15 The Archaeology ofChechem Ha Cave, Belize: A Late Classic Hiatus in Usage Holley Moyes Cave studies have traditionally relied heavily on ethnographic or ethnohistoric analogy to understand the sacred context of M~soamerican caves. What is little understood are the behavioral processes that produced the artifact assembla?es In c~ves and the nature ofthe relationship between caves and their ancient users residing in surrounding surface sztes. Thzs study demonstrates that caves can provide information that is useful in broader research arenas. A Late Classic hiatus in cave use is described and correlated with regional sociopolitical stress. This correlation dem~nstrates th~f caves were not just venues for worshipping rain deities but were important political spaces that requzred protectIOn from enemies. Introduction local and regional histories. Using a robust Beginning in the mid 1970's there strategy ofexcavation and radiocarbon dating, has been a steadily increasing amount of this study identifies a period of hiatus in cave research on Mesoamerican ritual in caves usage between A.D. 560 and A.D. 680 and (Brady and Prufer 2005). Work in cave examines the implications of this finding. archaeology has traditionally relied heavily on ethnographic or ethnohistoric analogy Setting to interpret the meaning of caves and their Chechem Ha Cave is an ancient Maya contents, but despite years of research little ceremonial site used from the Early Middle is known about the behavioral processes that Prec1assic (1100-700 B.C.) and possibly produced the artifact assemblages in caves or as early as 1300 B.C., to the Late Classic about the nature of the relationship between period (A.D. 600-900). The tunnel system caves and their ancient users residing in in the cave is 198m in length and consists surrounding surface sites. To integrate of over 300m of tunnels. For the purposes cave archaeology into wider sociopolitical of this paper the cave has been divided into research arenas requires new methodological three parts, Tunnel 1, Tunnel 2, and Chamber considerations that entail a shift in focus 2. Chamber 2 is located in the center of the from the use of ethnographic analogy as the tunnel system, 134m from the cave entrance. primary interpretive source to the artifact The cave is located in western Belize record itself. This paper demonstrates the on the western bank of the Macal River near utility of the approach at Chechem Ha Cave. the Guatemalan border (Awe et al. 2005) The cave has been under investigation by (Figure 1). It is positioned at the edge of the Western Belize Regional Cave Project the Vaca Plateau approximately 26km north (WBRCP) since 1998 under the direction of Caracol and 25km southeast of Naranjo. of Dr. Jaime Awe. The overall project goals The nearest surface sites are the middle-sized included the investigation of temporal depth centers of Minanha to the south, Las Ruinas and fluctuations in cave use-intensity in the to the north, and Xunantunich to the far north. region (Awe 1998:1). Research at Chechem The Chan site, an agricultural community is Ha contributes by identifying fluctuations in located between Las Ruinas and Xunantunich cave use and articulating these findings with 225 The Archaeology ofChechem Ha Cave e~ e \ i z e R \ -J }lackmlID Eddy 10 Baltoo"RaJ11Ie aalatg Pot Q:' ~. cu IO~$" ~ "7.. • ~ L,1S Rm Ie Arenal flpu {jCHECHEM HA CAVE • Chae-ra ~ukllatTllnicli I • :!! J Mmanha - ~/i~ CIl ~ CD C) ; Vaca Plateau Archaeological Sites U cave SIte La ~jo a I Ca~coJ I Map COurtI!sy of AIl8lllll FOld _ Belize R_ 5ettJemenl SuMy Figure 1. Digital elevation map (DEM) of western Belize showing location of Chechem Ha Cave and surrounding sites (Courtesy of Anabel Ford and the BRASS project). Methods Spectrometry (AMS) Laboratory. All samples In 1997 the WBRCP undertook an were wood charcoal with the exception of extensive mapping and survey program two samples of corn. Dates were calibrated supervised by the author to record the cave's using Oxcal 3 and are reported at the 2 sigma morphology, cultural features, and artifacts. probability. The survey produced a detailed map of the There were 1901 ceramic sherds, tunnel system and an artifact catalogue. whole, or partial vessels recorded in the cave Extensive test excavations were conducted in that represented at least 566 different vessels. 