Identifying the Chan Site and its Households

The Chan site was first identified by archaeologists in 1994, led by Wendy Ashmore, who were conducting settlement survey research in the region around the nearby polity capital of Xunantunich, which is located 4 km to the northwest of Chan (see Figure 1). The site had been long known to nearby residents and is named after its landowners—Dons Ismael and Derric Chan.

In order to explore interfluvial settlement, the Xunantunich settlement survey team extended a survey transect (Transect Archaeological 1) measuring 400 m wide and 8 km long, southeast from Xunantunich, which is situated along the Mopan branch of the Belize River, to the center of Dos Chombitos, situated along the Macal branch of the Belize River. This survey transect identified a settlement concentration and small site center, now known as Chan, equidistant (4 km) between Xunantunich and Dos Chombitos—a typical spacing distance for major and minor centers in the Belize Valley (see Driver and Garber Reference Driver, Garber and Garber2004). The Chan settlement cluster is both intuitively visible along the survey transect (Figure 5) and statistically identifiable based upon nearest neighbor and stem-and-leaf analyses (see VandenBosch in Ashmore et al. Reference Ashmore, Connell, Ehret, Gifford, Neff, VandenBosch and Leventhal1994).

Figure 5.

Transect Archaeological 1 of the Xunantnich settlement survey showing the settlement cluster identified as Chan.

To begin assessing variability in Chan's settlement, investigators grouped mounds and classified those mound groups into a seven-tiered typology (Table 3). A “Mound Group” was defined as one or more mounds and associated archaeological features (excluding agricultural terraces) in which individual features within the group were less than 25 m distant from one another and all other, non-associated features were more than 25 m distant. The 25 m cutoff point was derived both from archaeological and ethnographic evidence for the extent of ancient and contemporary Maya house lots (Ashmore et al. Reference Ashmore, Connell, Ehret, Gifford, Neff, VandenBosch and Leventhal1994; Hanks Reference Hanks1990; Killion Reference Killion and Killion1992; Neff et al. Reference Neff, Robin, Schwartz, Morrison, Leventhal and Ashmore1995). Excavations at 10% of Chan's mound groups consistently indicated that the Mound Group was a good proxy for the ancient household unit, as discussed further below for Chan's leading households (also see Blackmore Reference Blackmore and Robin2012; Hearth Reference Hearth and Robin2012; Kestle Reference Kestle and Robin2012; Robin Reference Robin2002, Reference Robin2013). The mound group and adjacent agricultural terraces is the best proxy for the ancient farmstead unit (Wyatt Reference Wyatt and Robin2012).

Table 3.

Mound-Group typology for the Chan site, Belize

The seven-tiered Mound Group typology was initially developed by Jennifer Ehret on the Xunantunich Settlement Survey (Ashmore et al. Reference Ashmore, Connell, Ehret, Gifford, Neff, VandenBosch and Leventhal1994). It utilizes four criteria to classify mound groups including: (1) the number of mounds and platforms, (2) the height of mounds and platforms, (3) the formality of mound arrangement, and (4) the presence or absence of a focal group. The smallest and least formal Type 1 mound group consists of isolated mounds and associated features that are 1 m or less in height. The largest and most formal (Type 7) mound group consists of a platform group with four or more mounds, at least one of which is 5 m in height. Type 2 through 6 Mound Groups represent variation in terms of size and formality between these two extremes.

There is only one Type 7 Mound Group, Chan's Central Group, as discussed above (see Figure 3). Type 5 and 6 Mound Groups represent Chan's largest residentially focused platform groups. These are found either adjacent to the Central Group or dispersed across the community, located roughly 700 m to 1 km from the Central Group. This article will focus on the two Type 5 and 6 Mound Groups located adjacent to the Central Group, which have been identified as the residences of Chan's leaders. Figure 6 shows Chan's Type 7 Central Group (C-001) and the two adjacent Type 5 and 6 Mound Groups, C-002 and C-003. Blackmore (Reference Blackmore and Robin2012) identified that the larger residential platforms dispersed throughout the site are the households of head residents of subgroups or neighborhoods across Chan.

