Early (300−100 B.C.) temple precinct in the Valley of Oaxaca, Mexico

Significance

Recent excavations at the site of El Palenque have recovered the earliest-known temple precinct in the Valley of Oaxaca, Mexico, dating to the Late Monte Albán I phase (300–100 B.C.). This precinct exhibits characteristics similar to the temple precincts of 16^th century Mesoamerican states. A walled enclosure contains differentiated temples, priests’ residences, and ritual features. We propose that the precinct’s components represent a hierarchy of temples staffed by a specialized full-time priesthood. A series of radiocarbon dates indicate that the El Palenque temple precinct was in use during the 300–100 B.C. period of archaic state emergence in Oaxaca.

Abstract

Archaeological investigations during the past two decades in Mexico’s Valley of Oaxaca have documented the appearance of key public buildings, such as the royal palace and multiroom temple, associated with the rise of an archaic state at ca. 300−100 B.C. A fuller picture is now emerging from the site of El Palenque, where recent excavations have defined a temple precinct on the east side of the site’s plaza. This precinct exhibits characteristics similar to those of the temple precincts of later Mesoamerican states described by Colonial period sources. The excavation data document a walled enclosure containing three multiroom temples, two special residences identified as priests’ residences, and an array of ritual features and activity areas. The temple precinct’s components are interpreted as comprising a hierarchy of temples staffed by a specialized priesthood. A series of radiocarbon dates indicate that the precinct’s differentiated components were all in use during the 300−100 B.C. period of archaic state emergence. The El Palenque temple precinct is the earliest temple precinct excavated thus far in the Valley of Oaxaca.

When archaeologist V. Gordon Childe (1) proposed his defining criteria of urban civilization, he highlighted monumental public buildings, especially temple precincts. Such precincts, consisting of one or more temples and their attached facilities, were considered an expression of occupational specialization within society, which he saw as a hallmark of civilization. Building upon Childe’s perspective, Adams (2) and Flannery (3) focused on the sociopolitical institutions associated with the state as a form of government. These authors characterized the state as highly centralized and hierarchical, managed by a professional ruling class whose legitimacy was often reinforced through the tenets and practices of a state religion maintained by full-time priests.

Adams (2) traced the development of Mesopotamian temple precincts in terms of the monumental scale of construction and increasing differentiation in function. The temple precincts at Uruk, the capital of a first-generation state, tended to be set apart from the community by an enclosure wall. The temples and temple precincts were variable in morphology and presumed function, and included attached living quarters for their attendant personnel.

Flannery (4) examined the archaeological indicators of early state institutions and proposed that a state religion was frequently associated with standardized, multiroom temples and special residences for full-time priests. Such priestly living quarters were often attached to the temples, or in close proximity behind or to one side of the temples, which the priests accessed by entryways or stairways different from those used by the public. Access to them was restricted and privacy achieved by narrow corridors and curtain walls. In some cases, priestly residences were similar in size and layout to each other, and shared the same astronomical orientation as the temple. Flannery (4) characterized the priestly residences of first-generation states as more modest in size and degree of architectural elaboration than those of second- and third-generation states. He pointed out that the full range of residences associated with high priests and minor priests remained to be identified in any archaic state.

Here we discuss our recent (2009−2011) excavations at the site of El Palenque, near San Martín Tilcajete in the Valley of Oaxaca, home of Zapotec speakers in southern Mexico (Fig. 1). We describe the temples and nearby residences that we excavated on the east side of El Palenque’s plaza and report the associated series of ¹⁴C dates that span the Late Monte Albán I phase (300–100 B.C.). We propose that these standardized, multiroom temples and associated residences make up a temple precinct staffed by a hierarchy of full-time priests during the Late Monte Albán I phase. The evidence of a specialized priesthood at the El Palenque temple precinct is concurrent with other archaeological indicators of a first-generation state in Oaxaca, currently recognized as the earliest state in Mesoamerica (5).

Fig. 1.

Mexico’s Oaxaca Valley, with Monte Albán and El Palenque indicated.

Temple Precincts of Later Mesoamerican States

Ethnohistoric accounts of the temple precincts of later Mesoamerican states underscore the architectural association of temples and the residences of the temple personnel within walled enclosures. For example, the major temple precinct in the Aztec capital of Tenochtitlán at the time of European contact consisted of eight or nine temples, close to one another within a large enclosure. Some temples were taller than others, some were more lavish than others; the grandest temple was dedicated to the principal deity (Huitzilopochtli). The neighboring multiroom temple, dedicated to Huitzilopochtli’s sister (Cihuacoatl), had an outer hall with a large stone divine brazier set into the floor, where priests received petitioners who came to offer incense. Only priests could enter the inner sanctum sanctorum, where stood a stone representation of Cihuacoatl, to whom they made offerings of incense and performed autosacrificial bloodletting regularly. Each temple had its own staircase, its special courtyard, its chambers, and adjoining sleeping quarters for the priests of the temple (6).

