Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Artificial plateau construction during the Preclassic period at the Maya site of Ceibal, Guatemala

  • Takeshi Inomata ,

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Visualization, Writing – original draft, Writing – review & editing

    Affiliation School of Anthropology, University of Arizona, Tucson, Arizona, United States of America

  • Daniela Triadan,

    Roles Funding acquisition, Investigation, Supervision

    Affiliation School of Anthropology, University of Arizona, Tucson, Arizona, United States of America

  • Flory Pinzón,

    Roles Investigation

    Affiliation Facultad de Ciencias Sociales, Universidad del Valle de Guatemala, Guatemala City, Guatemala

  • Kazuo Aoyama

    Roles Conceptualization, Funding acquisition

    Affiliation Faculty of Humanities, Ibaraki University, Mito, Ibaraki, Japan


Investigations at the Maya site of Ceibal, Guatemala, documented an artificial plateau, measuring 600 x 340 m in horizontal dimensions and 6 to 15 m in height. Unlike highly visible pyramids, such horizontally extensive constructions covered by the rainforest are difficult to recognize on the ground, but airborne laser scanning (LiDAR) revealed its planned form. Excavations carried out over many years provided data on its construction sequence, fill volumes, and labor investments. The initial construction of the plateau occurred around 950 B.C. when a formal ceremonial complex was built in its center. This was the period when the inhabitants of the Maya lowlands were adopting a new way of life with greater reliance on maize agriculture, full sedentism, and ceramic use. The inhabitants of areas surrounding Ceibal, who retained certain levels of residential mobility, probably participated in the construction of the plateau. In this regard, the Ceibal plateau is comparable to monumental constructions that emerged before or during the transition to agriculture or sedentism in other parts of the world. The data from Ceibal compel researchers to examine the social implications of monumental constructions in the Maya lowlands before the establishment of centralized polities with hereditary rulers. Unlike pyramids, where access to the summits may have been limited to privileged individuals, the horizontal monumentality of the plateau was probably more conducive to inclusive interaction. The Ceibal plateau continued to be built up during the Preclassic period (1000 B.C.-A.D. 175), and its fill volume substantially surpassed those of pyramids. Large-scale construction projects likely promoted organizational and managerial innovations among participants, which may have set the stage for later administrative centralization.


Pyramidal buildings are commonly viewed as the hallmarks of Maya and other Mesoamerican civilizations. At the height of population and political centralization during the Classic period (A.D. 250–950), these buildings were foci of collective work at many Maya centers and represented the power and authority of rulers and other elites. Importantly, some of the largest known pyramids in the Maya lowlands were constructed before the Classic period. For example, the Danta pyramid complex of El Mirador, built during the Late and Terminal Preclassic period (350 B.C.-A.D. 250), measures 72 m in height, dwarfing any pyramids of later periods [1,2]. It is likely that the construction of such buildings played a critical role in the development of centralized polities in the region. Recent investigations, however, hint at the presence of monumental constructions that emphasized horizontal dimensions, as opposed to the vertical ones of pyramids. Such artificial plateaus may have been even more important than pyramids during the Middle Preclassic period (1000–350 B.C.), but they have received little attention from researchers. A notable exception is the site of Cival, where Estrada-Belli uncovered the large-scale leveling of a hilltop area of 0.5 km2 dating toward the end of the Middle Preclassic [3,4].

Our airborne LiDAR (Light Detection and Ranging) survey and excavations at the site of Ceibal, Guatemala, also documented a large artificial plateau (Fig 1). Its initial version dates back to the beginning of the Middle Preclassic. Estimates of its fill volumes and construction costs indicate that the Ceibal plateau was a massive construction even at its early stages, achieved through significant labor investments by a large part of the population. Like the case of Cival, its volume significantly exceeds those of pyramids. The Ceibal plateau was among the earliest-documented monumental constructions in the Maya lowlands.

Fig 1. Map of the Maya area and adjacent regions with the locations of the sites discussed in the text.

Ceibal and its plateau

General setting

Ceibal is the largest Maya center in the southwestern part of the Maya lowlands, located on top of an escarpment overlooking the Pasión River (Fig 1). Willey directed the Harvard University Project (HP) at this site from 1964 through 1968 [57]. We revisited this important site in 2005 as the Ceibal-Petexbatun Archaeological Project (CPAP) and continued our field research until 2017. Permits for the CPAP research were granted by the Instituto de Antropología e Historia de Guatemala (IDAEH). Detailed stratigraphic and ceramic data, along with 182 radiocarbon dates obtained so far, allowed us to refine the chronology of Ceibal’s occupation, originally developed by Sabloff during HP (Fig 2)[6].

The ceramics of the Junco 1 phase generally align with those of the Terminal Preclassic in other parts of the Maya lowlands, and Junco 2–4 materials with those of the Early Classic. Nonetheless, the transition from Junco 1 to Junco 2 is not clear in some contexts. Thus, in the following analysis, we use the Xate phase as the end of the Preclassic period (A.D. 175), and the Junco 1 phase is included in the Early Classic period. For the discussion of the Maya lowlands in general, however, we use the more conventional date of A.D. 250 for the end of the Preclassic. The absolute dates are based on radiocarbon dates, which we analyzed with Bayesian statistics. All dates are calibrated. These radiocarbon dates and the Bayesian models are discussed in detail in our previous publications [810]. Since our last publication, we have obtained additional radiocarbon assays, resulting in a total of 182 dates. These recently obtained dates confirm our chronology and do not change the absolute dates of the ceramic phases published in our previous publications [11].

The residents of Ceibal began to use ceramics and to build durable structures at the beginning of the early Middle Preclassic Real phase around 1000 B.C. Ceibal continued to grow during the late Middle Preclassic Escoba phase (700–350 B.C.) and the Late Preclassic Cantutse phase (350–75 B.C.). After some domographic fluctuations during the Terminal Preclassic Xate phase (75 B.C.-A.D. 175), it experienced a significant demographic decline at the end of the Junco 1 phase around A.D. 300. The Early Classic Junco 2–4 phases (A.D. 300–600) are characterized by low populations and limited construction activity. Its population grew again during the Late Classic Tepejilote phase (A.D. 600–810). After a political disruption at the end of the Tepejilote phase, Ceibal regained political power during the Terminal Classic Bayal phase (A.D. 810–950) amid the so-called Classic Maya collapse. It was, however, deserted at the end of this period. During the Postclassic Samat phase (A.D. 1000–1200), small groups visited this place.

In 2015, we acquired airborne LiDAR data over an area of 470 km2 around Ceibal [12,13]. LiDAR involves laser scanning of terrain from equipment placed on a small airplane. Some laser pulses penetrate through the canopy and record the 3-dimensional morphology of the ground surface. In the tropical lowlands where thick vegetation hinders pedestrian surveys of wide areas, LiDAR has revolutionary, providing detailed data on subtle archaeological features over extensive areas [1419]. The crew of the National Center for Airborne Laser Mapping (NCALM) of the University of Houston, under the direction of Ramesh Shrestha and the coordination of Juan Carlos Fernandez Diaz, obtained LiDAR data over the area of Ceibal from March 18th to 23rd, 2015. They collected data for most parts of the study area from a flying altitude of 700 m above the ground level (AGL) and at a total pulse repetition frequency (PRF) of 450 kHz (150 kHz per channel for the three channels of Titan LiDAR), but they also used a total PRF of 750 kHz for some flight lines. For the central part of Ceibal, the team conducted canopy penetration tests with multiple settings, including 700 m AGL and 300 kHz total PRF, 600 m and 450 kHz, and 400 m and 150 kHz. The canopy penetration test flights resulted in 51 to 72 laser shots per m2, whereas regular mapping flight lines produced 15 to 19 shots per m2. Ground point densities vary widely by vegetation type, but the average for the central part of Ceibal obtained by the test flights, where the plateau is located, was 2.84 points/m2. The NCALM team then produced a digital elevation model (DEM, bare earth model after the removal of vegetation and buildings) at a horizontal resolution of 0.5 m. For a more detailed discussion of the LiDAR acquisition methods and settings at Ceibal, see [12,13,20]. Other applications of LiDAR in tropical areas include [1416,19,2125].