Figure 6.

Chan's community center (Central Group and West Plaza) and adjacent leading residences (C-002 and C-003). Chan's largest leading residences, Structure 2 in the Central Group and Structure 17 in Group C-003, are marked.

Chan in a Belize Valley Context

The Belize Valley region where Chan is located was a peripheral and provincial part of the Maya world throughout most of its history (Garber Reference Garber2004; LeCount and Yaeger Reference LeCount and Yaeger2010; Leventhal and Ashmore Reference Leventhal, Ashmore and Garber2004). During the Preclassic, Early Classic, and early Late Classic periods (spanning 800 b.c.–a.d. 670), numerous mid-sized centers jockeyed for power across the region. These centers were organized as competitive peers, but none became a paramount center in the region (Ashmore Reference Ashmore, LeCount and Yaeger2010; Ball Reference Ball, Sabloff and Henderson1993; Ball and Taschek Reference Ball and Taschek1991; Houston et al. Reference Houston, Stuart, Taube and Kerr1992; Leventhal and Ashmore Reference Leventhal, Ashmore and Garber2004; Reents-Budet Reference Reents-Budet1994; Taschek and Ball Reference Taschek, Ball and Kerr1992). Many of these centers have been the subject of extensive archaeological research or are currently under investigation so their political histories are better understood. Among these mid-sized centers are Actuncan (LeCount Reference LeCount2004), Baking Pot (Audet and Awe Reference Audet and Awe2004), Blackman Eddy (Garber et al. Reference Garber, Brown, Awe, Hartman and Garber2004), Buenavista (Ball and Taschek Reference Ball, Taschek, Inomata and Houston2001; Yaeger et al. Reference Yaeger, Cap and Peuramaki-Brown2009), Cahal Pech (Awe Reference Awe1992), Guacamayo (Ashmore Reference Ashmore, LeCount and Yaeger2010), Las Ruinas (Ball and Taschek Reference Ball and Taschek1991), and Xunantunich (Leventhal and Ashmore Reference Leventhal, Ashmore and Garber2004) (see Figure 1 for the locations of these sites).

As documented by the Xunantunich Archaeological Project, at the end of the Late Classic period (a.d. 670–800/830) the western portion of the upper Belize Valley was unified under the short-lived and late flourishing Xunantunich polity capital (LeCount and Yaeger Reference LeCount and Yaeger2010; Leventhal and Ashmore Reference Leventhal, Ashmore and Garber2004). Located 4 km to the southeast of Xunantunich, Chan is located within a few hours walk from this major center and made up part of the late Late Classic Xunantunich polity. Xunantunich likely served as a regionally focal ritual venue for at least certain members of the Chan community. The late Late Classic was also the period of maximum population and agricultural expansion at Chan. That there would be some political interaction between Chan and the polity capital of Xunantunich is perhaps unquestionable given their proximity and settlement histories. The question that emerges is: What was the nature of their political relationships? Through time, Chan interacted with numerous Belize River valley centers in complex and overlapping relations of influence and authority. The late intrusion of the polity capital of Xunantunich into the long-lived history of Chan provides an opportunity to explore how a farming community might be transformed by its interaction with a polity capital and as well how a rising polity capital might have needed to accommodate to pre-existing contexts in farming communities. As we will develop in this article, across Chan's long history, residents developed resources and production, and complex webs of sociopolitical and economic relations with numerous centers within and beyond the Belize River valley. These resources, production, and sociopolitical relations, provided a constraint on the network of power that the rising Xunantunich polity capital hoped to dominate, and indeed ultimately achieved for only a short period of time.

Placement of Chan's Leading Residence in the Central Group

Beginning in the Middle Preclassic period (650–350 b.c.), Chan's residents manipulated cosmological principles in designing and building their community center (Robin et al. Reference Robin, Meierhoff, Kestle, Blackmore, Kosakowsky, Novotny and Robin2012a). Residents oriented the Central Group to the cardinal directions (see Figure 4). On the east and west sides of the plaza of the Central Group sit the two pyramidal structures of its central religious buildings and ancestral shrine (Structures 5 and 7), to the north sits the leading residence (Structure 2) and, to the south, its administrative building (Structure 6).

