Sampling and Measurement

Our study relies on room counts, plaza areas, and degrees of plaza enclosure for sites with plazas throughout the Pueblo World. It does not include plazas in the Hohokam World (see Wallace et al. Reference Wallace, Lindeman, Kyle Woodson, Loendorf and Lewis2020), which are more loosely delineated, but they would make an interesting comparison in the future. We focus on size and accessibility because these two architectural attributes can be recorded consistently in Southwest pueblos and tied to different social uses. A total of 442 settlements with 861 total plazas are included in this analysis based on the availability of suitable data and scaled maps (Supplementary Table 1; Figure 2). Date ranges and room counts for each settlement included in this study were provided by cyberSW, a regional archaeological database of settlements, which is largely based on estimates provided by individual investigators.Footnote ¹

Figure 2.

Heat map of the total number of plazas in the Ancestral Pueblo Southwest (AD 850–1550). (Color online)

Plaza spaces are defined and reported in varied ways. Although there is a tendency to define plazas as fully enclosed architectural spaces, many plaza areas known ethnographically are not necessarily so bounded. For example, Kroeber (Reference Kroeber1917) mapped all named Zuni plazas, and some of these open spaces are adjacent to linear room blocks with ritual architecture (kivas) but are not fully enclosed (although they may have been more enclosed in the past). In the Hopi area, street-like plazas are often bounded on two sides by linear room blocks, especially in the historic period (Bernardini et al. Reference Bernardini, Wayne Lomayestewa, Schachner, Bernardini, Koyiyumptewa, Schachner and Kuwanwisiwma2021; Mindeleff Reference Mindeleff1891). Plazas are always adjacent to room blocks but may be bounded by berms (especially in the case of Chaco and post-Chacoan plazas), walls, or natural topographic relief.

Given this diversity in identification criteria, we used a standardized recording and measurement procedure so that we could compare plaza spaces across all sites in this study. Plazas were identified from published and unpublished works and from site plans in regional repositories and archives. When possible, plazas were identified by descriptions in reports or labels on the site plan itself. When no such descriptions were present, plazas were identified and delineated based on the architecture of the site and available topographic information.

Spaces were labeled as plazas if they met at least one of the following criteria:

1. 1. It is an open space notably larger than the largest masonry room, and it is fully enclosed by architecture (rooms or retaining walls) on all four sides.

2. 2. It is an open space enclosed by architecture on three sides.

3. 3. It is an open space bracketed on two sides by right-angled room blocks or opposing room blocks, also known as street plazas (Adams Reference Adams1991; Cameron Reference Cameron1999).

4. 4. It is an open space surrounded by an assortment of structures such as a great house, berms, kivas, or other architecture that may not be fully conjoined but encircles an open space.

Recorded Variables

Population Size. We relied on room counts as a means of estimating population size, using a simple conversion of one person per room. Room counts were taken from the values recorded for each site in cyberSW and were derived from site forms, maps, and other unpublished and published works. There are both advantages and disadvantages to using room counts as a proxy for population. One of the major advantages is that this variable is regularly recorded for pueblo architecture across the region. A disadvantage is that room sizes vary widely across the Pueblo World because of differences in household size and construction material availability (Cameron Reference Cameron1999). Another disadvantage specific to the Chaco great houses in our sample is that there are many rooms in these structures that were not used for residential purposes, and as a result population estimates for these sites tend to be lower than the total number of rooms. We address this problem in the sections on Chaco great houses.

Plaza Capacity. We assumed that the size of a plaza is related to its intended capacity. By calculating the size of plaza areas, we obtained one line of evidence for assessing whether a given plaza could have accommodated more or less than the entire pueblo’s resident population. We measured plaza areas by delineating the plazas on plan maps and then drawing polygon features over the map image using Adobe Photoshop. The area of a plaza was measured using the polygon tool in ImageJ (Schneider et al. Reference Schneider, Rasband and Eliceiri2012). Structures inside the plaza, such as masonry rooms, kivas, or wells, were measured and subtracted from calculated plaza areas. The area of the plaza was defined as follows (Figure 3):

1. 1. For a plaza with four enclosing walls, the interior space is enclosed by the walls (Figure 3a).

2. 2. For a plaza with three enclosing walls, a line drawn across the open direction, from the interior corner to the interior corner of the “wings” of the U-shaped room block formed the fourth edge of the perimeter (Figure 3b).

3. 3. For a plaza with two enclosing L-shaped walls, lines perpendicular to each wall were drawn until they met to form a corner. These two walls and two lines form the perimeter of the plaza (Figure 3c).