An important benefit of LiDAR is that its bird’s eye views without vegetation allow us to identify large-scale landscape modifications that are difficult to see from the ground level. HP researchers produced an excellent map of Ceibal through a pedestrian survey, recording many structures accurately. This map showed the central part of the site called Group A as a raised terrain in an amorphous shape (Fig 3). HP and CPAP archaeologists knew that parts of Group A were artificially modified, but dense vegetation prevented us from recognizing its overall shape accurately. LiDAR revealed that this area was an artificial plateau in a roughly rectangular shape, measuring 600 m in length (north-south) and 340 m in width (east-west) and rising 6 to 15 m above the surrounding ground surface. We use the term artificial plateau to describe this type of massive constructions to distinguish them from supporting platforms and other smaller buildings. Supporting platforms may measure 200 x 200 m or less and usually support multiple buildings on top of them. Artificial plateaus are larger than supporting platforms in horizontal dimensions, and the Ceibal plateau underlay multiple pyramids, supporting platforms, and plazas.

Fig 3. Group A plateau of Ceibal.

A. Map produced by the HP (1m contour lines). Redrawn from [5]. B. LiDAR-derived DEM. The area of confirmed Real-phase fills is indicated.


Data on the construction process of the Ceibal plateau derived from numerous excavations carried out by the HP and the CPAP (Fig 4). Detailed discussions of the HP excavations are found in [7,26], and those on the CPAP in [2733]. All excavated materials are stored at the National Museum of Archaeology and Ethnology and the Salon 3 storage facility of the Guatemalan government (7a Avenida y 6a Calle, Zona 13, Guatemala City). Researchers interested in those materials should request permission from the Institute de Antropología e Historia de Guatemala (12 Avenida 11–11, Zona 1, Guatemala City; phone +502 2208 6600;, Guatemalafor access to these facilities.

Fig 4. LiDAR-derived DEM of the Ceibal plateau with the locations of excavations and the lines of important sections shown in Figs 5 and 6.

Many of the CPAP excavations penetrated through construction layers of different periods to reach bedrock. Small HP excavations in plazas were also dug to bedrock, but most HP excavations placed over structures were limited to surface layers. These results indicated that the plateau was constructed throughout the history of Ceibal, but its initial construction started at the beginning of the Middle Preclassic, and a substantial part of its fill volume was placed during the Preclassic period.

The primary focus of the CPAP excavations was the southern part of the Central Plaza, where we identified the earliest formal ceremonial complex known in the Maya lowlands. It was built in a standardized configuration called an E-Group assemblage, which consisted of a western mound and an elongated eastern platform flanking a plaza [810,28,34]. The earliest version of the Ceibal E Group dates to 950 B.C., and this architectural format spread throughout the Maya lowlands in later periods. At Ceibal, the first construction episode involved the scraping of surface soil and bedrock to shape the initial E-Group buildings and the plaza out of bedrock. The original stage of the plateau may also have been formed at that time through the carving of bedrock, but it is difficult to determine the date of surface scraping in the peripheral parts of the plateau. The western structure and the plaza were then raised with earthen fills at least twice during the Real 1 phase (1000–850 B.C.). Builders continued to renovate the plazas and structures by adding fills over earlier constructions in subsequent periods. The fill thicknesses of the E-Group plaza reached 1.4 to 1.6 m by the end of the Real phase and 1.8 to 2.8 m by the end of the Preclassic, whereas the Classic portions of the fills were 0.3 to 0.4 m thick (Figs 5 and 6, and Table 1). Our excavation showed that Structure A-12, the northern portion of the eastern building, measured 8.7 m in height by the end of the Preclassic, while Classic constructions added only 0.9 m. The presence of numerous caches with greenstone axes suggests that the E-Group assemblage was the primary focus of communal ritual at Ceibal during the Middle Preclassic [28,30,33].

Fig 5. East-west lines of important sections.

For the locations of these section lines, see Fig 4.

Fig 6. North-south lines of important sections.

For the locations of these section lines, see Fig 4.

In the South Plaza, test pits excavated by the HP revealed even deeper Real and Preclassic fills [7]. The magnitude of Preclassic construction was also suggested by the excavations of supporting platforms. The A-24 Platform and the East Court reached heights of 3.4–6.0 m and 2.1 m respectively by the end of the Real phase, and 6.0–7.5 m and 4.7–5.2 m by the end of the Preclassic, whereas Classic-period fills were minimal. HP excavations revealed Real layers under Structure A-2. Other platforms, including Structure A-4, and A-18, appear to have been made mostly after the late Middle Preclassic Escoba phase, but substantial parts of their masses date to the Preclassic (Figs 5 and 6). These supporting platforms, arranged along the north-south axis of the E Group, formed a standardized site plan called the Middle Formative Chiapas (MFC) pattern during the Middle Preclassic period [8,29]. Contemporaneous MFC patterns are found at the Olmec center of La Venta and Chiapas sites along the Grijalva River and on the Pacific Coast [3538].

Substantial volumes of the pyramids in Group A were also built during the Preclassic period. Although most pyramids, except Structure A-20, were not excavated down to their earliest cores, shallow excavations were sufficient to uncover Preclassic constructions in Structures A-1, A-5, and A-6. The only exception to this pattern is Structure A-3, a small pyramid in the South Plaza, which was built entirely during the Terminal Classic [7,39]. Excavation data on peripheral parts of the plateau are more limited but still indicative of Preclassic constructions. Excavations in the northern part of the Central Plaza (Op. 202A) showed evidence of surface soil scraping, floors dating to the Real-Escoba transition, and substantial Preclassic fills, covered by thin Classic layers. To the west of the E Group (Op. 204A), we also found a Real layer and substantial Preclassic fills (Figs 46 and Table 1). The small North Plaza, however, appears to have been added during the Escoba phase, and we did not find Real layers in excavations placed just outside of the plateau (Ops. 206A, 201D, and 217A).

In the South Terrace of the plateau, fill thicknesses are modest, and Real fills are generally absent. It appears that the South Terrace was shaped mainly through the carving of surface soils and bedrock, particularly in its early construction stages. The LiDAR data show possible piles of back dirt along the western and southern edges of the South Terrace (Fig 7). These piles become particularly clear in the Red Relief Image Map (RRIM) visualization of the LiDAR data, which highlights subtle surface morphologies [12,40]. Builders probably dug into gentle slopes of bedrock to define the edges of the plateau with steep slopes. They seem to have dumped back dirt 15 to 30 m away from the plateau. The excavation of Operation 228B in the southern edge revealed carved bedrock under a thin layer of eroded soil (Figs 46). We could not determine when these edges were carved. Nonetheless, the excavation of Structure 64, a long mound along the southern edge (Op. 228A), indicates that this mound was initially constructed during the Escoba 3/Cantutse 1 phase or earlier (Figs 46). It is likely that the southern edge was carved and defined at least by the end of the Middle Preclassic.

Fig 7. LiDAR-derived DEM in the RRIM visualization showing the southwestern part of the plateau and a section showing possible back dirt and an edge of the plateau.

The vertical dimension of the section is exaggerated.

In the northwestern part of the plateau, a more regularly-shaped berm is visible roughly 30 m away from the plateau’s edge (Fig 8). We originally suspected that it was for directing surface runoff water to a depression located north of the plateau. The hydrological analysis of the DEM, however, shows that the channels along the berm led water to the north and south. Water-channeling does not seem to have been the primary function of the berm. One possible interpretation is that the berm was back dirt from the carving of bedrock in defining the plateau’s edge. Another possibility is that the Ceibal residents were preparing to expand the plateau in this area, and the berm defined the planned edge of the extended plateau. In this scenario, the builders never completed this expansion.

Fig 8. LiDAR-derived DEM in the RRIM visualization showing the northwestern part of the plateau and a section showing a berm and an edge of the plateau.

The vertical dimension of the section is exaggerated. Blue lines indicate the probable flows of water reconstructed with ArcGIS’s hydrological analysis functions.

These data show that the initial form of the plateau was built during the early Middle Preclassic Real phase (1000–700 B.C.). The early fills in the E-Group assemblage were placed around 950 B.C., and the carving of bedrock may have taken place slightly earlier. Although we cannot define the exact shape of the Real-phase plateau, it comprised a roughly 400 x 200 m area where Real fills have been documented, consisting of the Central and South Plazas, the East Court, the A-24 Platform, and Structure A-2 (Fig 3). If areas beyond this limit, including the South Terrace, were shaped during the Real phase through the carving of bedrock, the initial stage of the plateau would have been substantially larger. By the end of the Preclassic period around A.D. 175, a substantial part of the plateau was built, and it probably resembled the current form shown in the LiDAR-based DEM.

