Recent Investigations into Ancient Maya Domestic and Ritual Activities at Pook’s Hill, Belize

Introduction A fifth season of excavations was conducted between June and September 2005 at the ancient Maya site of Pook’s Hill, Belize (Helmke 2006a). The excavations were conducted as part of the Belize Valley Archaeological Reconnaissance Project (BVAR) under the direction of Jaime Awe (Belize Institute of Archaeology). The site is located in the Roaring Creek Valley of central Belize, in the Cayo District, approximately 14km southwest of the country’s capital, Belmopan (Fig. 1). The site is situated amidst the karstic foothills to the granitic Maya Mountains at an approximate elevation of 78m above mean sea level.


Introduction
A fifth season of excavations was conducted between June and September 2005 at the ancient Maya site of Pook's Hill, Belize (Helmke 2006a). The excavations were conducted as part of the Belize Valley Archaeological Reconnaissance Project (BVAR) under the direction of Jaime Awe (Belize Institute of Archaeology). The site is located in the Roaring Creek Valley of central Belize, in the Cayo District, approximately 14km southwest of the country's capital, Belmopan (Fig. 1). The site is situated amidst the karstic foothills to the granitic Maya Mountains at an approximate elevation of 78m above mean sea level.

Site Description
The principal complex of Pook's Hill is a residential site with masonry structures of various functions facing onto and delineating a quadrilateral plaza. Such residential groups are known as plazuelas and follow Maya architectural practices, in being quadrangular and aligned to major cardinal directions (Ashmore 1981, 49-54). The plazuela has been termed "Pook's Hill Group 1" (or PKH1) and now that it has been extensively cleared it is known to consist of the remains of nine masonry building platforms (cf.  Helmke 2003: 119;2006a: 42-43) (Fig. 2). The largest building (Structure 1A), measures 16.5m long (east-west) by c. 5.4m wide (north-south) and is over 2.5m high (above the ancient plastered plaza). The group as a whole encompasses an area of approximately 1106m 2 of which the plaza occupies just over 430m 2 .

Objectives
The objectives of the 2005 season were threefold: 1) to continue documentation of ancient Maya domestic and ritual activities at an intermediate-sized household group in the Roaring Creek Valley; 2) to fully excavate and ascertain the function of a presumed sweatbathing building; and 3) to complete architectural restoration works of five of the site's buildings that were scheduled for consolidation (as part of tourism development efforts) (Helmke 2006a: 39). The two most important finds made during the course of the 2005 season of investigations are the focus of this article, namely a structure that has been positively identified as a sweatbath (Structure 1B) and an interment associated with the largest building at the site (Structure 1A).