4. 4. For a plaza with two opposing room blocks, lines were drawn from the corners of the room blocks facing each other (Figure 3d).

5. 5. For a plaza encircled by noncontiguous masonry, lines connecting the interior corners of each feature defined the perimeter (Figure 3e).

Figure 3.

Schematic figure illustrating protocol for delineating plaza areas for different types of architectural site layouts. Dashed lines indicate projected boundaries when no physical boundaries are observable.

Accessibility. We also explored degrees of accessibility, asking how degrees of access control changed through time. Although we cannot reconstruct the full scope of past uses of plazas, certain architectural features can be illustrative, especially those that encourage or impede visual or physical access. Even though many studies focus on those physical features that are most imposing (e.g., the presence or absence of doors or high walls), Ashmore (Reference Ashmore and Stephen2007:190) reminds us that other features such as low walls, sequences of unaligned doorways, or narrow passages serve to “channel movement and sight without preventing them.” In our study, we defined accessibility based on how many walls or architectural features were used to bound the plaza space.

Dating. Date ranges for each site are based on those in cyberSW, which are usually those assigned by the original investigators with modifications from subsequent analyses. Temporal assignments are based on the period when the site reached its maximum extent, which is usually the latest period of occupation, rather than the midpoint used by Chamberlin (Reference Chamberlin, Donna and Van Keuren2011). This complements our approach to recording plaza attributes, which also relied on the final architectural footprint of each. We assigned each site to one of six time periods that overlap slightly in time (Table 1). Note that the post-Chaco period and Pueblo III overlap in time but are differentiated based on whether they have a great house (post-Chaco) or not (Pueblo III). Figure 4 shows the distribution of plazas and their sizes by different time periods throughout our entire study area.

Figure 4.

Plaza areas for all sites included in this study by period. (Color online)

Table 1.

Time Periods and Date Ranges Used in This Study.

Confidence. Based on available information, each plaza was assigned a confidence rating of high, medium, or low. Plazas enclosed by three or four conjoined walls of masonry or that were labeled clearly on a map received a high confidence rating. Plazas in front of a room block and encircled by berms or noncontiguous masonry or defined by walls that were projected and not fully excavated were labeled as medium confidence. Medium-confidence plazas were often irregular in shape, had significant unbounded edges, or were clearly marked plazas but with vague or poorly marked scales on their original maps. Low-confidence plazas were highly irregular in shape, with large unbounded areas or unclear boundaries, or they were located on maps in which the architecture was poorly or inconsistently mapped with vague or absent scales. Most L-shaped plazas were also labeled low confidence unless topography or other features were present to allow consistent measurement. Only high- and medium-confidence plazas are included in this study.

Great House Settlements

Early Chaco (AD 800–1020). The earliest recognizable plazas in pueblo settlements in the Southwest are constructed in the architectural layouts of Early Chaco great houses. Although not all early great houses (or “proto-great houses”) had plazas, those that did were relatively consistent in size, regardless of the size of the accompanying great house. Many early great houses have large middens with abundant ceramics (Windes Reference Windes and Stephen2007), and those with plazas were primarily located in the Chuska and Chaco Canyon subregions, although McPhee Pueblo in the Central Mesa Verde is a significant exception (Wilshusen Reference Wilshusen, Richard and Mendoza2017). In general, because of the continuous modification of the great houses in Chaco Canyon itself, we were not able to consistently measure plazas and count the associated rooms for the Early Chaco occupations of most of these sites. However, those with sufficient information about their early occupations demonstrate that early great houses had modest-sized plazas, and those in Chaco Canyon were accompanied by great houses with more rooms than Chuska settlements. Although the sample size for Early Chaco great houses is too small to be statistically significant, it is interesting that plaza size seems to have been independent of settlement size during this period. This suggests that early plazas were used for similar activities or by similarly sized groups, regardless of the size of the associated great house. Despite the lack of dependence on room count, Early Chaco plazas are quite variable in their configuration: There are examples of plazas with two, three, and four walls, indicating different degrees of accessibility at this time.

Late Chaco (AD 1020–1120). The Late Chaco period as delineated here is when Chaco great house construction peaked. As population increased and Chaco great houses became larger and more formalized during this period, plazas also increased in size. Although plazas of the Early Chaco period tended to vary independently of settlement size, both larger and smaller plazas were introduced during the Late Chaco period. With an r² value of 0.64, settlements during this period have the strongest relationship between room counts and plaza areas of any period in our study. In other words, a greater percentage of the variance in the Late Chaco period plaza area is accounted for by the relationship to room count than for any other period. As settlements increased in size, plaza areas also increased at a relatively predictable rate. Subsequent time periods have much more variability in this relationship.