2001. Twenty-two small shovel test pits and Ofthese 465 were typed for chronology (Jaime six larger units were placed throughout the Awe 1999; James Aimers 2003; Joseph Ball tunnel system (Moyes 2002). 1999; Joseph Ball and Jennifer Taschek 2005, Based on previous test results a broad Ishihara 2000; Kay Sunahara 2001)[personal horizontal excavation of Chamber 2, the area communications] using James Gifford's type- that demonstrated the heaviest utilization, variety-mode system (Gifford 1976). was undertaken in 2003 to evaluate its usage over time. The excavation consisted of 18 Settlement and Ceramic Chronology natural or cultural layers excavated to bedrock When compared with data from (Moyes 2004). Charcoal collected from each surface surveys the ceramic analysis from of the layers as well as 24 samples from Chechem Ha Cave demonstrated that surface and subsurface deposits were dated at cave ceramics provided a rough proxy for the University of Arizona Accelerator Mass settlement in the local area. The histogram 226  Holly Moyes in Figure 2 shows two data sets from local the entrance had been blocked offwith small settlement surveys juxtaposed with the to medium-sized boulders at least three and global set from Chechem Ha. The first set is possibly four times during the cave's history. compiled from Jennifer Ehret's preliminary Antonio Morales, the property owner, has survey of 242 mounds at Xunantunich (in reported that at the time of its discovery, Ashmore et al. 1994:283) and the second is limestone boulders blocked the entrance. from Cynthia Robin's survey of 100 mounds Morales' account is credible because it is at the nearby Chan site (Robin et al. almost certain that the cave would have been 2004:45). heavily looted had the entrance been easy to Their data represent the percentage find. Today, boulders surround either side of mounds containing a particular ceramic of the outside entrance and are also located complex. The data set from Chechem Ha is inside the entrance passage. Boulders were based on the percentage of the number of cleared from the center of the interior vessels representing each complex within passage by the owners to create a stairway the site. In the settlement survey data there for tourists leading down into the cave's is clearly a substantial occupation in the area entrance chamber. during the Middle Preclassic period but few A 1.5m x 1m excavation unit, Unit ceramics were found in the cave dating to 02-04, was placed in the center of the this period. Both settlement and cave use interior passageway at the base of the stairs. drop off in the Late to Terminal Preclassic The matrix of the entire unit consisted of and all data sets show continual increases small to medium sized boulders and loose from the Early to the Late Classic Periods. fill. The excavation revealed two use There is a very small Postclassic occupation surfaces beneath layers of limestone of areas around Xunantunich and within a boulders (Figure 3). The basal surface single structure at the Chan site that does not consisted of compacted clay and was AMS show up in the cave data. dated to the Middle Preclassic period (770- What is of interest here is that social 400 B.C.). Boulders and sediment covered disruptions occurring in short temporal the surface. A sherd dating to this period frames are not apt to show up in these kinds was horizontally embedded in the surface of data. Although ceramic chronologies and a medium-sized boulder sat directly on based on types are often used for dating top of the sherd. The charcoal sample was there -are inherent problems. First, some collected beneath the sherd. This suggested types may be prevalent throughout more that the cave was in-filled some time after than one time period (LeCount 2004) and the Middle Preclassic period. second, chronologies are not fine grained More boulders were noted below this enough to pick up small fluctuations in surface but these were impossible to remove settlement. Data collected from caves that without expanding the unit and building examine fluctuations in cave usage can aid supports to prevent