The priestly quarters comprised a range of accommodations according to one’s rank in the priestly hierarchy. The high priests and ordinary priests of the temple occupied adjoining palatial residences and apartments. Young priests in training were housed in monasteries consisting of many chambers. Assigned manual duties by the priests they assisted, these youths collected the firewood needed to keep the temple’s brazier burning and prepared the small blades used for autosacrificial bloodletting. Although prohibited from entering the temple, young women who served the temple resided in a separate cloister behind the temple. These maidens performed menial tasks, cooked the priests’ daily food, and spun and wove cloth. There were storerooms and granaries in the temple precinct for the foodstuffs, incense, and other offerings received by the temple, as well as the ritual paraphernalia used on ceremonial occasions (6, 7).

Similar to their Aztec counterparts, 16^th century Zapotec temple precincts consisted of two-room temples elevated on platforms, sacrificial altars, courtyards, and residences for the priesthood (and for the ruler and his retinue on occasion). The same distinction applied to the outer temple space, where many celebrants entered, versus the more sacred inner sanctum where the high priest (“great seer”) officiated (8). Only priests could enter this inner room, where they placed offerings and performed sacrifices before stone statues of supernatural figures. Best known is the temple precinct at Mitla, composed of two large courtyards enclosed by rectangular structures raised on platforms. The principal temple was reached by a broad staircase leading from the courtyard and through three entrances flanked by pillars. In the middle of the rectangular temple stood a line of monolithic columns. Priests performed divinatory and sacrificial rituals before a large stone altar on which sat the statues of supernatural figures. A narrow L-shaped corridor led from the rear of the temple to the adjoining high priest’s residence, made up of four rooms arranged around an interior courtyard. This palace of the high priest was where he lived and also held audiences with Zapotec lords who came to seek his counsel, because he served as an intermediary with the supernatural world. Ordinary priests resided in other structures bordering the courtyard, accessed by rear doors and secret passageways. These priestly residences were characterized by communal living conditions. The priests slept on mats and animal skins on the floor. The existence of monasteries adjacent to the temple for the training of young priests is recorded, as are the cloisters of young women in the service of the temple (8–10).

Zapotec temple precincts were staffed by a hierarchy of priests, ritual specialists, and servants. The high priest presided over major sacrificial rituals at the temple’s altar. Special sacrificers performed the human sacrifices. Ordinary priests oversaw the temples, received offerings from petitioners, and performed divinatory and sacrificial rituals on a nightly basis that involved burning incense, taking tobacco, sacrificing quail, turkeys, dogs, and deer, and performing autosacrifices by drawing blood from their earlobes, tongues, or other parts of their bodies. Special priests or diviners managed the ritual calendar that needed to be consulted before most decisions were reached. Young priests helped to maintain the temple and altar on a daily basis and kept the braziers burning during their nighttime vigils (9, 11, 12).

The ethnohistoric data on the temple precincts of later Mesoamerican states reveal certain common architectural, functional, and organizational elements that can serve as guidelines for the archaeological investigation of prehistoric temple precincts (4). First is the occurrence of multiple temples within a large walled enclosure, creating a separate precinct, access to which is restricted. A temple hierarchy may have existed, with a temple’s size and degree of architectural elaboration reflecting the relative importance of the supernatural to which the temple was dedicated. Temple space itself was dichotomized into an outer space, where priests could receive worshipers, and an inner space so sacred that only priests could enter. Second is the close architectural or spatial association between the temples and the special residences of the temple staff that could be accessed from the temple by rear passageways or side doors. Priestly accommodations were based on the individual’s position in the priestly hierarchy; they ranged from the palatial residences of high priests to the more communal sleeping quarters of ordinary priests and novices. Third is a hierarchy of priests, ritual specialists, and servants, who performed distinct duties within the temple precinct. The daily provisioning of the temples and priests, and their nightly ritual performances, might be detectable in the archaeological record (12).