Another important finding is that plaza area grows more slowly than room count, with a regression coefficient (slope) of about two-thirds. This is often referred to as a “sublinear” scaling relationship. Hanson and colleagues (Reference Hanson, Ortman, Bettencourt and Mazur2019) observed a similar slope for the relationship between population and forum/agora area in ancient Greek and Roman cities. They also argued that this relationship is to be expected because interactions in a social network (the links) grow faster than the number of individuals (the nodes). This means that each person can have the same number of interactions over time while using less space in a larger community. The implication of this logic is that plazas exhibiting a sublinear relationship with the associated settlement population were primarily used by community residents. In Chaco great houses, this sublinear relationship suggests that the practices taking place in plazas may have been conservative in their configuration, regardless of increasing populations.

An important issue, however, is that room counts may be a less accurate proxy for Late Chaco great house populations than for sites of other periods. For example, Pueblo Bonito with its 700 rooms has been estimated by various archaeologists to have housed from 100 to 1,200 people (Mills Reference Mills2023:Table 11.2). One recent survey of 13 archaeologists provided ranges between 100 and 800 people, with a median of 263 (Minnis Reference Minnis, Carrie and Plog2015). That Pueblo Bonito was at least partly residential is supported by the presence and scale of the associated middens, which contain ample evidence for residential activities (Crown Reference Crown2016; Plog Reference Plog and Peter2018; Plog and Watson Reference Plog and Watson2012; Plog et al. Reference Plog, Heitman, Watson, Barbara and Fowles2017) and assemblage compositions that do not differ appreciably from those of small houses (Crown Reference Crown2016:Figures 10.4 and 10.5). Another possibility is that for Late Chaco great houses, the room count is proportional to the number of people who periodically interacted within the great house but does not necessarily reflect the number of full-time residents. This latter possibility would also be consistent with the observed scaling relationship.

The Late Chaco period in this study corresponds with the maximum regional fluorescence of the Chaco World. And although the Chaco World clearly extended throughout the northern US Southwest, the nature of the open space and the corresponding activities varied significantly. The largest sites and the largest plazas are found in the Chaco Core. Great houses constructed in Chaco Canyon were clearly designed to accommodate large gatherings of people for plaza-based activities. For example, Crown (Reference Crown2020) notes that in its initial configuration Room 28 at Pueblo Bonito, which contained a large number of vessels used to serve a frothed cacao beverage to big groups, was directly accessible from the plaza in front of it. Although the Chaco Core emerged as a regional center during this time, several small great houses with small plazas were introduced along the western and southern edges of the San Juan Basin. In addition, sites with comparatively large plazas are established along the most southwestern fringes of the Chaco World along the Puerco River (of the West).

Plazas of this period, especially those in the Chaco Core, are predominantly bounded by four walls, indicating formally demarcated boundaries designed to clearly differentiate activities held within from those outside. Comparatively, Chaco great houses located on the peripheries of the Chaco World tend to have plazas with more permeable boundaries. Overall, this indicates that the Chaco Core emerged as a major hub of plaza-based activities that were more exclusive than those associated with great houses in other areas of the Chaco World.

Post-Chaco Great Houses (AD 1120–1275). In the early twelfth century, political power shifted from Chaco Canyon to the Middle San Juan area and the edges of the San Juan Basin. Post-Chaco great houses with plazas were constructed in a wider area than were the Late Chaco great houses. In general, these post-Chaco great house sites have much higher room counts and larger plazas than their Late Chaco precedents. At this time, new political centers that developed in the Middle San Juan and Mesa Verde regions maintained many architectural ideals established in Chaco Canyon and are referred to as post-Chaco great houses. Great houses in these regions are large sites with correspondingly large, four-walled plazas.

Elsewhere, these relationships are less consistent. Overall, the relationship between settlement size and plaza size is weaker than it was during the Late Chaco period, in that the r² value is lower but the scaling coefficient remains similar. This suggests two points. First, given that post-Chaco centers are typically interpreted as residential, the consistent slope of the room count to plaza area relationship between the post-Chaco and Late Chaco periods reinforces our suggestion that room counts in Late Chaco great houses are proportional to the number of people who gathered there periodically. Second, the weaker relationship between room count and plaza area suggests a somewhat looser relationship between settlement population and social gathering size.