the rock from collapsing. in defining periods of hiatus and social The presence of boulders beneath the early disruption. surface suggested an earlier infilling event as the above date was not the earliest date in Excavations the cave but was collected at the only Results of excavations conducted in possible entrance. This is a good indication the entrance area of the cave suggested that that the excavation did not reach the base of 227 The Archaeology ofChechem Ha Cave Comparative Ceramic Chronology Ma cal Valley Sites 100% 90% BD% 70% • X una ntuni ch 60% 50% (Ashmore et 40% a1.1994) 30% 20% II Chan (Robin et al. 10% 2004) 0% o Chechem Ha Cave- Ceramics Figure 2. Graph oflocal settlement survey data sets from Xunantunich (Ashmore et al. 1994:283) and Chan (Robin et al. 2004:45) compared with Chechem Ha ceramic types Chechem Ha Cave- Entrance Passage Unit 02-04 Profile " Surface- Cave Powder', '\ ! ........ , 0._'''. !Charcoal iAD 250-430 Features !~;t:;l Matrix ' \ Charcoal on Basal Surface n0-400 B, C, _Rock ";<H"'j Surface _Root _Sherd _ Charcoal 0,5 1 Meters • . Jute /\I Unit 02_04 Figure 3. Profile of Unit 02-04 placed in jumbled boulders in entrance passage. Two use-surfaces were overlain by boulders. the boulder jumble and that another surface destroys them. The presence of the deposit lay beneath the rock. suggests that the cave was unused for a The uppermost surface of the period of time and eventually covered by excavation consisted of compact blue-tinged sediment and rock. A piece of wood clay with a crusty crystalline powder charcoal removed from below the covering the top. This powder is most likely compacted surface dated to the Early Classic an autochthonous secondary mineral deposit Period (A.D. 250-430) and above the deposit referred to as "cave powder" (Hill and Forti Late Classic Spanish Lookout complex 1997:87). These types of crusts are very (A.D. 700-900) ceramics were scattered fragile and any trampling immediately over the surface and between boulders. The 228  Holly Moyes crust on the Early Classic floor suggests that on the top of the use-surface and its presence the cave sat dormant for some time between precluded trampling of the crust. At some A.D. 430 and the Late Classic period point it was protected by being covered over beginning about A.D. 700. with sediment and rock, possibly as the cave Rock overlying both of the was being unblocked prior to its Late excavated use- surfaces indicated that there Classic usage. Blocking of the entrance were at least three and possibly four during the early part of the Late Classic episodes of infilling in the entrance passage: period would explain why there was so little one possibly prior to or during the Middle sedimentation in Chamber 2 following the Preclassic, one after the Middle Preclassic, Early Classic. Not only would the lack of and another sometime after the Early Classic human usage reduce sedimentation, but by but before the end of the Late Classic. The blocking the entrance, the cave would also final blockage of the entrance had to have be inaccessible to bats, further eliminating occurred at the end of the Late Classic sedimentation. period as the latest radiocarbon date (A. D. 720-900) and presence of Late Classic Explaining the Hiatus ceramics suggests. Too often caves are thought of as Regarding the broad horizontal remote venues whose sole purpose is excavation in Chamber 2, AMS dates devoted to .the propitiation of rain deities. revealed that the sediment of the modem The sites are treated as un-integrated entities use-surface dated to the Early Classic and it is easy to forget that cave ritual was a period. The lack of sedimentation in the vital part of the ancient Maya community area during the Late Classic period was odd that may be expected to reflect the concerns as there were Late Classic sherds on the and fortunes of the ancient users. There are floor surface and on a small ledge above the no single blanketing explanations for chamber. disruptions in cave use just as there are no single explanations for socio/cultural ills. Radiocarbon Dating However, some factors must contribute The 37 calibrated AMS dates for the highly to disruptions in community life and Maya levels of the