Estimates of fill volumes and labor investments

To evaluate the social importance of plateau construction, we estimated its fill volumes and the labor investments required for this construction. Before the present study, Inomata estimated the earth-fill volume of the Central and South Plazas in 2014 [41]. This work was done before the LiDAR survey and the detection of the plateau and thus did not include the volumes of other parts of the plateau. The present study expands Inomata’s previous work through the production of detailed 3D digital models by incorporating additional excavation data and the LiDAR data (Fig 4, S1 Text).

Fill volume estimates.

We first calculated the total fill volume of the plateau, by subtracting the 3D digital model of bedrock estimated through excavations from the ground surface model provided by the LiDAR data. Thus, the estimate of the plateau fill volume included the fills of pyramids, supporting platforms, and other buildings located on top of it. We then separated the Preclassic and Classic portions of the fills. The Preclassic fills were further subdivided into the Real, Escoba, and Cantutse-Xate phases. Table 2 summarizes the estimated volumes of the plateau and all pyramids by period. We calculated possible lower and higher estimates for the entire fill volume and those of the Preclassic and Classic periods, but Table 2 shows what we think the most plausible estimates.

Table 2. Most plausible plateau volume estimates by period.

In reconstructing the elevations of bedrock, we had reasonably good data for the E-Group plaza, the East Court, and the central part of the South Plaza. Data were sparser in areas outside of these complexes. To show a range of possible error in volume calculations, we first created three versions of bedrock reconstruction: 1) the minimum volume given by the upper estimated elevation of bedrock; 2) the estimate that we think the most plausible; and 3) the maximum volume given by the lower estimated elevation of bedrock (Figs 9 and 10). For the southwestern part of the plateau where excavation data are lacking, we created conservative bedrock models by placing them close to the current ground surface. Thus, it is unlikely that our estimates of the plateau volume are unreasonably inflated. We should also note that the minimum volume estimate assumes that bedrock levels were high in the peripheral parts of the plateau. This scenario would imply that builders carved and removed a substantial quantity of bedrock to shape the plateau edges. The quantities of such removed material are not included in our fill volume estimates. We then calculated the total fill volume for all pyramids located on the plateau. These data show that even in our smallest estimate, the construction volume of the plateau is substantial. The total plateau fill is 7.8 to 11.8 times larger than the combined volume of the pyramids (Fig 11, Tables A and B in S1 Text).

Fig 9. Sections showing the ground surface, excavations, and three versions of bedrock: The upper estimate (minimum plateau volume); the most plausible estimate; and the lower estimate (maximum plateau volume).

Fig 10. DEMs showing the upper, most plausible, and lower estimates of the Preclassic (the end of the Xate phase) and bedrock surfaces.

Fig 11. Estimates of plateau fill volumes and yearly average volumes.

The volumes of pyramids are included in the plateau volumes. The volumes by period and those by Preclassic phase are based on the most plausible total volume and the most plausible Preclassic volume respectively. The yearly average volumes are calculated from the most plausible volume estimates.

To calculate the Preclassic and Classic portions of the plateau fill volume, we created three versions of the Preclassic fill model based on the most plausible model of bedrock (Fig 10). In the 2014 study, Inomata assumed that construction activity during the Junco phase was negligible, but subsequent investigations identified some buildings dating to this period. In the present study, we chose to calculate the combined volume for the entire Classic period, including the Junco, Tepejilote, and Bayal phases. These estimates suggest that 59.7% to 79.3% of the total plateau volume was constructed during the Preclassic period. We should caution that if we make such estimates for the minimum and maximum bedrock models, the error range for the Preclassic volume will be larger. Nonetheless, if we add the quantity of soil and bedrock that were removed, the total amount of moved materials during the Preclassic period would have been even larger. Thus, these estimates provide clear evidence that significant construction activity took place during the Preclassic period.

We then subdivided the second (most plausible) Preclassic model into shorter ceramic phases (Fig 12, S1 Text). We combined the Cantutse and Xate phases for volume calculations because the differentiation of ceramics from these phases was sometimes difficult. These estimates suggest that a considerable portion of the plateau was constructed during the Real and Escoba phases of the Middle Preclassic period. By 800 B.C., the A-24 Platform alone represented a monumental scale, reaching a height of 6 m and an east-west dimension of more than 34 m [8]. The mass of the entire plateau during the Real phase was substantially larger, possibly surpassing the combined volume of pyramids of all periods. Again, we need to consider ranges of error involved in these calculations, but a significant portion of the Real and Escoba fills in our models is concentrated in the central part of the plateau where we have abundant excavation data. It is unlikely that the true Middle Preclassic volume was much lower than our estimates.

Fig 12. Three-dimensional digital reconstructions of the Ceibal plateau through time.

Labor investment estimates.

Based on the estimates of fill volumes, we then calculated possible labor investments required for the plateau construction. We first calculated total labor investments in person-days (Table 3). Through a comparison with estimated populations at Ceibal (Table 4), we then assessed how long these constructions took (Table 5). It should be emphasized that the ranges of error for these estimates are even larger than those for the construction volumes. Although we provide low and high estimates for each category of calculation, the true figures may be substantially smaller or larger than our estimates. These estimates should be understood as heuristic exercises, which are intended to help us explore the social implications of construction activity.

Table 3. Estimated labor investments in the plateau construction.

Table 5. Estimates of days required for the plateau construction.

Table 3 shows the low and high estimates for labor investments by listing costs involved in the procurement of earth used in fills, their transport, and fill constructions. The formulae for these calculations derived from earlier experimental studies by various scholars [4246] (see S1 Text for a detailed explanation). The estimates were made based on the most plausible construction volumes for individual periods listed in Table 2. We first made these calculations, assuming that the entire plateau fill consisted of earth. Subsequently, we applied adjustments to the high estimates to account for the higher construction costs of masonry, which made up small portions of the plateau and the buildings standing on it. Our labor investment estimates were based only on the fill volume figures, and the scraping of surface soil and bedrock was not included. If we consider the costs of scraping, the total labor investments should be larger, particularly for the Real phase.

Population estimates were made previously by Tourtellot [26,47] and Inomata et al. [13] (Table 4). Tourtellot’s estimates were based on a ground survey and excavation results, whereas Inomata el al. mainly used LiDAR data. There are substantial debates about the methods of population estimates [48], and different assumptions can lead to divergent results even for the Classic period, for which we have relatively reliable settlement data. Inomata et al. calculated low and high estimates for this period, and here we use their high estimate to explore the shortest possible construction time. Estimates for the Preclassic period are even more problematic, and there is a substantial difference between the figures presented by Tourtellot and by Inomata et al. Tourtellot also suggested the possibility of a lower estimate through a time-weighted adjustment, which resulted in a Cantutse-Xate population figure similar to Inomata et al.’s [26]. Inomata et al. did not attempt to estimate populations for the Real and Escoba phases. As discussed below, Tourtellot likely underestimated these Middle Preclassic populations.

Through the analysis of LiDAR data, Inomata et al. defined zones of settlement distributions around Ceibal [13]. The structure density is high in the 5.41 km2 area around Group A demarcated by escarpments and poorly-drained terrains. Inomata et al. called this area the Ceibal center and assumed it to be the core of the Ceibal community. The terrain of 133.10 km2 surrounded by the Pasión River and wetlands is called the Ceibal horst. The residents of this area outside of the Ceibal center may have been loosely affiliated with Ceibal or may have been under its direct political control during different periods. Tourtellot estimated populations for his sampled area of 15.25 km2, which we converted to the figures for the Ceibal center and the Ceibal horst, by applying population ratios assessed from LiDAR data. For the Preclassic period, when the population was markedly concentrated in the Ceibal center, we estimated that the population ratio of the Ceibal center to Tourtellot’s sample area was 0.64 and the ratio of the Ceibal horst 2.26. For the Classic period, we applied figures of 0.53 and 3.28 respectively.

By dividing the total labor investments in person-days by population estimates, we obtained estimated numbers of days required for the construction of the Ceibal plateau (Table 5). We calculated two sets of estimates, one with the assumption that only the populations of the Ceibal center participated in construction and the other with the assumption that the entire populations of the Ceibal horst joined. For each set, we calculated low and high estimates. The low estimates are based on the low person-day estimates and the higher population estimates between Tourtellot’s and Inomata et al.’s, as well as the assumption that two persons from an average household of five contributed to construction. The high estimates were obtained with the high person-day estimates, the lower population estimates, and the participation rate of one person out of five. For the Classic period, we assumed that all construction volume of this period was placed during the Tepejilote-Bayal phase.