As Ashmore (Reference Ashmore1991) documents for larger civic centers across the Maya area, in constructing their cities and placing particular types of buildings in specific cardinal directions, Maya royalty referenced cosmological understandings to make powerful political and ideological statements through the architecture of a city (see also Ashmore and Sabloff Reference Ashmore and Sabloff2002; Hutson and Welch Reference Hutson and Welch2014). East and west within a city spatially mark the path of the sun and the location of the rising and the setting of the sun as seen on the horizon. Maya royalty often marked the east and west sides of plazas with ritual structures that commemorated these significant natural cycles as well as agricultural cycles. Along the sun's daily path, north (up) marks the sun's height at midday, associating north with the heavens and a position of supreme power. South (down) is associated with a worldly and underworldly position as the sun passes through the underworld at night. Across the Maya area, leaders of the largest Maya cities claimed the powerful north location for the construction of their residences or monuments commemorating their power, to make a vivid and visual statement about that power. The worldly affairs of administrative buildings are often found in southern locations.

At Chan, in the Middle Preclassic period, the north point of the Central Group was marked by a cache containing a single blue-green teardrop-shaped jade adorno. This small piece of jade marked the center of the north residence (Structure 2), which was at that time a 25 cm-high stone structure. While the north residence grew in size through time, ultimately rising to a height of 1.5 m, the position of the jade adorno remained the center point of the structure across its subsequent history.

Chan is certainly not unique among smaller centers in its manipulation of cosmological principles in site planning. Hutson and Welch (Reference Hutson and Welch2014) also demonstrate how the residents of Kancab, located in the northern Maya lowlands, constructed cardinally-directed causeways to focus their community around a sacred quadripartite and quincunx pattern. From small Maya centers such as Chan and others, to the largest royal centers, residents employed similar directional principles and spatial layouts to define community centers and invest salient social meanings in spaces to define Maya places. When similar social ideals appear at small and large sites, some archaeologists tend to argue that the small site is mimicking or emulating the ideas developed at the larger site. But the historical depth of the planning principles employed in the construction and reconstruction of Chan's community center shows that many of the ideas that later in time became part of a Maya noble and royal ideology and political strategy were, in fact, initially conceived by farming communities.

Houses

Although Structure 2 in the Central Group and Mound Groups C-002 and C-003 were among Chan's largest residences, with stone basal structures measuring over 1 m tall, perishable houses with thatch roofs surmounted these structures (see Figure 7). As mentioned previously, only one residence, Structure 17, had low (50 cm high) masonry foundation walls that supported full perishable walls. Chan's leaders, like its other residents, lived in perishable houses (Figure 8).

Figure 8.

Reconstruction of a farming residence at Chan, Belize, dating to the Late Classic period, based on evidence from household CN1. The perishable superstructure is composed of pole and thatch that sits upon a stone platform. Illustration by Jack Scott.

The perishable house construction of all residences at Chan at first seems unremarkable. It could perhaps be used simply to classify Chan's residents as situated at the lower end of the social spectrum in Maya society. But upon reflection, it tells us a great deal more about social life in the community. The variable elevation of stone house platforms across the community was a visible way that residents perceived and signaled to one another, on a daily basis, social differences within the community. But equally visible each day were the social commonalities among residents that cross-cut these differences: every resident from community leader to farmer, across 274 households in a 3.2 km² area, lived in a perishable house with a thatch roof. This point becomes even more striking in comparing Chan to neighboring communities.

Across the Maya area housing ranged in type from perishable constructions, such as those seen at Chan, to residences that have low or partial stone foundation walls that supported higher perishable walls and thatch roofs (like Structure 17 at Chan), residences that had full stone walls and thatch roofs and, finally, to the most elaborate residences completely made of stone, with stone walls and corbel vaulted roofs. That a full range of housing types is found at large civic centers across the Maya area, but not at Chan, perhaps goes without saying. But even when comparing Chan to other smaller communities in the Belize Valley area, one sees a greater variety of housing types at other communities of comparable size.