Temple Precinct at El Palenque

El Palenque was the first-tier center of a four-tiered settlement hierarchy associated with an independent state that emerged in the Ocotlán-Zimatlán subregion of the Oaxaca Valley in the Late Monte Albán I phase (300−100 B.C.) (Fig. 1). Our investigations at the site have been directed at exposing the building plans of the key institutions of this independent state, which we have argued resisted domination by the expansionistic Monte Albán state, centered in the Oaxaca Valley’s Etla-Central subregion, until the first century B.C. (5). In Area I, on the north side of El Palenque’s plaza, we excavated (in 1997–2000) the well-preserved remains of a palace complex on the highest ground overlooking the plaza (Figs. 2 and 3). Covering an area of 850 m², the palace consisted of a spacious residential component of rooms arranged around an interior courtyard, as well as a ceremonial component of low masonry platforms and flagstone pavements that faced the plaza (13). The construction of the Area I palace is associated with a conventional radiocarbon date (β-147540) of 2,300 ± 80 B.P. or 350 ± 80 B.C., at the interface between the Early Monte Albán I phase (500−300 B.C.) and the Late Monte Albán I phase (300−100 B.C.), making it the earliest excavated Zapotec palace. The abandonment of the palace occurred in a major conflagration that produced extensive deposits of burned adobes, carbonized roof beams, charcoal, and ash. A conventional radiocarbon date (β-143355) of 1,970 ± 60 B.P. or 20 ± 60 B.C., obtained from one of these carbonized deposits in the courtyard of the palatial residence, is the latest radiocarbon date associated with the palace, corresponding to the early years of the Monte Albán II phase (100 B.C.−A.D. 200) in the first century B.C.

Fig. 2.

Topographic map of the El Palenque site, indicating the plaza, lettered masonry buildings, and lettered excavation areas. Topographic elevations in meters above site datum.

Fig. 3.

The El Palenque plaza, showing all lettered masonry buildings, including the temple precinct and its component structures (St 20, St. 16, St. 29, St. 27, and St. 28). Topographic elevations in meters above sea level.

Our excavations in 2009–2011 revealed the extent and layout of a temple precinct that borders the entire east side of El Palenque’s plaza. Three multiroom temples were arranged symmetrically and faced west toward the plaza. The stone foundations of adobe-brick walls (1.2−1.3 m in width) spanned the intervening spaces between the temples, effectively limiting access to the temple precinct, which occupied a modified natural promontory above the plaza floor. As shown in Fig. 3, the temple precinct extended over some 5,000 m². Although our analysis of the materials recovered from the temple precinct is ongoing, we can report that behind the temples lay specialized activity areas, ritual features, and two structures of a residential nature that we suspect were priests’ residences. All of the buildings in the temple precinct were aligned to N 17° E (orientations are relative to magnetic north).

The largest temple platform [Structure (St.) 20] stood at the midpoint of the plaza’s eastern side. This platform was built of adobe bricks in an initial construction event associated with a conventional radiocarbon date of 2,200 ± 50 B.P. or 250 ± 50 B.C., and a 2-σ calibrated range of 390−110 B.C., obtained from charcoal in the construction fill (β-202178) (Table 1). On the platform stood a multiroom temple, 38 × 10 m in its overall dimensions, with stone foundations of adobe walls that measured up to 1.5−2.5 m in width. A broad staircase ascended from the plaza to a long and narrow (ca. 1-m-wide) landing. Entry to the temple’s principal hall (Room 1) from the landing involved a 30-cm-tall step through one of three doorways flanked by pillars (Fig. 4). Room 1 was Structure 20’s largest interior space and also the largest room within the temple precinct; it measured 34 × 6.7 m with symmetrical wings (6.5 × 4.5 m) at either end. The lime-stucco floor, although burned, was still present throughout the eastern half of Room 1, at an elevation of some 2 m above the floor of the plaza. Abundant ritual paraphernalia were recovered on the floor of Room 1, especially around two shallow hearths [Feature (F.) 68, F. 67]: ornaments of shell, mica, and alabaster, and ceramic effigy whistles, effigy vessels, and incense braziers. The distribution of obsidian blades, an obsidian lancet fragment, and three chert perforators suggest that the officiating priests were performing autosacrificial bloodletting here and in the adjoining wings of Room 1. The remains of turkey, dove, and other fauna on the floor and in one of the hearths (F. 67) reflect other sacrificial offerings. A human premolar tooth and a tiny fragment of a possible human long bone at the northern end of Room 1 indicate that human sacrifice might have been practiced here. Another activity detected at the northern end of Room 1 and in Room 2 was the reworking of obsidian, in view of the unused obsidian flakes, angular fragments, and waste flakes recovered (14).

Table 1.