We also calculated possible construction times in individual major construction episodes. Although small-scale building activity probably took place each year or every few years, larger-scale construction events may have taken place in longer cycles. Our excavations in the central part of the plateau revealed seven floors dating to the Real phase, five dating to the Escoba phase, and five dating to the Cantutse-Xate phase, hinting at such long cycles. We calculated the days required for each of the major construction episodes, assuming that such events took place every 20 years, which corresponds to the primary ceremonial cycle of the Classic period.

Although these estimates contain large ranges of error, the results suggest that Tourtellot’s estimates for the Real populations are probably too low. It is unlikely that builders engaged in construction activity more than three or four months a year. We should reiterate that the true labor investment for this period may have been substantially larger than our estimates because our calculations did not include the labor for scraping of surface soil and bedrock. The population during the Real phase was probably substantially larger, and the Escoba population estimates may also need to be adjusted upward.

The wide discrepancies in the population estimates for the Cantutse-Xate phase make the evaluation of this period difficult. However, it is reasonable to think that the construction of ceremonial buildings during this period was a significant social endeavor. During this period, many temple pyramids outside Group A were built, adding to labor demands. In comparison, the expansion of the plateau during the Classic period required a much smaller portion of social resources.

Political centralization

We have discussed political processes at Ceibal in our previous publications [9,13,28,29,34,49,50], and here we summarize main points. Some archaeologists use the scales of monumental buildings and population sizes as indicators of political centralization and social hierarchy [5153]. The purpose of this paper, however, is to examine the possible emergence of monumental constructions before the development of powerful elites, and thus we need to evaluate the degrees of political centralization and social inequality through other sets of data. We focus on large and elaborate residential complexes that may have been used by elites, rich burials, and material symbols of political authority.

In this brief review, we emphasize the following three points. First, during the early Middle Preclassic Real phase, when the initial version of the Ceibal plateau was built, emergent elites may have existed, but their power was probably limited. Second, during the late Middle Preclassic Escoba phase, Ceibal probably did not have rulers comparable to those of centers on the Gulf Coast and in the Grijalva River region. Third, during the Late-Terminal Preclassic Cantutse-Xate phase, rich tombs and symbols of authority used by later Maya rulers began to emerge in the central lowlands, but recognizable signs of marked inequality and political centralization are absent at Ceibal. These observations indicate that a substantial part of the Ceibal plateau was built before the development of highly centralized power.

Before 1000 B.C., the inhabitants of the Maya lowlands led mobile lifeways without the use of ceramics, combining heavy reliance on wild resources with the cultivation of maize and other crops [54]. Archaeologically-recognizable features from this period are scarce, not to mention any material remains indicating marked social inequality. Around 1000 B.C., the occupants of the Maya lowlands began to adopt a more sedentary way of life, along with greater reliance on maize agriculture and ceramic use. Most scholars agree that during the subsequent Middle Preclassic period, initial processes of social differentiation were underway, but the degree of hereditary inequality was still small [5557]. Ceibal may have been a community with more internal inequality than other Maya settlements, but the power of community leaders was most likely constrained. The presence of numerous caches with greenstone axes at Ceibal implies that some individuals played a leading role in obtaining these precious objects and organizing public ceremonies. Burial CB136, placed behind the eastern building (Structure Sib’) of the E-Group assemblage during the Real 3 phase, may have been an interment of such an individual, but it was not a particularly rich burial. It contained five ceramic vessels of common types, but no ornaments or exotic goods. A vexing problem is the nature of the A-24 Platform. It is not clear whether this large platform, built during the Real 1 phase and supporting multiple structures, was a residential complex of elite households or whether it was a communal compound used for collective events, such as meetings and feastings. Platform K’at, constructed during the Real 3 phase and later covered by the East Court, was the first case of a more clearly recognizable residential complex occupied by emergent elites [8,29].

During the Real phase, some groups may have adopted a sedentary lifestyle, but a significant portion of the Ceibal population appears to have lived in pole-in-ground structures without basal platforms, possibly changing their residences seasonally or every few years. Various anthropological studies suggest that it is difficult to impose coercive power on mobile or semi-mobile groups, who can vote with their feet [5861]. It is unlikely that the emergent elites of Ceibal had strong power over its population. During the subsequent Escoba phase, a larger number of more durable residences with basal platforms began to appear in the immediate vicinity of Group A. The practice of rebuilding residences in the same locations over generations, which characterized Classic-period Maya society, was probably gaining popularity during this period. Burial CB104, dating toward the end of the Escoba phase, contained obsidian blades and a core, a possible shell ink pot, and shell ornaments. The buried individual was probably of high status. Nonetheless, we have not found elaborate tombs, carved monuments (coarsely-shaped uncarved monuments are present), or other signs of highly centralized power at Middle Preclassic Ceibal.

The absence of such features at Ceibal contrasts with the Gulf Coast center of La Venta and the Grijalva center of Chiapa de Corzo. At La Venta, elites expressed their power through elaborate stone sculptures and possible tombs [62,63]. At Chiapa de Corzo, elaborate Middle Preclassic tombs were found in the western structure of the E Group, as well as in other buildings [37,64]. Ceibal and Chiapa de Corzo shared remarkable similarities in building arrangements and the placements of caches [30,33]. If powerful individuals that may be called rulers existed at Middle Preclassic Ceibal, we would expect to find their tombs in locations comparable to those at Chiapa de Corzo. Our tunnel excavation through the western E-Group structure, however, did not reveal any interments. The level of political centralization at Ceibal and other lowland Maya communities was probably significantly lower than those on the southern Gulf Coast and in the Chiapas Grijalva region [38].

At the end of the Middle Preclassic period, La Venta collapsed, and some Grijalva centers experienced depopulation or political disruptions [37]. As a pivotal place of communication between the Maya lowlands and Chiapas during the Middle Preclassic period, Ceibal may have suffered impacts of these changes in the surrounding regions more strongly than other lowland settlements. The possible elite residential complexes of the East Court and the A-24 Platform were transformed into open, flat areas at the beginning of the Late Preclassic Cantutse phase (350–75 B.C.). It is not clear where community leaders of Ceibal lived during the Cantutse phase. Although the population of Ceibal grew and residential groups spread to a wider area [12,26], we have not found rich burials from this period.

The main stage of political centralization in the Maya area probably shifted to the central lowlands. The construction of large pyramids at El Mirador and other centers, as well as the San Bartolo murals with glyphs and elaborate images, suggest that early forms of the practices and symbols tied to rulership may have been in development [6567]. The specific nature of political authority during this period, however, is still debated. It is not until the Terminal Preclassic period (75 B.C.-A.D. 250) that we recognize well-established rulers, along with unequivocal royal tombs, at Tikal and other centers [68]. The process of political centralization at Ceibal was probably slower. During the Terminal Preclassic Xate phase, Group D was constructed as a new focus of elite activity at Ceibal, but the scales of its pyramidal constructions were smaller than those at central lowland communities [7,69]. The first historically-known Ceibal ruler, who is retrospectively mentioned in a text, dates to A.D. 415 [70].


The earlier tradition of plateau construction is found outside the Maya lowlands, at the Olmec center of San Lorenzo, which reached its heyday between 1400 and 1150 B.C. [7173]. The initial plateau construction at Ceibal around 950 B.C. probably represents one of the first adoptions of this tradition in the Maya lowlands. Roughly two centuries later, an E Group and a plateau were constructed at Cival. In contrast to the gradual buildup of the Ceibal plateau, a substantial amount of the hilltop leveling at Cival appears to have been accomplished in a short time span. Estrada-Belli estimates that a Middle Preclassic volume of 1,304,000 m3 and a Late Preclassic volume of 556,000 m3 were placed [4]. Early platforms at Komchen, Yucatan, may be roughly contemporaneous with Cival [74]. As the Ceibal plateau continued to grow throughout the latter half of the Middle Preclassic and the Late Preclassic, large platforms were constructed at other Maya settlements, including Nakbe, El Mirador, and Yaxnohcah in the central lowlands, and Yaxuna, Xocnaceh, and Izamal in the northern lowlands [1,7577]. Given the difficulty of recognizing artificial plateaus covered by rainforest, similar constructions may remain undetected at other Maya sites.