The San Lorenzo hamlet (Yaeger Reference Yaeger, Canuto and Yaeger2000, Reference Yaeger and Smith2003), located 4 km from Chan, consists of 20 household groups—a fraction of the number of household groups at Chan (see Figure 1; hamlets are not shown here, but San Lorenzo is located just across the Mopan River from Xunantunich). San Lorenzo's smallest residences, like Chan's, were perishable constructions on low stone platforms, but larger residences had low or partial stone foundation walls that supported higher perishable walls, or had full stone walls, and at least one residence had a corbel vaulted roof. Despite being a smaller settlement, there was greater variability and more stratification in terms of housing at San Lorenzo. The Chaa Creek minor centers, CC1 (Stela Group) and CC18 (Tunchilen Group), located 5 km from Chan, are comparable to, but smaller than, Chan's Central Group and are surrounded by smaller settlement areas (Connell Reference Connell, Iannone and Connell2003, Reference Connell, LeCount and Yaeger2010) (see Figure 1). Leaders of the Chaa Creek minor centers lived in corbel vaulted residences in the Late Classic period.

Given these comparative data, it now seems remarkable that across 274 households in a 3.2 km² area at Chan, all residents lived in perishable houses. As much as this distinction is striking from a contemporary archaeological perspective, it also would have been striking to residents of Chan and various nearby communities. At Chan, leader to humble farmer lived in a perishable residence and this lesser degree of social stratification in housing types would have been visible across the community.

Craft Specialization and Luxury Items

Adjacent to Structure 2 in the Central Group are two low ancillary structures, Structures 3 and 4, consisting of low stone platforms that rose to 65 and 70 cm in height, respectively (see Figure 3). Craft production was one likely function to have taken place at Structures 3 and 4. Excavators identified obsidian debitage related to core maintenance activities suggesting low-intensity pressure blade production in a fill episode in Structure 3 dating to the Early Classic period, and a fill episode in Structure 4 dating to the Late Preclassic period (350 b.c.–a.d. 100/150) (Meierhoff et al. Reference Meierhoff, Golitko, Morris and Robin2012). Structures 3 and 4 are the only places where excavators identified obsidian blade production debris at Chan. This distribution suggests that Chan's leading residents either possessed the skills for obsidian blade production or associated themselves with attached specialists or itinerant specialists who may have worked out of these ancillary structures (Meierhoff et al. Reference Meierhoff, Golitko, Morris and Robin2012).

Chan's leaders were involved in small-scale Strombus shell ornament production in the Middle to Terminal Preclassic periods (650 b.c.–a.d. 250). Marine shell working detritus from this manufacture was identified in fill episodes in Structure 2, Structure 3, Structure 7, the Central Plaza, and Structure 8 in the West Plaza. Preclassic shell workers produced relatively simple ornaments such as perforated shell fragments and disk beads for distribution within the community (Figure 9). By the Late Classic period local shell working diminished, and Chan's leaders were distributing finely worked shell items produced elsewhere, including countersunk disks, rosettes, unperforated saucers, and an unusual toggle-bead type all made of conch (Keller Reference Keller and Robin2012).

Figure 9.

Illustrating the sequence of Strombus sp. shell bead production (scale in millimeters). Photos by Angela Keller.