In addition to the large, well-known post-Chaco centers established in the Middle San Juan and Mesa Verde regions, important post-Chaco communities also formed to the south and west of Chaco Canyon. These were especially concentrated in the Chuska Mountains, along the Upper Puerco River, and in the Upper Little Colorado region. Plazas established in regions outside the political centers of the Middle San Juan exhibit more variability in degrees of enclosure. Unlike the post-Chaco great houses to the north, post-Chaco great houses elsewhere tend to have more permeable boundaries. Higher numbers of two- and three-walled plazas suggest that these plazas were intentionally designed to be more permeable and less restrictive.

Non-Great-House Settlements

Pueblo III Settlements (AD 1100–1275/1280). Although post-Chaco great houses with plazas were established to the north and southwest of Chaco Canyon, there is also a proliferation of roughly contemporaneous Pueblo III settlements with plazas that do not feature great house architecture. In general, Pueblo III sites with plazas do not co-occur spatially with great houses.

The room count to plaza area relationship across Pueblo III (non-great-house) sites is both weaker and shallower than for Late Chaco and post-Chaco great houses. The coefficient for the overall period is strongly sublinear (0.33), suggesting that the size of the groups that met in plazas grew much more slowly than the community size. This may be consistent with increased exclusivity in plaza use, perhaps through greater use of the plazas by councils of representatives, rather than for public gatherings. Yet, there was also a resurgence in plazas with more loosely defined boundaries during this period. Of all Pueblo III plazas, 22% (23 of 103) are bounded by only two walls. This is the highest percentage of two-sided plazas of any period. Although a two-sided plaza may not fit most definitions of a plaza, as discussed earlier, we included plazas that were clearly bounded by berms, walls, and other topographic features. The increased fluidity and accessibility of plazas may indicate less managerial oversight of the activities that took place in plazas (Potter Reference Potter and Katherine1998:145–146).

The overall relationship between room counts and plaza areas of Pueblo III plazas is the most variable in size and configuration of any period (r² = 0.08). Some of this variability is reduced when these patterns are examined spatially. Settlements established in the Upper Little Colorado are generally very similar to Chaco great house plazas. They tend to be large, with spacious, four-walled plazas, similar to the pattern established in Chaco great houses with plazas. Much of the variability of the Pueblo III period comes from elsewhere.

Compared to the large settlements and plazas in the Upper Little Colorado, communities established in the Lower Little Colorado and Middle Little Colorado regions and along the Verde River tend to have larger plazas than expected, given their small room counts. In particular, the Middle Little Colorado region features several smaller communities with modest plazas. Although most of these sites also have relatively low room counts, plazas in the Middle Little Colorado are extremely variable in size and configuration, with many examples of permeable two- and three-walled plazas. Small settlements with small plazas area were also established along the southern peripheries of the San Juan Basin, extending into the Mogollon Rim region. These communities in the Mogollon Rim tend to have small, four-sided enclosures.

Early Pueblo IV (AD 1275/1280–1325). During the Early Pueblo IV period, settlements explode in size, and large, bounded plazas emerge as a core component of settlement layouts. In addition, the variability seen across twelfth- and mid to late thirteenth-century sites is replaced by a much clearer relationship between settlement size and plaza area during this period. A regression coefficient of 0.78 indicates that as settlement size increased, the scale of their plazas also increased in a similar fashion to that observed across earlier Late Chaco and post-Chaco great houses. However, an r² value of 0.4 indicates that more than half the variation in plaza area is not accounted for by room count. Early Pueblo IV period plazas are also far more enclosed than preceding time periods. Most plazas are enclosed by either three or four walls (83.2%); only 16.7% are defined by two walls.

When the scaling relationship is calculated at a subregional scale, however, a new scaling relationship between settlement size and plaza area clearly emerged among the Eastern Pueblos of the Northern Rio Grande at this time (Figure 9). A regression coefficient of 1.16 for the Eastern Pueblos is far higher than ever seen before. Coefficients greater than one are often referred to as “superlinear.” In this case, the superlinear relationship indicates that larger settlements maintained more plaza space per resident than did smaller settlements. Plazas had far more space than was needed merely for residents to interact: This pattern indicates that plazas in larger communities could accommodate far more participants in plaza-based activities than the resident population of that pueblo itself. Ortman and Coffey (Reference Ortman, Coffey and Scott2019) argue that this pattern marks the emergence of feast-day-type ceremonialism, in which plazas were designed not merely for social mixing but also to accommodate large groups of dancers in formation, with an audience gathered around the perimeter. They also show that plaza space is proportional to the total number of people who lived within a day’s walk of these villages, consistent with a model in which people attended dances in several adjacent villages over the course of a year. Market economic exchange along the lines suggested by Abbott and others (Reference Abbott, Smith and Gallaga2007) for the Phoenix Basin likely also occurred as part of these ritual events.