Outside of the Maya lowlands, excavations at the Gulf Coast center of La Venta revealed artificial fills measuring 1.5 to 2 m in thickness in the Ceremonial Court of Complex A [58,78]. At the MFC site of San Isidro in the Grijalva River region, Early and Middle Preclassic fills of the E-Group plaza measured roughly 2 m in thickness [79]. Although it is not clear whether La Venta and San Isidro had well-defined plateaus comparable to that of Ceibal, the substantial fill accumulations in plazas or other open areas suggest that their residents shared similar practices of artificially creating extensive ceremonial spaces.

Early plateau constructions in the Maya lowlands occurred during the transition from mobile lifeways and mixed subsistence strategies without the use of ceramics to a more sedentary way of life with greater reliance on maize agriculture and ceramic use. HP researchers assumed that the Real people formed a small community nucleated around Group A and that there was no settlement hierarchy in the region [26,39]. Our study suggests that the Real population estimate based on this assumption is too low. Although our estimate of the plateau construction during the Real phase contains a wide range of error, we can reasonably state that the sacle of its construction necessitated a larger population. It was likely that there were semi-mobile groups in areas outside the Ceibal center, even though their archaeological traces are extremely difficult to detect. These peripheral groups probably participated in construction events and rituals that were periodically held at Ceibal. In this sense, Ceibal was a primary ceremonial center within a broader network of settlements, which attracted a significant number of people from surrounding areas on certain occations.

Emergent elites may have been present at Ceibal during the Real phase and may have played a leading role in organizing this unprecedented construction project. Social inequality at Ceibal, however, was still limited, and it is unlikely that community leaders possessed coercive power over groups that retained some degree of residential mobility. The initial mechanism for the construction of the Ceibal plateau was likely not coercion by elites but willing participation by community members, possibly motivated by the excitement of creating a new building on an unprecedented scale and of taking part in public ceremonies held there.

Plateau constructions may be characterized as horizontal monumentality, which contrasts with the verticality of pyramids. Horizontal building forms may have been preferred by early communities in the Maya lowlands with unpronounced social differentiation. Emphasis on horizontal monumentality is also found in other non-centralized societies in the world, such as prehistoric bison hunters of the North American Plains [80]. These observations are not meant to equate certain built forms with levels of political centralization. San Lorenzo with a large plateau, for example, had established rulers. In opposite examples, some archaic societies without pronounced social inequality, such as those in the American Southeast, built tall mounds reminiscent of pyramids (see below) [81]. With these cautions in mind, we suggest that plateaus with extensive space and open access are more conducive to inclusive interaction than pyramids, which tend to limit access to their summits to the privileged few.

The earliest pyramids in southern Mesoamerica are found outside the Maya lowlands on the Pacific Coast of Chiapas and Guatemala, where the level of political centralization was more pronounced, toward the end of the Early Preclassic and at the beginning of the Middle Preclassic [8284]. The Middle Preclassic residents of Ceibal and Cival also built pyramids, but their volumes were substantially smaller than those of the plateaus. As shown by the example of Ceibal, the tradition of plateau construction continued in the Maya lowlands during later periods. For example, the basal plateau supporting the Danta complex of El Mirador, dating to the Late Preclassic, has dimensions similar to the Ceibal plateau, measuring 600 x 310 m horizontally and 10 to 11 m in height [65]. Nonetheless, pyramids as foci of collective construction activity gradually overshadowed plateaus after the late Middle Preclassic period. In particular, enormous pyramidal buildings at Nakbe, El Mirador, and other sites in the central Maya lowlands may have been correlated with increasing political centralization in that region [1,2,85].

This trend led to the construction of steep pyramids as dominant built forms during the Classic period, which served as symbols of now well-established dynastic rule and as places of elite burials. Pyramids at Ceibal, however, were substantially smaller than the Preclassic portions of the plateau in terms of construction volume. The Preclassic constructions of the Ceibal plateau are impressive even when they are compared to better-known examples of Classic-period pyramids. Webster and Kirker, for example, estimated the construction volume of Tikal Temple 1 and Copan Temple 26 at 18,260 and 31,905 m3 respectively, and their labor investments at 90,000 and 124,000 person-days [45]. Even if the true construction costs of the Tikal and Copan pyramids were significantly higher than the estimates made by Webster and Kirker, we can say that Preclassic constructions of the Ceibal plateau, even that of the Real phase, surpassed Tikal Temple 1 and Copan Temple 26 in terms of construction volume, and perhaps with regard to labor investment as well. Obviously, many other buildings were constructed at Tikal and Copan during the Classic period. Still, when we consider that the populations of Tikal and Copan during the Classic period were substantially larger than Preclassic-period Ceibal, the social importance of the Ceibal plateau during the Preclassic period is evident.

The construction of the Ceibal plateau may be comparable to early monumental constructions in other parts of the world, which emerged before or during the transition to agriculture or sedentism. Examples include: Göbekli Tepe and related Neolithic sites in the Near East; Watson Brake, Poverty Point, and other mound complexes in the Archaic lower Mississippi area; Sechín Bajo, Caral, and other monumental constructions in the Andes; and large shell mounds in the southeastern US and Brazil [81,8691]. Although archaeologists traditionally associated monumental architecture with societies with pronounced political centralization and social hierarchy [51], these examples force us to reconsider this assumption [92,93].

We should note that there are certain differences between the Ceibal plateau and these early monumental constructions from other parts of the world. First, Ceibal dates later than most of these cases. Whereas Göbekli Tepe, Watson Brake, and other examples emerged before or during the period of initial plant and animal domestications in the regions, Ceibal postdates the domestication of maize, beans, squash, and other key crops in Mesoamerica by several millennia. Second, while the earliest monumental buildings in the Near East and American Southeast were constructed probably by groups without marked hereditary inequality, Ceibal during the early Middle Preclassic period possibly had emergent elites, albeit with limited power. In addition, the residents of Ceibal were most likely aware of the presence of more hierarchical groups on the southern Gulf Coast and in Chiapas.

A factor contributing to the late emergence of monumental constructions in the Maya area may be the nature of subsistence strategies related to maize productivity. For a long period after the initial domestication, maize cob size remained small, and maize was a small part of the diet in Mesoamerica [94,95]. Mesoamerican groups continued to rely heavily on wild resources and maintained a high degree of mobility in many regions long after the initial domestication of maize. The inhabitants of the Maya lowlands were particularly slow to adopt sedentary lifeways [96]. Various scholars suspect that around 1000 B.C. or slightly earlier, the productivity of maize increased, stimulating greater reliance on this crop [97101]. The construction of the Ceibal plateau began at this monument of profound change in subsistence strategies and lifeways.

In this regard, the social condition surrounding the emergence of the Ceibal plateau may not be dissimilar to that of Göbekli Tepe. In both cases, unprecedented construction activities appear to have been motivated by subsistence changes: initial domestication in the case of Göbekli Tepe, and the intensified production of an existing crop at Ceibal. Large-scale construction events in those areas probably mediated collaboration and negotiation among diverse groups during the period of radical change in lifeways [50]. We should note that such changes are less obvious in the Andes and American Southeast. In the former, the combination of hunting, fishing, gathering, and horticulture persisted for millennia, whereas in the latter, the builders of monumental structures relied on wild resources without the benefit of agriculture. Nonetheless, the emergence of some centers in those regions may have been associated with rapid increases in population and intensified resource use [92,102].

These variabilities in social circumstances provide important clues to underlying processes related to early monumental constructions from across the world. An important factor in this regard may be seasonal ritual gatherings [92]. Mobile or semi-sedentary groups possibly gathered periodically and worked together to built structures on unprecedented scales. These early monumental constructions represented transformations of the lived landscape, which possibly mirrored the emergence of a new social order tied to the new ways of life [8,38,103]. The construction of prominent landmarks also paralleled people’s changing relationships to lands and landscapes. In the course of these construction projects, participants may have developed new organizational forms that coordinated different responsibilities among them. These construction events and large public gatherings probably instigated and framed new modes of political negations among participants regarding affairs of common interest. Inomata et al. have analyzed this process by applying the concept of the public sphere [50,104]. These events may also have encouraged participants to adopt new subsistence strategies and new lifestyles in some cases. In other words, we need to examine not only how political and economic bases enabled large-scale constructions but also how early building projects stimulated later political and economic changes.