Leaders were the only residents at Chan to be involved in the manufacture of exotic items, though residents elsewhere in the community were involved in types of craft production that involved local materials such as the production of chert bifaces or limestone block (Hearth Reference Hearth and Robin2012; Kestle Reference Kestle and Robin2012). Chan's leaders had quantitatively more luxury items such as jade, shell, and obsidian than other residents across the community, again marking their distinction in visible ways (Blackmore Reference Blackmore2011, Reference Blackmore and Robin2012; Keller Reference Keller and Robin2012; Meierhoff et al. Reference Meierhoff, Golitko, Morris and Robin2012). Still, all residents across the community had access to luxury items of comparable quality and style in their homes and for their bodily adornment, and this would have re-enforced commonalities across the community. Comparable access to similar marine shell ornament types of similar quality and obsidian blades from the three primary Guatemalan sources, Ixtepeque, San Martin Jilotepeque, and El Chayal (Keller Reference Keller and Robin2012; Meierhoff et al. Reference Meierhoff, Golitko, Morris and Robin2012), are two examples (Figure 10). As seen in housing, while the larger quantity of luxury items leading residents possessed indicates their high status, the comparability of the types of luxury items possessed by all of Chan's residents suggests a heterarchical arrangement of difference in terms of access to luxury goods. Again, this distribution differs from that seen at neighboring small communities. At San Lorenzo higher-status residents had access to more elaborate shell ornament types than lower status residents (Yaeger and Robin Reference Yaeger, Robin, Lohse and Valdez2004). At Chaa Creek community leaders amassed locally unprecedented wealth including greenstone beads, polychrome vessels, chert and obsidian eccentrics, shell pendants, and high-quality tools and ground stone (Connell Reference Connell, Iannone and Connell2003). At Chan, all residents lived in perishable houses, goods were more equitably distributed, and social stratification was less pronounced.

Figure 10.

Examples of shell ornaments distributed across the Chan community. Illustrations by Carmen Ting.

Chan's leaders had access to a single exotic item not found elsewhere in the community: non-Guatemalan obsidian (from Pachuca and Paredon, in Mexico, and La Esperanza, in Honduras; these sources are located anywhere from 320 to 1,070 km from Chan). One hundred percent of the obsidian from Chan was examined using a portable X-Ray fluorescence (XRF) device at the Field Museum of Natural History to determine geological sources for the entire obsidian assemblage (Meierhoff et al. Reference Meierhoff, Golitko, Morris and Robin2012). The four pieces of non-Guatemalan obsidian date to the late Late Classic and Terminal Classic periods (a.d. 670–900). Because they do not occur in the assemblages of Chan's leaders prior to the late Late Classic period (when Xunantunich rose to power) and are from the most distant location of any items found at Chan, one plausible interpretation for the presence of these items is that they might be political gifts from the more regionally connected rulers at Xunantunich, who may have had access to more distant trade networks than residents at Chan. But, there is another plausible explanation: Chan's leaders had played a role in obsidian production and exchange at Chan since the Preclassic period. They may have built upon this historical role to expand their regional networks in relation to obsidian in the late Late Classic period. Thus leaders' long term relationships to distant places, rather than their emerging relationship with the Xunantunich polity capital, may account for the presence of non-Guatemalan obsidian at Chan.

Understanding what political-economic connections allowed Chan's leaders to procure distant non-Guatemalan obsidian requires the sourcing of obsidian at Xunantunich and other major centers in the Belize Valley, which unfortunately has not been undertaken to date. One of the pieces of non-Guatemalan obsidian found at Chan comes from the Pachuca source (located 1,070 km distant from Chan, the most distant item found at the site). Pachuca obsidian has a distinctive green color that allows it to be sourced visually. Thus it is possible to explore how at least this one exotic item arrived at Chan. Kindon and Connell (Reference Kindon and Connell1999) visually sourced the obsidian from the Xunantunich Archaeological Project collections, which include obsidian from Xunantunich and nearby smaller settlements. No Pachuca obsidian was identified at Xunantunich. Other than the Chan example, the only Pachuca obsidian in the Xunantunich area comes from the minor center of Chaa Creek (three pieces) and the hamlet of San Lorenzo (one piece). Researchers have also visually identified Pachuca obsidian from recent excavations at the nearby major centers of Buenavista and Cahal Pech, located in the Belize Valley (Jaime Awe, personal communication 2011; Jason Yaeger, personal communication 2011). Given that Xunantunich's rulers did not possess Pachuca obsidian, they likely did not have any special claim on the distant trade networks that brought Pachuca obsidian into the Belize Valley. Conversely, major Belize Valley centers such as Buenavista and Cahal Pech, which were not as powerful as Xunantunich in the late Late Classic period, did have access to Pachuca obsidian, as did the residents of Chan. It is impossible to know exactly how Chan's leaders acquired Pachuca obsidian. (Through long-standing political economic relationships with Buenavista, or Cahal Pech, or some other center?) Nevertheless, it is clear that it was not through political gifting from Xunantunich. Plausibly, the political and economic means through which Chan's leaders acquired Pachuca obsidian in the Late and Terminal Classic periods were a continuation of connections they had established earlier with other political centers with whom they were involved. Chan's long-term political and economic ties with a range of political centers in the Belize Valley, and even beyond the valley, seem to have provided some of the resources that allowed Chan's leaders to develop some form of political constraint on the networks of power that the late rising Xunantunich polity capital hoped to develop, and ultimately dominated, for only a short period of time.