Radiocarbon dates from El Palenque, San Martín Tilcajete, Oaxaca

Lab. No.	Archaeological context	14C years B.P.	14C years B.C./A.D.	Calibrated 2-σ range
β-202178	Mound B adobe construction, subfloor deposit beneath St. 20	2,200 ± 50	250 ± 50 B.C.	390−110 B.C.
β-332860	Mound B adobe construction, subfloor deposit at western base of St. 20 (F. 92, above F. 77)	2,080 ± 30	130 ± 30 B.C.	180−40 B.C. and 10−1 B.C.
β-332859	Mound B adobe construction (above F.77), subfloor deposit beneath St. 20	1,990 ± 30	40 ± 30 B.C.	50 B.C.−A.D. 70
β-299210	Hearth (F. 82), Area B/D, SE of St. 20	2,030 ± 30	80 ± 30 B.C.	100 B.C.−A.D. 30
β-171545	Hearth (F. 65) in St. 20, Rm. 3	1,990 ± 60	40 ± 60 B.C.	160 B.C.−A.D. 130
β-299211	Mound G subfloor deposit beneath St.16, Rm. 1	2,120 ± 40	170 ± 40 B.C.	350−300 B.C. and 210−40 B.C.
β-299212	Mound G subfloor deposit beneath St. 16, Rm. 1	2,120 ± 30	170 ± 30 B.C.	340−330 B.C. and 200−50 B.C.
β-202179	Mound G subfloor deposit beneath St. 16, Rm. 1	2,050 ± 40	100 ± 40 B.C.	170 B.C.−A.D. 40
β-143353	Carbonized deposit (F. 22) on NW surface of St. 16	1,980 ± 70	30 ± 70 B.C.	165 B.C.−A.D. 155
β-332857	Directly east of rear wall of St. 29, Rm. 1	2,100 ± 30	150 ± 30 B.C.	200−40 B.C.
β-299208	Directly south of southern wall St. 29, Rm. 4	2,080 ± 30	130 ± 30 B.C.	180−30 B.C.
β-299213	Mound D subfloor deposit beneath St. 29, Rm. 1	2,060 ± 30	110 ± 30 B.C.	170 B.C.−A.D. 10
β-332858	Directly south of southern wall of St. 29, Rm. 4	1,920 ± 30	A.D. 30 ± 30	A.D. 20−130
β-288315	Mound A subfloor deposit beneath St.27, Rm. 2, F. 78	2,140 ± 40	190 ± 40 B.C.	360−290 B.C. and 240−50 B.C.
β-288311	Mound A subfloor deposit beneath St. 27, Rm. 2	2,090 ± 40	140 ± 40 B.C.	200−10 B.C.
β-288312	Mound C subfloor deposit beneath St. 28, Rm. 2	2,160 ± 40	210 ± 40 B.C.	360−90 B.C.
β-288309	On floor of St. 28, Rm. 2	2,070 ± 40	120 ± 40 B.C.	190 B.C.−A.D. 10
β-288314	Mound C subfloor deposit beneath St. 28, Rm. 2	2,060 ± 40	110 ± 40 B.C.	180 B.C.−A.D. 20
β-288316	Mound C subfloor deposit beneath St. 28, Rm. 2	2,020 ± 40	70 ± 40 B.C.	150−140 B.C. and 110 B.C.−A.D. 60

Fig. 4.

Plan of Structure 20 temple.

The southeastern corner of the Structure 20 temple consisted of a small landing (Room 3) at the top of a 2.4-m-wide staircase of three tiers that offered a separate entrance at the rear of the temple. Alongside the staircase lay a large adobe-lined cooking facility (F. 65) with a firebox on its eastern side and a circular chamber (1.5 × 1 m in area and 30-cm deep). Feature 65 was evidently used to prepare food for a group larger than the usual household, considering its significantly larger size (by a factor of 16.6) than a firebox hearth in the kitchen of a residence (St. 9) that we excavated in Area P at the site (Fig. 2). On the exterior floor around the large hearth lay two grinding stones, fragments of ceramic griddles (comales), braziers, and used obsidian flakes. Another activity performed by the kitchen staff may have been spinning cotton fiber, as suggested by the four ceramic disks, most having a drilled hole 3.5−5 mm in diameter for use as spindle weights, recovered in the kitchen area (15, 16). Several ceramic vessels characteristic of the Late Monte Albán I phase were recovered from the fill of Feature 65 (14), along with charcoal (β-171545) that yielded a conventional radiocarbon date of 1,990 ± 60 B.P. or 40 B.C. ± 60 and a 2-σ calibrated range of 160 B.C.−A.D. 130 (Table 1). This date probably pertains to the final use of the hearth before the abandonment of the Structure 20 temple.