cave are listed in Table 1. therefore disruptions in ritual practices. The general areas from which dates were Disruptions on local community or regional collected are indicated as Tunnell, Tunnel scales will likely correlate with 2, Crawl 3, and the Chamber 2 excavations. sociopolitical problems or environmental The dates are listed in chronological order. stress. Note that there is a gap in the dates from The Late Classic hiatus at Chechem A.D. 560-680. There are no overlapping Ha (A.D. 560-680) is not an isolated dates in any cave area and almost no occurrence but in fact overlaps the Classic ceramics that date to this time period, Hiatus Phenomenon (A.D. 534-593) suggesting that there was a hiatus in cave identified and described by Gordon Willey use. (1987:72-73). Willey considers the hiatus to Excavations in the entrance passage be a phenomenon of the southern lowlands revealed an Early Classic use floor with during which there is a marked drop-off in boulders and sediment covering the top stela carving and dedication for indicating that the cave entrance was approximately 60 years dividing the Early blocked sometime after A.D. 430. Recall and Late Classic periods. The hiatus is not that a fragile crust or cave powder formed indicated in the northern Maya areas or at 229 The Archaeology of Chechem Ha Cave Palenque and some northern sites fluoresce Sum ofPercent Deviation at this time. Although Willey offers no rromMean n: 150 explanation for the occurrence, Richardson Gill (2000:318) struggles to argue that the , ~ hiatus is caused by aberrations in the world ~ ~ u 1600 climate around A.D. 536. Gill mentions that 11 ~~ . ~ l:"" 1400 Curtis et al. (1996) record a drought at Punta 1200 Laguna in Quintana Roo. The problem is 1000 ~ that the northern areas were not the ones 800 affected by the phenomenon, which weakens C/llC/IemH. h--.--.~~-i--,-,,-,---,--,---,, LCHlatul his argument. f..-U-ll-¥=--+-LL..L...l~~!!:d:::::~=-1 600 If the hiatus was in fact a result of 400 climatic manifestations, we would expect ~~_._- 200 aberrations in the local climate record. James Webster (2000) conducted a 'IZB.(;. 200 paleoclimate study using a speleothem from 400 ~ 1Il the Vaca Plateau in western Belize, which 4e88.C. 600 provides an excellent local paleoclimate 800 proxy. Webster evaluated the thickness and Wetter Drier frequency of bands, their color, luminescence, and isotopic ratios (8 180). Figure 4. Reconstruction of moisture conditions for Although the rings were dated using the Vaca Plateau for the past 2,700 years. The bar radiocarbon methods, the speleothem was of chart indicates drier conditions in black. Diagonal a known age so that old carbon could be lines illustrate the hiatus in cave usage at Chechem Ha. Note the overlap with the Maya Hiatus. (after factored as a variable in the calibration using James Webster 2000:197, fig.8.2) a standard calculation. Webster's climate record illustrates that moisture was within an During the hiatus there are wars average range during the entire hiatus between Tikal, Naranjo, and Caracol. Tikal (Figure 4). The hiatus followed one of the is defeated by Caracol by A.D. 562, the wettest episodes in Classic Maya history for beginning of the Chechem Ha hiatus. the local area therefore there was no Following Tikal's defeat there is a war preceding drought. It is clear that something between Caracol and Naranjo that lasts until else was occurring in the area at this time. A.D. 680 when Caracol goes into a hiatus Although there is little epigraphy in lasting until A.D. 700. Naranjo continues its the Belize Valley, numerous texts are domination until the Late Classic period and available for the closest large sites, Caracol is mentioned on texts from Xunantunich in and Naranjo. Table 2 is an abbreviated the early 9th century. The cave is blocked compilation of deciphered texts spanning the off (A.D. 560-680) during the same time Hiatus 3 time period. The table was period that conflicts arise between Caracol assembled from the Notebook for the 28 th and Naranjo (A.D. 596-680). It is unlikely Hieroglyphic Forum at the University of that this is a coincidence. Recall that Texas at Austin. War events, skirmishes, Chechem Ha is located halfway between