Feasting and Festivals

Two lines of evidence suggest that Chan's leaders hosted feasts as part of their community building festivals: (1) larger quantities of ceramics from ancillary Structures 3 and 4, and (2) the deposition of an enormous amount of jute shells (freshwater snails) between the Central Group and the leading residences at C-002.

Excavators identified the highest quantities of ceramics on the surfaces of ancillary structures (Structures 3 and 4) at Chan. Kosakowsky's (Reference Kosakowsky and Robin2007) ceramic analysis also identified greater amounts of ceramics in the fill from Structures 3 and 4 than from any other structures in the Central Group. The higher quantity of ceramics at these ancillary structures suggests that at least one function of these buildings, in addition to craft work, was the preparation of food—meals for leading residents and, perhaps, festival foods for the feasts that accompanied ceremonies hosted by leaders in the Central Group.

Between the Central Group and the leading residences at C-002 are two midden areas largely comprised of jute shells, referred to as the “jute midden” (see Figure 6). On the basis of Kosakowsky's (Reference Kosakowsky and Robin2007) ceramic analysis, the jutes themselves were likely consumed in the Middle to Late Preclassic period, as the dense jute deposit is interspersed with redeposited Middle and Late Preclassic sherds. In the Late Classic period the “jute midden” was likely moved and redeposited at its current location. The “jute midden” contained 46,000 shells per m³, equaling roughly 2.7 million jute shells in total (Pachychilus indorium [96%], Pachychilus glaphyrus [2%], and Pachychilus largillierti [2%]). Keller (Reference Keller and Robin2012) likens the deposit to a vast ritual sea of jutes, drawing upon archaeological and ethnographic findings about their ritual use and symbolic meaning (see, for example, Cap Reference Cap and Robin2012; Halperin et al. Reference Halperin, Garza, Prufer and Brady2003; Jamison Reference Jamison and Leventhal1992). Using Healy and colleagues' (Reference Healy, Wright and Emery1990) nutritional data, the Chan jute midden contains roughly 23,000 meals of 500 calories each, or possibly the remains of 115 gatherings of 200 people—this size gathering would include the majority of Chan's residents in the Late Preclassic period when the jutes were consumed (Keller Reference Keller and Robin2012). Estimates for Chan's Late Preclassic population based on settlement survey and surface collection ceramic data range from 195 to 330 people (Robin et al. Reference Robin, Wyatt, Kosakowsky, Juarez, Kalosky, Enterkin and Robin2012c).

For each major period of occupation we developed minimum and maximum population estimates drawing upon a range of published procedures in Mesoamerican archaeology (see, for example, Culbert and Rice Reference Culbert and Rice1990). Population was estimated for an area with a 1 km radius around the Central Group—corresponding to the intuitively and statistically identifiable settlement concentration around that group. This area includes the core of the community and surrounding subgroups. In calculating population estimates, several variables need to be considered: number of structures, percentage of structures determined to be residential, percentage of structures dated to the phase in question, percentage of structures inhabited contemporaneously, and number of individuals living in each structure. Depending on how one estimates any one of these variables, the projected population estimate can vary widely. Given that 562 mounds were identified across the 3.2 km² Chan survey area, we estimate that there were 552 mounds within a 1 km radius of the Central Group. Using excavation data to estimate how many mounds were residential, we estimate that 436 mounds were residential. To calculate a minimum population estimate, we used the following equation: 436 residential mounds × percent dated to phase × 75% contemporaneity × 4 persons/residence. For our maximum estimate we used 436 residential mounds × percent dated to phase × 90% contemporaneity × 5.6 persons/residence. A more extensive discussion of our population estimating procedures can be found in Robin et al. (Reference Robin, Wyatt, Kosakowsky, Juarez, Kalosky, Enterkin and Robin2012c:Note 2). Population estimates allow us not only to compare populations at Chan through time, but also to compare populations at Chan to other Maya communities where similar estimating procedures have been used. We see generating ancient population estimates not as a means of getting at absolute population levels, but as a means of identifying relative population levels for comparative purposes.