Preclassic artificial plateau constructions in the Maya lowlands have received little attention from researchers. This is partly because such extensive features covered by the rainforest are difficult to recognize and because archaeological studies strongly focused on pyramids, which became the hallmarks of Mesoamerican civilizations in later periods. Our analysis of the artificial plateau at Ceibal demonstrates that its Preclassic constructions were substantially larger than pyramids and other Classic-period buildings. The Ceibal plateau was constructed at the time of social transformation from mobile lifeways relying on mixed subsistence strategies without the use of ceramics to full sedentism with a strong commitment to maize agriculture and ceramic use. The construction volume of the Ceibal plateau and its labor investment during the early Middle Preclassic Real phase suggest that this scale of construction cannot be explained with the previously-proposed population size of this community. The Real-phase construction events likely involved groups living outside of the Ceibal center, who possibly retained certain levels of residential mobility and whose traces are difficult to detect archaeologically.

Increasing data on artificial plateaus and platforms at Ceibal and other lowland Maya sites show that monumental constructions existed in the Maya lowlands before the development of strong coercive power held by centralized authorities. Although archaeologists sometimes use the scales of buildings as proxies for political centralization [51], these examples cast doubt on this assumption. We need to explore not only how monumental constructions resulted from political centralization in certain contexts but also how large construction projects stimulated social changes. These artificial plateaus transformed the lived landscape of the community and provided settings for public gatherings and rituals. Large constructions probably reflected and promoted the emergence of a new social order tied to the new ways of life. The creation of an ordered yet inclusive space may have motivated many individuals to participate in unprecedented construction endeavors, in the absence of coercion by elites. At the same time, such collective work on a large scale likely required and promoted organizational and managerial innovations among participants, which may have set the stage for later administrative centralization.

These new data from the Maya lowlands add to the emerging understanding that monumental structures were built before or during the transitions from hunting, gathering, and fishing to agriculture or mobile lifeways to sedentism in various parts of the world. The Ceibal plateau, which was constructed during the intensification of maize cultivation rather than initial plant domestication, reflects diversity in social circumstances surrounding early monumental constructions across the world, as well as their commonality in their emergence before the development of marked political centralization.