Figure 9.

Regression graphs comparing log-transformed room count and plaza areas in Western Pueblo and Northern Rio Grande settlements in the Early Pueblo IV and Late Pueblo IV periods. (Color online)

At this time, many comparably sized settlements were established beyond the Northern Rio Grande. Small settlements that initially formed along the Little Colorado River in the Pueblo III period were replaced by large, aggregated villages with correspondingly large plazas often bound by four walls. Expansion into the Mogollon Rim region and the upper Salt and Gila drainages occurred at this time as well, and small clusters of settlements with relatively large plazas were constructed. During this period many areas of the northern Southwest were depopulated, and migrants contributed to the process of coalescence in these Early Pueblo IV settlements (Hill et al. Reference Hill, Clark, Doelle and Lyons2004). In these sites, the sublinear relationship between room counts and plaza area persists (coefficient = 0.74), suggesting continued use of plazas primarily by the residents of these settlements, many of which were multicultural.

Late Pueblo IV (AD 1325–1550). The most dramatic shift in settlement dynamics and the use of public space in the Ancestral Pueblo World occurred in the Late Pueblo IV period. The patterns of dispersal that characterize the Early Pueblo IV period are replaced by widespread aggregation into a smaller number of settlements that are much larger than those of any previous period. These large settlements also feature the largest plazas ever constructed. Continuing the pattern established in the Northern Rio Grande during the Early Pueblo IV period, a regression coefficient of 1.136 indicates that Late Pueblo IV plazas offered far more space than the resident population needed for everyday activities; this suggests the solidification of feast-day-type ceremonialism that persists to the present. Late Pueblo IV plazas maintain the pattern of enclosure established in the Early Pueblo IV period and are almost always enclosed by either three or four walls (82.9%). They are incorporated into highly defensible site layouts.

Outside the Northern Rio Grande area, Late Pueblo IV period plazas are more variable. Although the sample sizes for Late Pueblo IV Hopi and Cibola sites are too low to be statistically sound, the differences remain illustrative. Hopi settlements had an even higher regression coefficient (1.68) than the Northern Rio Grande (1.18), driven by a few exceptionally large plazas, the largest of which is at Sikyatki (Bernardini et al. Reference Bernardini, Wayne Lomayestewa, Schachner, Bernardini, Koyiyumptewa, Schachner and Kuwanwisiwma2021:Figure 8.22). This site’s plaza could have accommodated many times the number of people indicated by the number of rooms; these people would have been drawn from other Hopi villages. By contrast, Late Pueblo IV Cibola sites are much more compact and have a much lower regression coefficient (about half that of the Northern Rio Grande at 0.63).

Settlement Scaling

Although plazas occur in small numbers in Early Chaco great houses, relationships between settlement size and plaza size appear to crystallize in the Late Chaco period. There is a predictable relationship between Late Chaco settlement sizes and plaza sizes, which is consistent with the use of plazas primarily as places for social interaction among residents. After about AD 1000, the Chacoan plaza is formal, predictable, and often bounded by four walls. The stark difference between Chaco great house plazas (of all time periods) and Late Pueblo IV period plazas suggests that Chaco and Late Pueblo IV period plazas were not used in the same way, especially those in the Northern Rio Grande and Hopi areas. Importantly, the dramatic difference between the sublinear scaling coefficients of the Late Chaco and post-Chaco periods and the superlinear scaling coefficient of the Pueblo IV period suggests that contemporary Northern Rio Grande Pueblo public ritual practices originated during the Pueblo IV period that were not part of the earlier, much more exclusive gatherings in Chaco great houses. The exclusivity of Chacoan plazas suggests that events held within them were likely restricted and intimate, offering only limited participation.