Sweatbath
In 2001, excavations at Pook's Hill partly exposed the frontal facade of Structure 1B. The architectural features discovered in association with the exterior of that building were strongly reminiscent and compared favourably in terms of type and dimensions to roughly coeval sweatbathing structures at the important sites of Piedras Negras and Tikal in Guatemala (Satterthwaite 1952;Jones 1996: 75-77;Helmke 2006a: 53-54;2006b: 13, Fig. 6). In order to test the presumed sweatbathing function of that building in 2005, we completely cleared the interior of the once vaulted room of Structure 1B (Fig. 3). Based on the many features consistent with sweatbaths uncovered it can now be conclusively stated that the vaulted room of Structure 1B functioned primarily as a sweatbath (see Helmke 2006a: 53-68). Though most of the vault had collapsed in  (1999)(2000)(2001)(2002)(2003)(2004)(2005).
antiquity, we found that the sweatbath comprised a well-preserved, ovoid room (measuring 2.25m wide by 2.85m long). Based on well-preserved portions, it is clear that the room would have had a domed corbel vault (approximately 2.35m in height; Helmke 2006a, 61-62, 64-65, Figs. 9, 13). All features of the sweatbath have heat promotion and retention as the basis of their design. The smallness of the room (5.3m 2 and c. 9.6m 3 ) would have amplified heat and is entirely in keeping with other Lowland sweatbaths (which range between 2.7m 2 and 21.3m 2 ) (Helmke 2006a: 66, table 4). Similarly, the low doorway into the room (98cm high) was designed to retain heat within, while the two high masonry benches that dominate the inside would have raised bathers to the warm air gathered in the upper reaches of the room. In the rear of the room we also exposed the 'hearth' and 'firebox' of the sweatbath, where heated stones were sprinkled with water to produce steam. The rapid cooling of stones, brought about by contact with water, induced pervasive cracking and spalling of the masonry framing the hearth. Water was also used by sweatbathers to wash and rinse as is suggested by the bench Figure 3. Plan of the terminal phase architecture of the sweatbath of Structure 1B and parts of the adjoining Structures 1A and 2A. The extensive diagonally-hatched area represents the artefactual deposit that was deposited during the terminal phase of the site's occupation. Note how the deposit filled the sunken passage and obstructed the entrance to the circular sweatbathing room, thereby terminating its primary function.
and floor surfaces that are noticeably graded towards the entrance to promote drainage. The exterior features of the sweatbath also include a so-called 'sunken passage' which would have drained runoff from the sweatbath, southward onto the plastered plaza. The two high exterior benches that frame the 'sunken passage' in turn would have been places where bathers could disrobe prior to bathing and rinse-off thereafter (Helmke 2006a: 59, 61).
Structure 1B stands apart from all others at Pook's Hill not only in terms of its primary function but also since it is one of the latest buildings constructed at the site (Helmke 2006a: 67-68, 81-82). As a late addition, dated to the Terminal Classic (c. AD 830-950), it nonetheless witnessed considerable usage based on re-plastering of interior bench surfaces as well as the wear-polish documented along the verges of the interior benches and its floor. The primary function of the sweatbath was eventually terminated late in Figure 4. Plan of Burial 1A-1. Note the halved shell inkwell (no. 14) behind the cranium, the fragmented carved bone plaque (shaded in light grey) (nos. 7 and 11), and the pyrite mosaic mirror (no. 8) in relation to the poorly-preserved skeleton.
The two limestone capstones that sheltered the cranial area are marked by dashed outlines.
the site's history by an extensive artefactual deposit (Fig. 3). The deposit had been cast into the vaulted room (prior to its collapse) and filled the sunken passage thereby partly obstructing entry into the room. Though still the focus of ongoing analyses, comparable deposits have been documented at several other sites, including Aguateca and Tikal in Guatemala (Harrison 1999, 193-198;Inomata et al. 2001, 292, 296-297).