the births, and accessions are included in the two sites and could easily be pillaged by table. The dates that include Hiatus 3 are either side. highlighted in gray. 230  Holly Moyes AZ Lab # Period Area Radiocarbon Calibrated Date Age 2 Sigma AA57293 LC Tl 1187±33 AD 720-900 AA57288 LC Tl 1210±31 AD 690-900 AA59754 LC Tl 1224±38 AD 680-900 AA59753 LC Tl 1239±36 AD 680-890 AA57291 LC Tl 1244±31 AA57271 EC Ch2 excL2 1587±34 AA57290 EC Tl 1605±32 AD 390-540 AA57310 EC Tl 1607±32 AD 380-540 AA57307 EC Tl 1638±42 AD 260-540 Beta EC Ch2 excLl 1660±40 AD 250-540 AA57301 EC Tl 1685±32 AD 250-430 AA59755 EC Tl 1696±36 AD 250-430 AA57272 EC Ch2 excL3 1673±34 AD 250-440 AA57273 EC Ch2 excL4 1668±34 AD 250-440 AA57274 EC Ch2 excL5 1685±39 AD 240-440 AA57289 EC Tl 1714±33 AD 240-420 AA57299 EC Tl 1716±36 AD 240-410 AA57275 EC Ch2 excL6 1744±40· AD 130-420 120 BC-AD AA57311 LPC Tl 1944±71 250 AA57291 LPC T2 2096±33 200 BC- AD 0 AA57276 LPC Ch2 excL7 2120±34 350-40 BC AA57306 LPC Tl 2156±34 360-60 BC AA57308 LPC T2 2130±34 360-40 BC AA57312 LPC T2 2135±32 360-50 BC AA57309 LPC Tl 2275±34 400-200 BC AA57313 LPC Tl 2295±34 410-200 BC AA57314 LPC Tl 2309±37 410-200 BC AA57298 MPC Tl 2339±42 800-200 BC AA57302 MPC Tl 2432±33 600-400 BC AA57300 MPC Tl 2465±33 770-400 BC AA57296 MPC T2 2517±37 800-510 BC AA57278 EPC Ch2 excL9 2755±35 1000-820 BC AA57279 EPC Ch2 excLI0 2760±34 1000-820 BC Beta170518 EPC Ch2 excL12 2780±40 1010-820 BC AA57277 EPC Ch2 excL8 2826±34 1130-890 BC AA57280 EPC Ch2 excLll 2865±33 1190-920 BC AA57282 EPC Ch2 excL13 2847±34 1320-910 BC AA57281 EPC Ch2 excL12 2931±62 1320-930 BC EMPC=Early LC=Late EC=Early LPC=Late MPC=Middle Middle Classic Classic Preclassic Preclassic Preclassic Table 1. AMS dates for the Maya Levels of Chechem Ha Cave. Late Classic gap in cave use higWighted 231 The Archaeology ofChechem Ha Cave AD 514 Caracol Stela 13 AD 531 Caracol Stela 15 AD693-698 Naran" 0 Stela 22,29 Table 2. Selected events in the eastern Maya lowlands recorded in epigraphic texts from A.D. 514-700. Time spanning Hiatus 3 at Chechem Ha highlighted in table. In a recent publication Brady and have been allied with Naranjo and that the Colas (2005) demonstrate from epigraphic cave was blocked as a defensive maneuver. evidence that caves can be places for aggression in war events. Three seventh Conclusion century panels that were looted and thought Although methods that relied heavily to have come from the Piedras Negras on ethnographic analogy have been vitally region, record the story of the ruler Nikte important in understanding the sacred Mo' who scatters fire into the cave of ruler meaning of the cave context, it is the K'ab Chante. Following this event Nikte archaeological record that offers information Mo' is beheaded and the authors interpret the on how caves were used by the ancient fire scattering event as act of war. Maya. These data can be useful in broader Although we do not know exactly archaeological contexts. By shifting the where Maya wars were fought it makes focus of cave study to a behavioral approach sense that soldiers moving between the sites it becomes clear that caves were not may have frequented rural areas between the regarded exclusively as ritual spaces but as two creating a war zone. Because the cave political spaces as well. Correlating periods was re-opened after the fall of Caracol, this of usage and abandonment with events suggests that the users of Chechem Ha may occurring in the local and regional areas 232  Holly Moyes sheds light on the nature of cave use and Occasional Paper No.1, edited by Jaime J. demonstrates the importance of the cave to Awe, pp. 1-22, University of New Hampshire, Durham. the local community. These sites were not just venues for worshipping rain deities but Awe, Jaime 1., Sherry Gibbs, and Cameron Griffith were important political spaces that required 2005 Stelae and Megalithic Monuments in protection from enemies. These data help to the Caves of Western Belize. In In the Maw integrate ritual cave use with local and of the Earth Monster: Mesoamerican Ritual Cave Use, edited by James E. Brady and regional sociopolitical events and with them Keith M. Prufer, pp. 223-248, University of in hand it is possible to better understand