Additional activity areas lay on either side of the Structure 20 temple, behind the walls demarcating the temple precinct. Two or three rectangular cells (ca. 2.75 by 1.75 m) abutted the southeastern corner of the temple (Fig. 4). A hearth (F. 82) lay nearby with four masonry blocks that might have originally lined its sides to form a firebox (ca. 70 × 45 cm). Charcoal (β-299210) from the hearth dated to 2,030 ± 30 B.P. or 80 ± 30 B.C. with a 2-σ calibrated range of 100 B.C.−A.D. 30 (Table 1), and probably corresponds to its final use. The midden deposits exposed here (F. 59, F. 81, F. 83) were notable for the quantities of ash and ritual paraphernalia. The deposits featured shell ornaments, mica, fragments of ceramic incense-braziers and incense-burners, a ceramic figurine portraying a standing human figure wearing an opossum mask, and an incised polished-stone celt; they also contained food remains. On the northeast side of the Structure 20 temple in Area B-NE, we exposed an inner adobe wall (0.81- to 0.95-m wide), parallel to the precinct’s outer wall and less than 1 m inside (Fig. 3). Directly east of the inner adobe wall lay an octagonal firebox lined by eight burned upright adobe bricks (F. 89), measuring 1 m in overall diameter and 20-cm deep. Its fill contained bits of charcoal, burned stones, and lime. Alongside F. 89 lay another less well-preserved burned adobe brick feature (F. 91). The principal activity in Area B-NE seems to have centered on these ritual fireboxes.

Flanking the principal temple stood a pair of smaller multiroom temples of similar size and layout. The Structure 16 temple in Area G measured 22 × 8 m overall, with stone foundations of adobe walls that measured 1.5-m wide in places. A three-tiered staircase led from the plaza to the first of two contiguous rectangular rooms (Fig. 5). This outer room (Room 2) measured 12.8 × 2.3 m, and remnants of its lime-stucco floor were evident in its northeastern corner. A step through a 1.65-m-wide doorway led to the smaller inner room [Room (Rm.) 1] of the temple. Room 1 measured 9.8 × 2.2 m, and would have stood at an elevation of ca. 0.80-m above the plaza floor. Small rooms or cubicles were accessed from the interior of Room 1; Room 3 on the north measured 3.4 × 2.2 m, and Room 4 on the south measured 2.7 by 2.2 m. Lime stucco was still present on portions of the temple’s exterior aprons, which we labeled the northwest and southwest surfaces. The 2009 excavations in Area G directly behind Structure 16 exposed a rear staircase, leading from Room 1’s midpoint. This 4-m-wide staircase of four tiers provided a separate entrance at the rear of the temple.

Fig. 5.

Plan of Structure 16 temple.

The Structure 29 temple in Area D measured 23 × 10 m overall, with stone foundations of thick adobe walls, most notably those measuring 1.1- to 1.8-m wide that enclosed the inner Room 1. A staircase composed of three steps led from the plaza into the first of two contiguous rectangular rooms (Fig. 6). This outer hall (Rm. 3) measured 12 × 3.29 m. Four adobe pilasters projected from the hall’s eastern wall: two on the north side of the single 2.8-m-wide entrance into the temple’s inner room (Rm. 1) and two on the south side. A single step led up into Room 1, which measured 9 by 3.45 m, and stood 0.80- to 0.85-m above the plaza floor. At either end of Structure 29 lay a small room. Room 2 at the northern end measured 4.2 × 4.1 m. Room 4 at the southern end was only partially excavated because of the extent of accumulated gravel from postabandonment slope wash, overlain by a dense rock pile, a by-product of recent farming in the adjacent plaza. It was possible to expose an exterior apron fronting the plaza at Structure 29’s southwestern corner, the dimensions of which were 4.4 × 1.5 m. Although the original floors of Structure 29 were poorly preserved, the recovery of lime in practically every square meter of excavation unit of Room 1, Room 3, and the exterior southwestern surface, together with the presence of white adobe fragments, allow us to propose that the Structure 29 temple rooms had whitewashed adobe walls and lime-stucco floors. A 2-m-wide staircase of three tiers, situated at the northeastern corner of the temple’s inner Room 1, gave access from the rear of the temple directly into its inner sanctum.

Fig. 6.

Plan of Structure 29 temple.