Supporting information


  1. 1. Hansen RD. Continuity and disjunction: the Pre-Classic antecedents of Classic Maya architecture. In: Houston SD, editor. Function and meaning in Classic Maya architecture Washington, D.C.: Dumbarton Oaks Research Library and Collection; 1998. p. 49–122.
  2. 2. Hansen RD. Cultural and environmental components of the first Maya states: a perspective from the central and southern Maya lowlands. In: Traxler LP, Sharer RJ, editors. The origins of Maya states Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology; 2016. p. 329–416.
  3. 3. Estrada-Belli F. The first Maya civilization: ritual and power before the Classic period. London: Routledge; 2011.
  4. 4. Estrada-Belli F. Early civilization in the Maya lowlands, monumentality, and place making: A view from the Holmul region. In: Burger RL, Rosenswig RM, editors. Early New World monumentality Gainsville: University Press of Florida; 2012. p. 198–230.
  5. 5. Willey GR, Smith AL, Tourtellot G III, Graham I. Excavations at Seibal, Department of Peten, Guatemala: introduction: the site and its setting. Cambridge, MA: Harvard University; 1975.
  6. 6. Sabloff JA. Excavations at Seibal, Department of Peten, Guatemala: ceramics. Cambridge, MA: Harvard University; 1975.
  7. 7. Smith AL. Excavations at Seibal, Department of Peten, Guatemala: major architecture and caches. Cambridge, MA: Harvard University; 1982.
  8. 8. Inomata T, Triadan D, Aoyama K, Castillo V, Yonenobu H. Early ceremonial constructions at Ceibal, Guatemala, and the origins of lowland Maya civilization. Science 2013;340(6131):467–471. pmid:23620050
  9. 9. Inomata T, MacLellan J, Triadan D, Munson J, Burham M, Aoyama K, et al. Development of sedentary communities in the Maya lowlands: coexisting mobile groups and public ceremonies at Ceibal, Guatemala. Proc Natl Acad Sci USA 2015 Apr 7;112(14):4268–4273. pmid:25831523
  10. 10. Inomata T, Triadan D, MacLellan J, Burham M, Aoyama K, Palomo JM, et al. High-precision radiocarbon dating of political collapse and dynastic origins at the Maya site of Ceibal, Guatemala. Proc Natl Acad Sci USA 2017;114(7):1293–1298.
  11. 11. Sharpe AE, Emery KF, Inomata T, Triadan D, Kamenov GD, Krigbaum J. Earliest isotopic evidence in the Maya region for animal management and long-distance trade at the site of Ceibal, Guatemala. Proc Natl Acad Sci USA 2018;115(14):3605–3610. pmid:29555750
  12. 12. Inomata T, Pinzón F, Ranchos JL, Haraguchi T, Nasu H, Fernandez-Diaz J, et al. Archaeological application of airborne LiDAR with object-based vegetation classification and visualization techniques at the lowland Maya site of Ceibal, Guatemala. Remote sensing 2017;9(6):563.
  13. 13. Inomata T, Triadan D, Pinzón F, Burham M, Ranchos JL, Aoyama K, et al. Archaeological application of airborne LiDAR to examine social changes in the Ceibal region of the Maya lowlands. PLoS ONE 2018;13(2):e0191619. pmid:29466384
  14. 14. Chase AF, Chase DZ, Weishampel JF, Drake JB, Shrestha RL, Slatton KC, et al. Airborne LiDAR, archaeology, and the ancient Maya landscape at Caracol, Belize. J Archaeol Sci 2011;38(2):387–398.
  15. 15. Chase AF, Chase DZ, Awe JJ, Weishampel JF, Iannone G, Moyes H, et al. Ancient Maya regional settlement and inter-site analysis: The 2013 west-central Belize LiDAR Survey. Adv Archaeol Practice 2014;6(9):8671–8695.
  16. 16. Hutson SR. Adapting LiDAR data for regional variation in the tropics: A case study from the Northern Maya Lowlands. J Archaeol Sci 2015;4:252–263.
  17. 17. Evans D. Airborne laser scanning as a method for exploring long-term socio-ecological dynamics in Cambodia. J Archaeol Sci 2016;74:164–175.
  18. 18. Canuto MA, Estrada-Belli F, Garrison TG, Houston SD, Acuña MJ, KovÃ¡Ä M, et al. Ancient lowland Maya complexity as revealed by airborne laser scanning of northern Guatemala. Science 2018;361(6409):eaau0137. pmid:30262469
  19. 19. Rosenswig RM, López-Torrijos R. Lidar reveals the entire kingdom of Izapa during the first millennium BC. Antiquity 2018;92(365):1292–1309.
  20. 20. Fernandez-Diaz JC, Carter WE, Glennie C, Shrestha RL, Pan Z, Ekhtari N, et al. Capability Assessment and Performance Metrics for the Titan Multispectral Mapping Lidar. Remote Sensing 2016;8(11):936.
  21. 21. Chase AF, Chase DZ, Fisher CT, Leisz SJ, Weishampel JF. Geospatial Revolution and Remote Sensing LiDAR in Mesoamerican Archaeology. Proc Natl Acad Sci USA 2012 Aug 7;109(32):12916–12921. pmid:22802623
  22. 22. Chase AF, Chase D, Awe J, Weishampel J, Iannone G, Moyes H, et al. The use of LiDAR in understanding the Ancient Maya landscape. Advances in Archaeological Practice 2014;2(3):208–221.
  23. 23. Evans DH, Fletcher RJ, Pottier C, Chevance JB, Soutif D, Tan BS, et al. Uncovering archaeological landscapes at Angkor using lidar. Proc Natl Acad Sci USA 2013 Jul 30;110(31):12595–12600. pmid:23847206
  24. 24. Fisher CT, Fernández-Diaz JC, Cohen AS, Cruz ON, Gonzáles AM, Leisz SJ, et al. Identifying ancient settlement patterns through LiDAR in the Mosquitia region of Honduras. PLoS ONE 2016;11(8):e0159890. pmid:27560962
  25. 25. Prufer KM, Thompson AE, Kennett DJ. Evaluating airborne LiDAR for detecting settlements and modified landscapes in disturbed tropical environments at Uxbenká, Belize. J Archaeol Sci 2015;57:1–13.
  26. 26. Tourtellot G III. Excavations at Seibal, Department of Peten, Guatemala: peripheral survey and excavation, settlement and community patterns. Cambridge, MA: Harvard University; 1988.
  27. 27. Inomata T, Triadan D, Aoyama K. After 40 years: revisiting Ceibal to investigate the origins of lowland Maya civilization. Ancient Mesoam 2017;28(1):187–201.
  28. 28. Inomata T, Pinzón F, Palomo JM, Sharpe A, Ortíz R, Méndez MB, et al. Public ritual and interregional interactions: excavations of the Central Plaza of Group A, Ceibal. Ancient Mesoam 2017;28(1):203–232.
  29. 29. Triadan D, Castillo V, Inomata T, Palomo JM, Méndez MB, Cortave M, et al. Social transformation in a Middle Preclassic community: elite residential complexes at Ceibal. Ancient Mesoam 2017;28(1):233–264.
  30. 30. Inomata T, Triadan D. Middle Preclassic Caches from Ceibal, Guatemala. Maya Archaeol 2016;3:56–91.
  31. 31. Palomo JM, Inomata T, Triadan D. Mortuary rituals and cranial modifications at Ceibal: from the early Middle Preclassic to the Terminal Classic period. Ancient Mesoam 2017;28(1):305–327.
  32. 32. Aoyama K, Inomata T, Triadan D, Pinzón F, Palomo JM, MacLellan J, et al. Early Maya ritual practices and craft production: Late Middle Preclassic ritual deposits containing obsidian artifacts at Ceibal, Guatemala. J Field Archaeol 2017;42(5):408–422.
  33. 33. Aoyama K, Inomata T, Pinzón F, Palomo JM. Polished greenstone celt caches from Ceibal: the development of Maya public ritual. Antiquity 2017;91(357):701–717.
  34. 34. Inomata T. The emergence of standardized spatial plans in southern Mesoamerica: chronology and interregional interactions viewed from Ceibal, Guatemala. Ancient Mesoam 2017;28(1):329–355.
  35. 35. Lowe GW. The Mixe-Zoque as competing neighbors of the Early Lowland Maya. In: Adams REW, editor. The Origins of Maya Civilization Albuquerque: University of New Mexico Press; 1977. p. 197–248.
  36. 36. Clark JE, Hansen RD. Architecture of early kingship: comparative perspectives on the origins of the Maya royal court. In: Inomata T, Houston SD, editors. Royal Courts of the Ancient Maya, Volume 2: Data and Case Studies Boulder: Westview Press; 2001. p. 1–45.
  37. 37. Clark JE. Western kingdoms of the Middle Preclassic. In: Traxler LP, Sharer RJ, editors. The origins of Maya states Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology; 2016. p. 123–224.
  38. 38. Inomata T. The Isthmian origins of the E Group and its adoption in the Maya lowlands. In: Freidel DA, Chase AF, Dowd AS, Murdock J, editors. Early Maya E Groups, solar calendars, and the role of astronomy in the rise of lowland urbanism Gainesville: University Press of Florida; 2017.
  39. 39. Willey GR. Excavations at Seibal, Department of Peten, Guatemala: general summary and conclusions. Cambridge, MA: Harvard University; 1990.
  40. 40. Chiba T, Kaneta S, Suzuki Y. Red relief image map: new visualization method for three dimensional data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 2008;37(B2):1071–1076.
  41. 41. Inomata T. Plaza builders of the Preclassic Maya lowlands: the construction of a public space and a community at Ceibal, Guatemala. In: Tsukamoto K, Inomata T, editors. Mesoamerican Plazas: Practices, Meanings, and Memories Tucson: University of Arizona Press; 2014. p. 19–34.
  42. 42. Erasmus CJ. Monument building: some field experiments. Southwestern Journal of Anthropology 1965;21:277–301.
  43. 43. Abrams EM. How the Maya built their world: energetics and ancient architecture. Austin: University of Texas Press; 1994.
  44. 44. Abrams EM. Economic specialization and construction personnel in Classic period Copan, Honduras. American Antiquity 1987;52(3):485–499.
  45. 45. Webster D, Kirker J. Too many Maya, too few buildings: investigating construction potential at Copán, Honduras. Journal of Anthropological Research 1995;51(4):363–387.
  46. 46. Ortmann AL, Kidder TR. Building Mound A at Poverty Point, Louisiana: Monumental Public Architecture, Ritual Practice, and Implications for Hunter‐Gatherer Complexity. Geoarchaeology 2013;28(1):66–86.
  47. 47. Tourtellot G III. Population estimates for Preclassic and Classic Seibal, Peten. In: Culbert TP, Rice DS, editors. Precolumbian population history in the Maya lowlands Albuquerque: University of New Mexico Press; 1990. p. 83–102.
  48. 48. Culbert TP, Rice DS. Precolumbian population history in the Maya lowlands. Albuquerque: University of New Mexico Press; 1990.
  49. 49. Inomata T. Theories of power and legitimacy in archaeological contexts: the emergent regime of power at the Formative Maya community of Ceibal, Guatemala. In: Kurnick S, Baron J, editors. Political strategies in Pre-Columbian Mesoamerica Boulder: University Press of Colorado; 2016. p. 37–60.
  50. 50. Inomata T, MacLellan J, Burham M. The construction of public and domestic spheres in the Preclassic Maya lowlands. Am Anthropol 2015;117(3):519–534.