The sharing of food was a means through which Chan's leaders built community and asserted their authority. The size of the Central and West Plazas also suggests that the feasts and festivals hosted by Chan's leaders were community-wide events. At the height of its construction in the late Late Classic period, the Central Plaza had an area of 1,154 m², and the West Plaza had an area of 1,928 m² (including its broad south stair). Drawing on ethnographic research on the number of people who can be accommodated in open spaces during public events, Inomata (Reference Inomata2006 [following Moore Reference Moore1996]) estimates that participants attending events in Maya plazas would require between .46 and 3.6 m²/person of plaza space. The lower number corresponds to a tightly packed event and the higher number corresponds to an event in which participants had substantial space to move around. Using these numbers to gauge a minimum and maximum number of people that could participate in events in Chan's plazas in the late Late Classic period, we suggest that between 857 and 6,700 people could have attended events there. The larger of these numbers is significantly more than the number of residents at Chan at its population peak, and the smaller of these numbers would have still included a broad spectrum of residents from across the community. Using settlement survey and ceramic surface collection data, estimates for Chan's late Late Classic population range from 994 to 1,670 people (Robin et al. Reference Robin, Wyatt, Kosakowsky, Juarez, Kalosky, Enterkin and Robin2012c). Ritual events sponsored by Chan's leaders at its community center can be correctly referred to as community-wide festivals that involved the broad membership of the community. The community-wide nature of festivals at Chan diverges from what Inomata (Reference Inomata2006:811–814) identifies at the largest of Maya ceremonial centers, such as Tikal, where plazas could only hold a select portion of community residents. Instead, it is comparable to what he (Inomata Reference Inomata2006) finds at small royal centers such as Aguateca, where, as at Chan, all residents of a community could have attended events in its central plazas.

The simultaneous presence of the entire Chan populace for community-wide festivals furthers our understanding of the power of these events in building community. Festivals were times when community members came into face-to-face contact with each other. Residents experienced together the convening with ancestors, the burning of incense, and the deposition of sacred objects that consecrated community. All of these phenomena would have given a physical and experiential reality to the Chan community.

Community-wide rituals would also have embodied the divisions and competition within the community (Blackmore Reference Blackmore2011, Reference Blackmore and Robin2012; Cap Reference Cap and Robin2012). Social differences among community members would have been as visible at these events—through what people wore and where they were able to stand to view events—as the shared community concept that they were constructing. As well, hidden in the crowded space of a public event in the Central Plazas, residents may have been able to find the privacy to voice concern or dissent (see Scott Reference Scott1990).

Across its 2,000-year history Chan's community center was a place where residents came together for community-wide ceremony and political events. While these were not daily, they were neither devoid of, nor detached from, the quotidian. Residents and ritual practitioners typically chose objects that they used in daily domestic life, not a specialized set of objects produced solely for ritual use, for placement in caches, burials, and terminal deposits. The ceramic vessels placed in ritual deposits, the most numerous items they placed in these deposits, were previously used items. They were people's possessions and containers for food and other items (Kosakowsky et al. Reference Kosakowsky, Novotny, Keller, Hearth, Ting and Robin2012). Even the luxury items ritual practitioners incorporated into deposits show evidence of use. They included shell detritus from marine shell bead production in ritual deposits in the Preclassic period (Keller Reference Keller and Robin2012). Jade objects, beads, celts, and other polished pieces were used, broken, or formerly strung. There were a few items that appear to have been specially produced for ritual use, such as a unique 38 cm-long honey brown-colored chert blade from Northern Belize and eccentric chert objects (crescents, serpents, and celestial forms) that do not show signs of use, and are not found in domestic, nonritual inventories. But, most of the objects that ritual practitioners interred in ritual deposits were items people had used in daily life.