We did not recover abundant artifacts in situ from the Structure 16 temple and the Structure 29 temple, in part because of the poor preservation of the original floors; nor was there evidence of any subfloor features. Along with grayware ceramics characteristic of the Late Monte Albán I phase (17), the ritual paraphernalia recovered here (shell ornaments, a deer bone awl, obsidian blades, chert perforators, fragments of ceramic incense braziers and effigy vessels, especially from Structure 16) are evidence of the ritual activities performed in these temples (14). The fragment of the left humerus of an ungulate, possibly a juvenile deer, may represent a sacrificial offering left in the outer room (Rm. 3) of the Structure 29 temple.

These two companion temples may have been constructed a little after Structure 20 in the Late Monte Albán I phase (300−100 B.C.), based on the conventional ¹⁴C dates obtained from charcoal in subfloor deposits pertaining to their initial construction (Table 1). The three multiroom temples share a common location on the plaza’s east side, a common position facing west, and a common orientation (N 17° E). The temples also share an elongated rectangular plan consisting of a pair of contiguous rectangular rooms flanked by additional smaller rooms accessed from their inner sancta. All three temples display exceptionally thick walls, up to 1.5−2.5 m in width. The temples were accessed from the plaza by means of stairways that led into their outer halls. The three temples were also equipped with a separate entrance at the rear of the temple, with direct access into the innermost sancta.

Despite their overall morphological similarities, Structure 16 and Structure 29 are not smaller replicas of the prominent Structure 20 temple. The artifactual remains associated with the Structure 20 temple show evidence of a greater range of divinatory and sacrificial bloodletting rituals, performed by officiating priests in the central hall, than do the ritual paraphernalia recovered from the smaller temples. The two companion temples exhibited no evidence of reworking obsidian, spinning cotton fiber, and preparing food, as was observed in the Structure 20 temple. The differences in size, architectural elaboration, and range of activities associated with the three temples suggest to us that the El Palenque temple precinct displayed a temple hierarchy. We lack clear evidence bearing on the temples’ dedication to specific supernatural forces or figures (e.g., refs. 18 and 19), other than the fragments of modeled grayware bottles depicting the Zapotec supernatural Cociyo (Lightning) recovered from Structure 20 and Structure 16 (11, 14).

The three temples were probably abandoned at the same time, in the early years of the Monte Albán II phase (100 B.C.−A.D. 200). Their destruction by fire is manifested by their burned surfaces, burned adobe wall fall, and carbonized deposits. Charcoal (β-143353) from one of the carbonized deposits and burned adobe wall fall (F. 22) resting on Structure 16’s exterior aprons yielded a conventional radiocarbon date of 1,980 ± 70 B.P. or 30 ± 70 B.C. and a 2-σ calibrated range of 165 B.C.−A.D. 155 (Table 1). The slightly earlier ¹⁴C dates associated with the final uses of the two hearths (F. 65, F. 82) directly behind the Structure 20 temple (β-171545 and β-299210) (Table 1) are also consistent with that abandonment date.

At the eastern end of the temple precinct, in a direct line behind the Structure 20 temple, stood a pair of structures on low mounds A and C, situated at a right angle to each other, and aligned to N 17° E (Figs. 2 and 3). When we mapped and surface-collected them, we suspected they might be residences, yet the overall size and configuration of their wall foundations and internal surfaces differed from the surface characteristics of the many house mounds located across the site of El Palenque (cf. 20). Our excavation Area A (Mound A) exposed Structure 27 (Fig. 7), a two-room structure that measured 14.6 by 12.7 m and faced west. A stairway of three tiers led across a flagstone-paved platform at the building’s western front and up to a narrow rectangular vestibule defined by adobe walls atop stone foundations and a hard-packed earthen floor (Rm. 1). From Room 1, a step onto a paved zigzag passageway led through a massive wall (2.2- to 2.4-m wide) and down a 34-cm step into the structure’s sunken back room (Rm. 2). Room 2’s hard-packed earthen floor measured 8 × 5 m, and was delimited by adobe walls on its northern, eastern, and southern sides of the same thickness as its western wall. A posthole on its floor (F. 76) and a fragment of multiple cane-impressed burned daub recovered from its southeastern corner provided evidence of posts and possible partitions within Room 2. At its center lay a rectangular, burned adobe-lined firebox (F. 79), 1 × 0.74 m in exterior dimensions, and a maximum depth of 0.13 m below the floor. Alongside the firebox lay a subfloor adobe-lined box (F. 78) (0.84 × 0.55-m wide and 0.35-m deep), which may have served as a subfloor cache. It had been looted in antiquity, when its stone-slab lid was displaced. Charcoal from beneath F. 78 (β-288315) yielded a conventional radiocarbon date of 2,140 ± 40 B.P. or 190 ± 40 B.C., and a pair of 2-σ calibrated ranges of 360−290 B.C. and 240−50 B.C. (Table 1). Charcoal (β-288311) recovered from the subfloor deposit beneath Room 2 dated to 2,090 ± 40 B.P. or 140 ± 40 B.C., with a 2-σ calibrated range of 200−10 B.C. (Table 1).