Texas Press, Austin. ritual practice within the social and natural environmental contexts. Brady, James. E. and Keith M. Prufer 2005 Introduction: A History of Mesoamerican Cave Interpretation. In In the Acknowledgements I would first like to Maw of the Earth Monster: Mesoamerican thank Dr Jaime Awe, the Director of the Ritual Cave Use, edited by James E. Brady Institute of Archaeology and principal and Keith M. Prufer, pp. 1-18, University of investigator for the Western Belize Regional Texas Press, Austin. Cave Project (WBRCP). I would also like Brady, James E. and Pierre R. Colas to thank members of the Belizean Institute 2005 Nikte Mo' Scattered Fire in the Cave of of Archaeology. Appreciation is extended to K'ab Chante: Epigraphic and Archaeological the Morales and Plytez families; to the 2003 Evidence for Cave Desecration in Ancient crew: Jim Aimers, Mark Aldenderfer, Maya Warfare. In Stone Houses and Earth Nathan Craig, Tony Menoa, Grant Polley, Lords: Maya Religion in the Cave Context, edited by Keith M. Prufer and James E. and Connie Price. Thanks to Joseph Ball Brady, University Press of Colorado, and Jennifer Taschek for their assistance Boulder. with problematic ceramics and to Anabel Ford for the use of the BRASS spatial data. Curtis, Jason H., David A. Hodell, and Mark Brenner Various aspects of the project were funded 1996 Climate Variability on the Yucatan Peninsula (Mexico) during the Past 3500 by an NSF dissertation improvement grant Years, and Implications for Maya Cultural to Dr. Ezra Zubrow, the Foundation for the Evolution. Quaternary Research 46:37-47. Advancement of Mesoamerican Studies, the Cave and Karst Conservancy, the Cave Gifford, James C. Research Foundation, and the Mark 1976 Prehistoric Pottery Analysis and the Ceramics of Barton Ramie in the Belize Diamond fund at the University at Buffalo. Valley. Memoirs of the Peabody Museum of Archaeology and Ethnology 18, Harvard References Cited University, Cambridge, Mass. Ashmore, Wendy, Sam V. Connell, Jennifer J. Ehret, Gill, Richardson B. L. Theodore Neff, and Jon C. Vandenbosch 2000 The Great Maya Droughts: Water, Life, 1994 The Xunantunich Settlement Survey. In and Death. University of New Mexico Xunantunich Archaeological Project 1994 Press, Albuquerque. Field Season, Manuscript on file, Xunantunich Archaeology Project. Hill, Carol A., and Paolo Forti 1986 Cave Minerals of the World. National Awe, Jaime J. Speleological Society, Huntsville, Alabama. 1998 The Western Belize Regional Cave Project: Objectives, Context, and Problem Ishihara, Reiko. Orientation. In The Western Belize Regional 2000 An Investigation of the Ancient Maya Cave Project: A Report of the /997 Field Ritual Cave Activity at Actun Chechem Ha, Season, Department of Anthropology 233 The Archaeology ofChechem Ha Cave Cayo District, Belize. Unpublished Honors Thesis, University of Tsukuba, Japan. LeCount, Lisa J. 2004 Looking for a Needle in a Haystack the Early Classic Period at Actuncan, Cayo District. In Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2003 Belize Archaeology Symposium, edited by Jaime J. Awe, John Morris, Sherilyne Jones, pp. 27-36. Institute of Archaeology, National Institute of Culture and History, Belmopan. Moyes, Holley 2002 Excavations at Chechem Ha Cave: Preliminary Results of the 2002 Field Season. Manuscript on file at the Institute of Archaeology, Belmopan. Moyes, Holley 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 Robin, Cynthia, William D. Middleton, Santiago Juarez and Mary K. Morrison 2004 Surveying an Agrarian Community: The 2002 Field Season at the Chan Site, Belize. In Research Reports in Belizean Archaeology Vol.J: Archaeological Investigations in the Eastern Maya Lowlands: Papers of the 2003 Belize Archaeology Symposium, edited by Jaime J. Awe, John Morris, and Sherilyne Jones, pp. 38-48, Institute of Archaeology, National Institute of Culture and History, Belmopan, Belize. Webster, James W. 2000 Speleothem Evidence of Late Holocene Climate Variation in the Maya Lowlands of Belize Central America and Archaeological Implications. Unpublished Ph.D. dissertation, University of Georgia, Athens. Willey, G. R. 1987 Essays in Maya Archaeology. University of New Mexico Press, Albuquerque 234