  51. 51. Trigger BG. Monumental architecture: a thermodynamic explanation of symbolic behavior. World Archaeol 1990;22:119–132.
  52. 52. Abrams EM. Architecture and energy: an evolutionary perspective. Archaeological method and theory 1989;1:47–87.
  53. 53. Price BJ. Competition, productive intensification, and ranked society: speculations from evolutionary theory. In: Ferguson B, editor. Warfare, culture, and environment New York: Academic Press; 1984. p. 209–240.
  54. 54. Lohse JC. Archaic origins of the lowland Maya. Latin Am Antiq 2010;21(3):312–352.
  55. 55. Hammond N. The genesis of hierarchy: mortuary and offertory ritual in the Pre-Classic at Cuello, Belize. In: Grove DC, Joyce RA, editors. Social patterns in Pre-Classic Meosamerica Washington, D.C.: Dumbarton Oaks Research Library and Collection; 1999. p. 49–66.
  56. 56. McAnany PA, Storey R, Lockard AK. Mortuary ritual and family politics at Formative and early Classic K'axob, Belize. Ancient Mesoamerica 1999;10(1):129–146.
  57. 57. McAnany PA. Living with the Ancestors: Kinship and Kingship in Ancient Maya Society. Austin: University of Texas Press; 1995.
  58. 58. Clastres P. Society against the state. Oxford: Blackwell; 1977.
  59. 59. Gilman A. The development of social stratification in Bronze Age Europe. Curr Anthropol 1981;22(1):1–23.
  60. 60. Roscoe PB. Practice and political centralisation: a new approach to political evolution. Curr Anthropol 1993;34(2):111–140.
  61. 61. Scott JC. Against the grain: a deep history of the earliest states. New Haven: Yale University Press; 2017.
  62. 62. Drucker P, Heizer RF, Squier RH. Excavations at La Venta, Tabasco, 1955. Washington, D.C.: Smithsonian Institution; 1959.
  63. 63. González Lauck RB. The architectural setting of Olmec sculpture clusters at La Venta, Tabasco. In: Guernsey J, Clark JE, Arroyo B, editors. The place of stone monuments: context, use, and meaning in Mesoamerica's Preclassic tradition Washington, D.C.: Dumbarton Oaks Research Library and Collection; 2010. p. 177–205.
  64. 64. Bachand BR, Lowe LS. Chiapa de Corzo's Mound 11 Tomb and the Middle Formative Olmec. In: Lowe LS, Pye ME, editors. Arqueología reciente de Chiapas: Contribuciones del encuentro celebrado en el 60⁰ aniversario de la Fundación Arqueológica Nuevo Mundo Provo, Utah: Brigham Young University; 2012. p. 45–68.
  65. 65. Hansen RD, Suyuc L. E editors. Mirador. Guatemala City: FARES Guatemala; 2016.
  66. 66. Saturno WA, Stuart D, Beltrán B. Early Maya writing at San Bartolo, Guatemala. Science 2006:1281–1283. pmid:16400112
  67. 67. Saturno WA, Taube KA, Stuart D. The murals of San Bartolo, El Peten, Guatemala, Part 1: the north wall. Ancient America 2005;7:1–56.
  68. 68. Coe WR. Tikal Report No. 14: Excavations in the Great Plaza, North Terrace and North Acropolis of Tikal. Philadelphia: University Museum, University of Pennsylvania; 1990.
  69. 69. Bazy D, Inomata T. Multiple waves of political disintegration in the Classic Maya collapse: new insights from the excavation of Group D, Ceibal, Guatemala. Journal of Field Archaeology 2017;In press.
  70. 70. Stuart D. "The fire enters his house": architecture and ritual in Classic Maya text. In: Houston SD, editor. Function and Meaning in Classic Maya Architecture Washington, D.C.: Dumbarton Oaks Research Library and Collection; 1998. p. 373–426.
  71. 71. Coe MD, Diehl RA. In the land of the Olmec. Austin: University of Texas Press; 1980.
  72. 72. Cyphers A, Murtha T. Early Olmec open spaces at San Lorenzo, Veracruz. In: Tsukamoto K, Inomata T, editors. Mesoamerican plazas: arenas of community and power Tucson: University of Arizona Press; 2014. p. 71–89.
  73. 73. Cyphers A. The Early Preclassic Olmec: an overview. In: Traxler LP, Sharer RJ, editors. The origins of Maya states Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology; 2016. p. 83–122.
  74. 74. Andrews EW V, Bey GJ III, Gunn CM. The earliest ceramics of the northern Maya lowlands. In: Brown MK, Bey GJ, III, editors. Pathways to complexity: a view from the Maya lowlands Gainesville: University Press of Florida; 2018. p. 49–86.
  75. 75. Reese-Taylor K. Founding landscapes in the Central Karstic Uplands. In: Freidel DA, Chase AF, Dowd AS, Murdock J, editors. Maya E Groups: calendars, astronomy, and urbanism in the early lowlands Gainesville: University Press of Florida; 2017. p. 480–513.
  76. 76. Gallareta Negrón T. At the foot of the hills: early monumentality at Xocnaceh, Yucatan, Mexico. In: Brown MK, Bey GJ, III, editors. Pathways to complexity: a view from the Maya lowlands Gainesville: University Press of Florida; 2018. p. 276–291.
  77. 77. Robles Castellanos F, Ceballos Gallareta T. The genesis of Maya complexity in the northern region of the Yucatan peninsula. In: Brown MK, Bey GJ III, editors. Pathways to complexity: a view from the Maya lowlands Gainesville: University Press of Florida; 2018. p. 223–251.
  78. 78. Drucker P. La Venta, Tabasco: a study of Olmec ceramics and art. Washington, D.C.: Smithsonian Institution; 1952.
  79. 79. Lowe GW. Olmec horizon defined in Mound 20, San Isidro, Chiapas. In: Coe MD, Grove D, editors. The Olmec and their neighbors Washington, D.C.: Dumbarton Oaks Research Library and Collection; 1981. p. 231–256.
  80. 80. Zedeño MN, Ballenger JAM, Murray JR. Landscape engineering and organizational complexity among Late Prehistoric Bison Hunters of the northwestern Plains. Curr Anthropol 2014;55(1):23–58.
  81. 81. Saunders JW, Mandel RD, Saucier RT, Allen ET, Hallmark CT, Johnson JK, et al. A mound complex in Louisiana at 5400–5000 years before the present. Science 1997 Sep. 19;277(5333):1796–1799.
  82. 82. Hodgson JG, Clark JG, Gallaga Murrieta E. Ojo de Agua Monument 3: A New Olmec-Style Sculpture from Ojo de Agua, Chiapas, Mexico. Mexicon 2010;32:139–144.
  83. 83. Love M. Early states in the Southern Maya Region. In: Traxler LP, Sharer RJ, editors. The origins of Maya states Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology; 2016. p. 271–328.
  84. 84. Rosenswig RM. Agriculture and monumentality in the Soconusco region of Chiapas, Mexico. In: Burger RL, Rosenswig RM, editors. Early New World Monumentality Gainesville: University Press of Florida; 2012. p. 111–137.
  85. 85. Laporte JP, Fialko V. Reencuentro con Mundo Perdido, Tikal, Guatemala. Ancient Mesoam 1995;6:41–94.
  86. 86. Schmidt K. Göbekli Tepe—the Stone Age sanctuaries: new results of ongoing excavations with a special focus on sculptures and high reliefs. Documenta Praehistorica 2010;37:239–255.
  87. 87. Kidder TR. Transforming Hunter-Gatherer History at Poverty Point. In: Sassaman KE, Holly DH, editors. Hunter-gatherer archaeology as historical process Tucson: University of Arizona Press; 2011. p. 95–119.
  88. 88. Fuchs PR, Patzschke R, Yenquec G, Briceño J. Del Arcaico Tardío al Formativo Temprano: las investigaciones en Sechín Bajo, Valle de Casma. Boletín de Arqueol PUCP 2009;13:55–86.
  89. 89. Shady Solis R, Haas J, Creamer W. Dating Caral, a preceramic site in the Supe Valley on the Central Coast of Peru. Science 2001 20111110;292(5517):723–726. pmid:11326098
  90. 90. Sassaman KE. Complex Hunter-Gatherers in Evolution and History: A North American Perspective. J Archaeol Res 2004;12:227–280.
  91. 91. Fish PR, Fish SK, DeBlasis P, Gaspar MD. Monumental shell mounds as persistent places in southern coastal Brazil. In: Thompson VD, Waggoner JC Jr, editors. The archaeology and historical ecology of small scale economies Gainesville: University Press of Florida; 2013. p. 120–140.
  92. 92. Stanish C. The evolution of human co-operation: ritual and social complexity in stateless societies. Cambridge: Cambridge University Press; 2017.
  93. 93. Burger RL, Rosenswig RM editors. Early New World Monumentality. Gainsville: University Press of Florida; 2012.
  94. 94. Blake M. Maize for the Gods: unearthing the 9,000-year history of corn. Oakland, California: University of California Press; 2015.
  95. 95. Staller JE, Tykot RH, Benz BF. Histories of mazie: multidisciplinary approaches to the prehistory, linguistics, biogeography, domestication, and evolution of maize. Walnut Creek: Left Coast Press; 2010.
  96. 96. Clark JE, Cheetham D. Mesoamerica's tribal foundations. In: Parkinson WA, editor. Archaeology of Tribal Societies Ann Arbor: International Monograph in Prehistory; 2002. p. 278–339.
  97. 97. Blake M, Clark JE, Voorhies B, Love MW, Chisholm BS. Prehistoric subsistence in the Soconusco region. Curr Anthropol 1992;33(1):83–94.
  98. 98. Arnold PJ III. Settlement and subsistence among the Early Formative Gulf Olmec. Journal of Anthropological Archaeology 2009;28:397–411.
  99. 99. Kennett DJ, Thakar HB, VanDerwarker AM, Webster DL, Culleton BJ, Harper TK, et al. High-precision chronology for Central American maize diversification from El Gigante rockshelter, Honduras. Proc Natl Acad Sci U S A 2017 Aug 22;114(34):9026–9031. pmid:28784803
  100. 100. Clark JE, Pye ME, Gosser DC. Thermolithics and corn dependency in Mesoamerica. In: Lowe LS, Pye ME, editors. Archaeology, Art, and Ethnogenesis in Mesoamerican Prehistory: Papers in Honor of Gareth W. Lowe Provo, Utah: Brigham Young University; 2007. p. 23–42.
  101. 101. Rosenswig RM, VanDerwarker AM, Culleton BJ, Kennett DJ. Is it agriculture yet? Intensified maize-use at 1000 cal BC in the Soconusco and Mesoamerica. Journal of Anthropological Archaeology 2015;40:89–108.
  102. 102. Clark JE, Gibson JL, Zeidler J. First Towns in the Americas: Searching for Agriculture, Population Growth, and Other Enabling Conditions. In: Bandy MS, Fox JR, editors. Becoming Villagers: Comparing Early Village Societies Tucson: University of Arizona Press; 2010. p. 205–245.
  103. 103. Inomata T. The Emergence of Standardized Spatial Plans in Southern Mesoamerica: Chronology and Inter-Regional Interactions Viewed from Ceibal, Guatemala. Ancient Mesoamerica 2013;To be submitted.
  104. 104. Habermas J. The structural transformation of the public sphere: an inquiry into a category of Bourgeois society. Cambridge, Mass.: MIT Press; 1989.