Burial 1A-1
Ritual activity was signalled by Burial 1A-1, the sole 'special deposit' encountered in 2005. The interment was discovered as part of trenching excavations aligned to the 'primary axis' (Loten and Pendergast 1984: 3) of Structure 1A, which were designed to document construction sequences and test for the incidence of votive dedicatory deposits. The trench (2m east-west by 4.3m north-south) penetrated into the core of the terminal outset Stair 1 (Str. 1A-1 st ) and revealed the presence of an earlier engulfed Stair 2. At present, it remains unknown if Stair 2 represents a 'construction stair' (see Loten and Pendergast 1984: 6) to the terminal phase architectural refurbishments (Str. 1A-1 st ) or the stair to the penultimate phase construction (Str. 1A-2 nd ) (Helmke 2006a: 72-73). It was in the core of Stair 2 that we found Burial 1A-1. The grave was built directly atop unmodified bedrock as a cist with two large marl capstones spanning the cranial area (Fig. 4). The skeleton was in a prone position, oriented north-south, with head to the south, as was customary at Pook's Hill and the greater Belize Valley from Middle Formative times (600 BC onwards) (see Helmke 2003, 122). The skeletal re- mains of the interment were poorly preserved on account of the effect of bone against limestone bedrock but the interment was notable for its artefactual inclusions. The burial contained, in addition to several items of personal adornment, a halved conch shell that may have served as a pendant and portable inkwell (Figs. 4.14 and 5a) (see Coe and Kerr 1997: 101, 105, 150-151;Reents-Budet 1994: 36-38, 42-43), a fragmentary carved bone plaque and a complete circular mosaic mirror (Helmke 2006a: 75-80;Stanchly 2006: 107). The carved bone plaque represents an iconographic panel showing an anthropomorphic figure, holding what may be a feather-tipped blood-letter, seated cross-legged, within a lunar ancestor cartouche (Figs. 4.7, 4.11 and 5b). The absence of ceramic containers in the interment complicates matters of dating, though based on the dates ascribed to adjoining architecture and the style of the carved bone plaque, the burial can be provisionally dated to the Late Classic (c. AD 700-830) (Helmke 2006a: 69-70, 80-81). The mosaic mirror is one of the most complete of its kind and is composed of a solid, smoothed slate backing, to which 45 iron pyrite tesserae were adhered (Figs. 4.8 and 5c). The position within the burial and the perforations of the slate backing indicate that the mirror was worn as an item regalia on a belt assemblage at the small of the back in keeping with burial evidence from other sites and iconographic depictions thereof (Figs. 4.14 and 5d) (Miller and Taube 1993: 114-115;Taube 1992: 172-177).

Discussion and Conclusions
The finds made at Pook's Hill in 2005 provide a varied picture of local social processes during the 'collapse' period that characterises the Terminal Classic period (c. AD 830-950). In this regard, the sweatbath and the burial are particularly illustrative. What sets the sweatbath of Pook's Hill apart is its circular plan, which contrasts to the quadrilateral examples seen at the important centres of Piedras Negras, Palenque, Tikal and Copan (Cheek 2003: 134-135;Jones 1996: 75-77;Robertson 1985: 79-80;Satterthwaite 1952). The circular shape is, however, seen in the earliest domed examples (Hammond and Bauer 2001) as well as in modern cases in the Mexican and Guatemalan Highlands (see Satterthwaite 1952). The pattern that emerges is that of two competing sweatbathing traditions, one regally-sanctioned and associated with monumental rectangular sweatbaths (i.e. c. 11-24m 2 ), versus a vernacular tradition of great longevity and geographic breadth, often associated with smaller domed sweatbaths (i.e. 3-8m 2 ) (Helmke 2006a, 66-67). As such, the Classic period examples of domed sweatbaths (Pook's Hill for example), may be seen as direct precursors the ones that have been typical in Mesoamerica since the period of Spanish contact (see Arreola 1920;Cresson 1938: 90-96;Satterthwaite 1952: 7-10;Hammond and Bauer 2001: 684;Helmke 2006a: 67).
In turn, one of the most interesting aspects of Burial 1A-1 is the placement of the mosaic mirror as worn by the deceased. The practice of wearing mirrors at the small of the back is one of many traditions introduced to the Maya area in the late Early Classic (AD 375 onwards) under the influence of Mexico's great and distant metropolis Teotihuacan (Taube 1992: 172-174;Miller and Taube 1993: 114;Nielsen 2006: 5, 6). With the waning of Teotihuacan's power and its eventual abandonment in the seventh century, many Teotihuacan-inspired practices fell into disuse. A great revivalism in Teotihuacan iconography and regalia has, however, been noted at major centres such as Tikal, Piedras Negras and Copan during the early part of the eight century (Martin and Grube 2000: 45, 142, 207-208;Stone 1989;Stuart 2000: 490, 495-498;2005: 387-393). As Burial 1A-1 has been dated to the Late Classic it is contemporaneous to the revivalistic practices currently seen elsewhere in the Lowlands and thus can be deemed to be a local adherence to these trends.
Though the results presented here are provisional, it is hoped that with continued analyses these finds will contribute to our understanding of the particular social processes involved in the transitional Terminal Classic period in this area of the Maya world.