In the chaotic reorganization that followed the breakdown of Chacoan regional authority during the twelfth and thirteenth centuries, this predictable relationship was disrupted and replaced with high degrees of variability in settlement size and plazas. By the Early Pueblo IV period, settlements were much larger than ever before, and plazas had reemerged as a core component of architectural layouts. Although many well-known Early Pueblo IV settlements were constructed in the Northern Rio Grande, our study also highlights the emergence of comparable plaza-centered settlements elsewhere, especially in the Tusayan and Cibola regions and throughout the Mogollon Rim. Most Early Pueblo IV settlements feature a sublinear relationship similar to that of the Late Chaco and post-Chaco periods. However, a superlinear relationship emerges for the first time in settlements in the Northern Rio Grande: It is consistent with mass dances with audiences drawn from the entire social network of a community. Plazas in large Northern Rio Grande settlements at this time became much larger than expected for social interaction only among residents and likely accommodated periodic asynchronous events such as market exchange across villages. By the Late Pueblo IV period, this superlinear scaling relationship was adopted elsewhere, especially among highly aggregated Tusayan and Cibola villages.

Accessibility

What does the analysis of Chaco great house plazas tell us about how these villages are different from later Pueblos? First, given that one (but not the only) use of plazas is for ritual performances, Chaco great house ceremonialism was much more exclusive than that of the Pueblo IV period. Consider Pueblo Bonito, which had two plazas bisected by a north–south wall. Each half would not have accommodated a large number of people, and performances were likely more intimate. There is a good argument to be made that the smaller, more compact Chaco great house plazas were used by a subset of the entire community. Given that great houses are interpreted as the residences of higher-status groups (Plog and Heitman Reference Plog and Heitman2010), political control over the performance spaces would have been an important aspect of their use. Whether this control was in the hands of lineage or sodality leaders, or both, is more difficult to interpret from size alone.

At several great houses, including two with the largest plazas—Pueblo Bonito and Chetro Ketl—storage rooms containing objects used in ritual performances were found in spaces adjacent or close to plazas. These artifacts include cylinder jars for frothing and drinking cacao (Crown Reference Crown2020), shell trumpets (Mills and Ferguson Reference Mills and Ferguson2008), painted wooden objects used as altar pieces and dance regalia (Vivian et al. Reference Gwinn, Dodgen and Harrison Hartmann1978), and the pigments, prepared paints, and associated tools used in painting them (Hanson Reference Hanson2023). Moreover, the density of roofed ceremonial architecture in or adjacent to great house plazas—especially at Pueblo Bonito—suggests the involvement of multiple groups, including ritual sodalities. Great kivas within or adjacent to plazas were the largest of these roofed ritual structures. Reconstructions of great kiva capacities in other areas of the Southwest (Dungan and Peeples Reference Katherine A. and Peeples2018) suggest that hundreds of people could have participated, but not the thousands that Pueblo IV plazas contained.

Exclusivity in plaza access does not mean that people outside a specific Chacoan great house network did not participate in larger-scale ceremonialism. Most great houses were multistoried, a pattern that extends into the Pueblo IV period and even to the present day: their roofs are commonly used ethnographically to observe activities taking place within plazas. Given the density of settlements in what is known as “Downtown Chaco” within the Chaco Core area, it is likely that those excluded from the activities that took place within great house plaza spaces would have known about their scheduling and execution. The sounds from performances could have been heard well beyond the walls of the pueblo (Van Dyke et al. Reference Dyke, Ruth, Primeau, Throgmorton and Witt2024; Witt and Primeau Reference David E. and Primeau2019).

In addition, Chaco great house communities had many exterior spaces, including roads and causeways, that were likely used for processions and other ceremonial activities. Several authors have suggested that the trash mounds of Pueblo Bonito were places on which elite orators and other ritual leaders could have stood to project sounds to public gatherings (Van Dyke Reference Dyke and Ruth2008; Witt and Primeau Reference David E. and Primeau2019; but see Chamberlin [Reference Chamberlin, Donna and Van Keuren2011] for a similar interpretation for Pueblo IV period plazas). Wills and colleagues (Reference Wills, Williams, Murphy, Przystupa and Wetherbee2021) document the presence of wooden structures anchored to the mesa side between Chetro Ketl and Pueblo Bonito, which along with balconies would have provided views of processions coming down the staircase behind Chetro Ketl. They further conclude that the Hillside Ruin, immediately to the east of Pueblo Bonito, was intentionally constructed to extend the use of elevated “observational space overlooking public areas” (Wills et al. Reference Wills, Williams, Murphy, Przystupa and Wetherbee2021:17) during the early twelfth century. They compare the amount and accessibility of these two kinds of spaces—intramural plazas and extramural open spaces between buildings—to suggest different practices and concerns about visibility and access related to each kind of open space.