Fig. 7.

Plan of Structure 27 priests’ residence.

In Area C we excavated Structure 28 (Fig. 8), which measured 10 × 10 m and faced south. The narrow rectangular vestibule (Rm. 1) at its southern entrance had a hard-packed earthen floor that measured 7.7 × 0.80 to 1-m wide. Room 1’s western wall was indented some 0.60 m from the rest of Structure 28’s western edge. The rear room (Rm. 2) was reached from Room 1 by a step onto a paved zigzag passageway through a thickset wall (1.6- to 1.7-m wide) and then a 30-cm step down into the lower Room 2. The hard-packed earthen floor of Room 2 measured 7 × 5.6 m. Abutting Room 2’s western wall was a rectangular, adobe-lined firebox (F. 74), reddened from use, measuring 0.92 × 0.65 m and 0.35-m deep. Some fragments of Late Monte Albán I phase grayware ceramics (17) lay on the floor. Charcoal (β-288309) from the floor dated to 2,070 ± 40 B.P. or 120 ± 40 B.C., and had a 2-σ calibrated range of 190 B.C.−A.D. 10 (Table 1); it probably pertains to Structure 28’s occupation during the Late Monte Albán I phase (300−100 B.C.). Charcoal recovered from beneath the floor of Room 2 (β-288312) yielded a date of 2,160 ± 40 B.P. or 210 ± 40 B.C., and a 2-σ calibrated range of 360−90 B.C. (Table 1). The ¹⁴C dates indicate that Structure 28 and Structure 27 were constructed and occupied contemporaneously.

Fig. 8.

Plan of Structure 28 priests’ residence.

We propose that Structure 27 and Structure 28 were priests’ residences, an interpretation that is based on several lines of evidence, beginning with their placement behind the prominent Structure 20 temple. The ground plans of Structure 27 and Structure 28 display strong similarities to one another in: (i) layout, (ii) construction with thick walls, (iii) indirect access through narrow vestibules and blind passageways, and (iv) habitation rooms with sunken floors equipped with fireboxes, probably for light and warmth. Although their thick walls and dichotomy between outer and inner spaces are reminiscent of temples, the resemblance to the multiroom temples reported here is limited to those two architectural elements. The presence of internal fireboxes is a feature that Structure 27 and Structure 28 share with the unambiguous residential structures at El Palenque, including the elaborate Area I palace as well as the ordinary residences excavated in Area P (St. 9) and Area X (St. 30) (Fig. 2). At the same time, the ground plans of Structure 27 and Structure 28 clearly distinguish them from the other residences.

The artifact assemblages recovered from Structure 27 and Structure 28 further differentiate them from both ordinary houses and temples. Fig. 9 is a scatter plot of the relative frequency (percentage of total diagnostic potsherds) of plainware jars (ollas) plus griddles (comales) against fine grayware serving plates. The samples used in this plot were recovered through the complete horizontal excavation of ordinary houses (St. 9 and St. 30), multiroom temples (St. 16 and St. 29), and priests’ residences (St. 27 and St. 28). The Area I palace and Structure 20, in view of their uniqueness, are not included in this analysis. The scatter plot distinguishes among the assemblages associated with the ordinary houses, temples, and priests’ residences (Fig. 9). For example, there is relatively little evidence in the temples and priests’ residences of the daily domestic activities performed in ordinary houses, which centered on the preparation of food with utilitarian ceramics (the jars and griddles). Also, we note that the temples (St. 16 and St. 29) and the priests’ residences (St. 27 and St. 28) lacked the grinding stones (manos, metates) used for processing maize. The relatively high frequencies of fine gray serving plates recovered in Structure 27 and Structure 28 imply that food was prepared elsewhere and then consumed by the priests in their living quarters. In contrast to the temples, ritual paraphernalia were not recovered in the priests’ residences, although one possible exception is a small polished stone disk, incised on one side (2 cm in diameter), found on the floor of Room 2 in Structure 27, which may have been used for divination or games.

