear (west) walls were exposed, defining a room that measures about 5 meters wide. This house was built over the collapsed soil and cobbles of the earlier Late-to-Terminal Classic dwelling, suggesting a hiatus between occupations. Conclusions The earliest confirmed occupation at Akab Muclil takes place during the Early Classic. Preceramic sherds were recovered from fill contexts, though no secure deposits from this time period were recorded. We know that by the Early Classic, and possibly the end of the Early Classic, this site appears to have been an important settlement in the hinterlands between th e larger centers of Blue Creek and Gran Cacao. 228 228

PAGE 227

Investigations at Akab Mucli, Northwestern Belize As with most other s ites in the region, the peak in occupation occurred in the Late to Terminal Classic. The main building, Structure 1, was probably added onto at this time, though our exposures did not record specific arch itectural additions. What we did see were several houses arranged around the main plaza and its central pyramid, establishing a site plan that later inhabitants of th e site would preserve and maintain. Most occupation took place on the main platform though activities that included possible food-preparation areas and trash deposits were also conducted on the lower, southern platform. We have evidence of elaborate domestic rituals involving multiple burials during the Late Classic. Strong directional symbolism is implied by the alignments of the interments, with north and south orientations to the east and west, respectively, representing up and down. It is possible that this de posit was intended to reflect the movement of the sun around the earths plane. This potent symbolism is seen elsewhere in Maya site s and art, though has been recorded only infrequently in a common domestic setting such as this. Other material evidence from the site indicates more typical domestic activities grinding implements; utilitarian bifaces, hammerstones, and recycled stone tools; heavily utilized obsidian blade fragments; and patterned data suggesting food preparation. The long stonewall bounding the sites south side during the Terminal Classic perhaps tells us something of regional stability at this time. This sites remote setting probably provided access to diverse environments and lots of food resources we see evidence for this in the faunal record. However, it also would have left these people somewhat vulnerable in the event of social unrest or even hostilities. We have no firm evidence for violence at the site, but our data do imply that security and defensibility were important considerations in the Terminal Classic. Also for the Terminal Classic, the setting of the low, unadorned stela is potentially significant. If nothing else, it suggests that site oc cupants were heavily invested in the sanctity of this location. We suggest that this stel a tells us something about the reduced political weight or importance of nearby centers. Akab Muclil is believed to be the first recorded Early Postclassic occupation site in the region. Identifying buildings from this time period is difficult for at least two reasons: first, not all Postclassic constructions utilized well-cut stone blocks, sometimes making them hard to distinguish from other rock outcrops. Second, a thin sheet midden of Postcla ssic pottery seems to overlie almost the whol e site. Unlike earlier inhabitants, later occupa nts at the site appear to have been very messy. Most Postclassic constructions occurred directly on t op of earlier buildings; Structure 6 to the northeast is the exception to this. Only in Structure 4, though, do we have a strong case for occupational continuity between these two periods. Perhaps most of the site was abandoned by the end of the Terminal Classic, and then reoccupied a short time later. Future excavations will resolve this question. Our work also indicates that future research in the region should consider environmentally diverse settings such as this one when searching for other Postclassic habitation. So far no evidence indicates that people were still living in towns and cities that were abandoned by the Terminal Classic. Instead, data reflect brief pilgrimages and visitations squatting and stela resetting at La Milpa (Hammond and Bobo 1994) and the small offerings at Chan Chich (Guderjan 1991) and Dos Hombres (Houk 1996) reveal the temporary nature of 229 229

PAGE 228

Antonio E. Padilla 230 230 these revisits (Padilla et al. 2006b). One pattern that seems to be emerging, then, is that after the collapse of urban centers, at least some hinterlanders either stayed or returned to the region to continue their lives as best they could. La rger, central places remained important as reminders of the past, though in some places rural traditions persisted a century or more after urban sites were abandoned. References Cited Chase, Arlen F, and Prudence M. Rice (editors) 1985 The Lowland Maya Postclassic. University of Texas Press, Austin. Chase, Diane Z. 1985 Ganned But Not Forgotten: Late Postclassic Archaeology and Ritual at Santa Rita Corozal, Belize. In The Lowland Maya Postclasic, edited by Arlen F. Chase and Prudence M. Rice, pp. 10425. University of Texas Press, Austin. Graham, Elizabeth A. 1985 Facets of Terminal to Postclassic Activity in the Stann Creek District, Belize. In The Lowland Maya Postclassic, edited by Arlen F. Chase and Prudence M. Rice, pp. 215230. University of Texas Press, Austin. Guderjan, Thomas H. 1991 Chan Chich. In Maya Settlement in Northwestern Belize: The 1988 and 1990 Seasons of the Rio Bravo Archaeological Project, edited by T. H. Guderjan, pp. 35 50. Maya Research Program and Labyrinthos, California. Hammond, Norman, and Matthew R. Bobo 1994 Pilgrimages Last Mile: Late Maya Monument Veneration at La Milpa, Belize. World Archaeology 26:19. Houk, Brett A. 1996 The Archaeology of Site Planning: An Example from the Maya Site of Dos Hombres, Belize. Ph. D. dissertation, Department of Anthropology, University of Texas at Austin. Levi, Laura J. 1994 Ceramics from RB43. Collections from Looters Trenches. In Programme for Belize Archaeological Project 1993 Field Season, edited by R.E.W Adams. University Texas at San Antonio Martin, Simon and Nikolai Grube 2000 Chronicle of the Maya Kings and Queens: Deciphering the Dynasties of the Ancient Maya. Thames & Hudson Ltd, London. Masson, Marilyn A. 1993 Changes in Maya Community Organization from the Cl assic to Postclassic Periods: A View from Laguna De On, Belize. Unpublished Ph. D. dissertation. Department of Anthropology, University of Texas at Austin. 1999 Postclassic Maya Communities at Progresso Lagoon and Laguna Seca, Northern Belize. Journal of Field Archaeology, 26(3):285306. Masson, Marilyn A., and Robert M. Rosenswig 1998 Belize Postclassic Project 1997: Laguna de On Progresso Lagoon, Laguna Seca Occasional Publication of the Institute of Mesoamerican Studies No. 2. Institute of Mesoam erican Studies, Albany. Padilla, Antonio E., Molly Morgan, and Jon C. Lohse 2006a Excavations at Akab Muclil, Northwestern Belize. In Final Season Summaries of the Blue Creek Regional Political Ecology Project: 2005 Investigations in Upper Northwestern Belize, edited by Jon C. Lohse, pp. 1239. Maya Research Program, Ft. Worth, and The University of Texas at Austin. 2006b Investigations of Akab Muclil: A Hinterland Site in Northwestern Belize. Paper presented at the 71st Annual Meeting of the Society for American Archaeology, April 2630, San Juan, Puerto Rico. Pendergast, David M. 1981 Lamanai, Belize: Summary of Excavation Results, 19741980. Journal of Field Archaeology 8(1):29.