Because the objects ritual practitioners incorporated into ritual deposits were those residents used in daily life and could typically be found in domestic inventories, ritual and everyday life were linked at Chan. Residents' concerns for the sacred and supernatural were tied to, not separate from, the experiences of everyday life, through the incorporation of ordinary objects in ritual. As community members gathered on special occasions to observe even the largest of community-wide ritual events at Chan's community center, they experienced not only a sacrosanct periodic event, but through the manipulation of daily objects in these events, recognized the way the sacred and the meaning of special events was deeply tied to the conditions of their daily lives.

Burial

In the Middle Preclassic period, a single ancestor burial was identified in the Central Plaza, presumably one of Chan's leading residents. This individual, an adult aged 20–24 of unknown sex (Burial 1), was placed in a shallow grave cut into the underlying limestone bedrock (Novotny Reference Novotny and Robin2012) (Figure 11). This grave was reentered for centuries across the Middle Preclassic and Late Preclassic periods (650 b.c.–a.d. 100/150). At each reentry the people visiting the grave deposited fragments of human bone, serpentine, jade, shell, and slate above the grave. These objects were either objects people had originally placed with the deceased and later removed and replaced during the reentries, or they were objects deposited later, at the time of the reentry to commemorate visitation of the interred individual. Regardless, residents remembered and revered the death and burial of this individual for generations and, in fact, centuries, interring them with the most offerings of any ancestor or person buried at Chan, and thus initiating a practice of remembering and revisiting ancestral burials across time at Chan.

Figure 11.

Plan map of Burial 1, Chan, Belize, from the Central Plaza of the Central Group. Drawing by Carmen Ting.

Beginning in the Late Preclassic and continuing into the Terminal Classic period, Chan's eastern and western shrines (Structures 5 and 7) became the community's primary location for ancestral burials (Figure 12). Adults and children, men and women, were among the 20 individuals buried in 14 graves that were excavated in these structures (Novotny Reference Novotny and Robin2012).

Figure 12.

Profile of the central structure (Structure 5, center) of the eastern tripartite shrine at the Central Group, Chan, Belize. Profile shows stratigraphy and voids indicate the location of graves. Illustration by Nasario Puc.

Focus on the individual ancestor as a part of ritual and political process was most marked at Chan in the Preclassic period, when ritual practitioners interred their dead in the Central Plaza and the eastern and western shrines with more grave offerings than their Classic period counterparts (Novotny Reference Novotny and Robin2012). Residents buried curated Middle Preclassic figurines, each with unique facial characteristics that may represent actual portraiture, with Late Preclassic ancestors, possibly to link the deceased with founding members of Chan (Kosakowsky et al. Reference Kosakowsky, Novotny, Keller, Hearth, Ting and Robin2012). In terms of shell ornamentation, only two shrine burials, Late Preclassic Burial 10 and Terminal Preclassic Burial 2, were accompanied by shell ornaments that marked individual identity (Keller Reference Keller and Robin2012).

In the Classic period, burial and caching in the shrines shifted from a focus on individual ancestors to a focus on the community as a whole (Robin et al. Reference Robin, Wyatt, Kosakowsky, Juarez, Kalosky, Enterkin and Robin2012c). This change is evidenced by a shift from burying ceramics and other material objects with individual ancestors to burying these items in caches and terminal deposits, foregrounding the community, rather than the specific ancestor, as the focus of ritual activity (Kosakowsky et al. Reference Kosakowsky, Novotny, Keller, Hearth, Ting and Robin2012), and an increase in the number of multiple (over single) interment ancestor burials (Novotny Reference Novotny and Robin2012). Changes in Classic period rituals in the eastern and western shrines also indicate a heightened focus on highlighting the community as a whole in ritual practices.