Fig. 9.

Scatter plot of plainware jars (ollas) plus griddles (comales) against grayware serving plates from excavated temples, houses, and priests’ residences at El Palenque.

Despite their similarities, Structure 27 and Structure 28 were not mirror images of one another. Structure 27 was ∼85% larger than Structure 28. In addition, the flagstone-paved platform at the front of Structure 27 was not duplicated in Structure 28. This architectural elaboration of Structure 27 is reminiscent of the flagstone-paved platform (St. 8) in front of the palatial residence (St. 7) of El Palenque’s rulers in Area I (13). Both Structure 27 and Structure 28 had rectangular fireboxes, but only Structure 27 included a subfloor cache, whose stone-slab lid had evidently been pried open in antiquity and the contents removed. These differences in size and architectural elaboration may reflect a status difference between the priestly occupants of Structure 27 and those of Structure 28. In addition, we suspect that the cell-like rooms behind the southeast corner of the Structure 20 temple could have served as sleeping quarters, probably more suitable for young priests or initiates (Fig. 4).

The abandonment of Structure 27 and Structure 28 was accompanied by a conflagration, which resulted in the burned surfaces and burned adobe wall fall recorded during their excavation. An adult individual of indeterminate sex (Burial 6) was rather hurriedly interred in Structure 27’s firebox, in a cramped position that resulted in the skull resting at the firebox’s southeast corner and a knee poking above its northwest corner. The individual’s lower legs overlay an incised gray bowl of the Late Monte Albán I phase ceramic microtype G.12b (17).

Evidence of a special ritual feature (F. 77) turned up in the temple precinct, behind the Structure 20 temple. A masonry-lined rectangular entrance to a vaulted tunnel lay 4-m east of the temple’s rear wall, at the building’s midpoint (Fig. 4). A series of 16 steps descended a steep corbel-vaulted stairway to the base of Structure 20’s rear masonry foundation, where a stone lintel capped the entry into a vaulted, masonry-lined tunnel that passed under Structure 20 and proceeded at an orientation of W 17° N toward the plaza. The narrow passageway under the temple was 1.07-m tall and 0.45-m wide. The tunnel continued beyond the western base of the Structure 20 temple for nearly 5 m, where it exited onto the plaza floor, making a total length from entrance to exit of 27 m. Charcoal (β-332860) recovered in the adobe construction above the tunnel at the western base of the building dated to 2,080 ± 30 B.P. or 130 ± 30 B.C., with a pair of 2-σ calibrated ranges of 180−40 B.C. and 10−0 B.C. (Table 1). Based on ethnohistoric analogy (6, 7), we propose that priests and costumed impersonators used this facility to secretly access the public plaza on special ceremonial occasions. Issuing as it does from the largest temple onto the public plaza, this elaborate stone-lined tunnel hints at the importance of such ritual performances practiced by the priests of the El Palenque temple precinct (12, 21).

Conclusions

Several lines of archaeological evidence support the proposed existence of a specialized temple precinct at El Palenque in the Late Monte Albán phase (300−100 B.C). The temples and attached facilities making up the precinct were raised above the public plaza and demarcated by an enclosure wall, which effectively limited access from the plaza. The differences in the sizes, architectural elaboration, and attached facilities of the three member temples probably reflect a temple hierarchy. The range of activities performed by the temple staff, as indicated by the associated features, activity areas, and artifacts, is also consistent with such a hierarchy. The priests and servants who staffed the principal Structure 20 temple were performing a more diverse array of ritual activities and provisioning tasks than were the priests assigned to the two smaller companion temples. These tasks included tending braziers and ritual fireboxes, performing ritual sacrifices, and provisioning the temple and its officiating priests with obsidian blades and other ritual paraphernalia, food, and possibly cloth.

Two special residences directly behind the Structure 20 temple differed in nature and content from ordinary residences at El Palenque. A hierarchy is also evident in the sizes and architectural elaboration of these priestly residences, which probably corresponds to the accommodations of high priests versus ordinary priests. Young priests in training might have slept in cells attached to the rear of the Structure 20 temple.

The radiocarbon dates associated with the temples, priests’ residences and special ritual features and activity areas span the Late Monte Albán I phase (300−100 B.C.), for which there is considerable evidence of state organization (5). The contemporaneity of the temple precinct’s differentiated components is consistent with the existence of a full-time, specialized priesthood in the service of a state religion by 300–100 B.C. As of this writing, the temple precinct at El Palenque is the earliest excavated temple precinct in the Valley of Oaxaca.