Part 5: Milpa Survey and Mapping
We mapped five cleared fields in 2003, 23 in 2002, and eight during the 2001 season, for a total of 36 milpas, encompassing 52.99 hectares (Figure 11). Inspection of aerial photographs of Mayapán and its nearby environment from flights in 2002 to assess hurricane damage also confirm that we have mapped all of the major fields available. Our sample in the northeastern portion of the city remains smaller than our samples of other areas due to the prevalence of ranching rather than farming activities in this area. Many of the fields were previously mapped by the Carnegie, but our maps document finer details such as additional structures and albarrada houselot wall enclosures that will give us better information about the settlement organization, dwelling size and elaboration, and size of residential clusters associated with albarrada-defined solares. With the use of an EDM and control points established by survey grade GPS units, Hare’s maps are accurate to within 1cm. We also designed a classification system for different types of settlement units (particularly residences) that will aid us in quantitatively comparing feature distributions in different parts of the city. One problem we faced was that cleared milpa areas often did not encompass entire solares. Instead, we have many samples of partial albarrada enclosures. We made an effort to document (with GPS) the remaining parameters of albarrada walls that extended into uncleared areas. All survey and mapping work was performed under the supervision of Timothy S. Hare.
The laboratory side of this work using GIS programs has been extremely time consuming, As a result, we are able to describe the features recorded and to provide maps of all work accomplished, and analysis of the large amount of data we have collected is currently in progress. We are analyzing variables as structure size, design, and elaboration, the number of structures per solare and varying occupational density over space, and the size of various solares across the city. Brown has analyzed some of these variables with the smaller sample of solares he studied for his dissertation (1999), and we plan to follow a similar approach. We will use the settlement data in conjunction with the results of artifact analysis from surface collections and test pits. In general, we note that residential and artifact density as well as overall complexity of features tapers off with distance from the monumental center (with a few exceptions). Are these trends attributed to different functions of various parts of the city? Was gardening more prominent closer to the city walls? Or were the residents of these areas more impoverished? Perhaps the outlying settlement zones were settled later, resulting in less urban congestion and less debris. Our analyses will address these important possibilities.
Albarrada Analysis. A GPS survey of features outside of cleared milpas was also performed in 2002. A major focus of this survey was the documentation of albarrada wall alignments, particularly double albarradas that could represent alleys or pathways through the city. Double albarradas are most often formed by pairs of walls that outline adjacent houselots, but these also probably served as pedestrian walkways throughout Mayapán. This survey began by documenting known areas with double albarradas and also targeted key features such as cenotes and major gates of the city. The GPS survey also documented the location of modern features such as roads and trails to assist in future navigations through the site. The results of this survey document many segments of potentially significant pathways at the city. We have begun our analysis of albarradas (walled, partitioned spaces) within the city of Mayapán, following up on Bullard’s pioneering work on these boundary walls at the site during the Carnegie investigations. We have two sources of data. First, we have the mapped enclosures and structures within our sample of 36 cleared milpa areas documented over the past three years. Second, we obtained unpublished copies of additional boundary wall maps documented by Bullard at Mayapán from the archives of the Peabody Museum of Harvard. Hare has digitized Bullard’s data and our own into ArcGIS, enabling us to perform a more robust spatial analysis.
Several research subjects can be addressed through the study of albarradas at Mayapán. Some albarradas delineate paths or lanes between residential and open areas and hence reflect patterns of pedestrian traffic through the city’s neighborhoods. Albarradas that delineate houselots, or solares, allow us to compare sizes of residential and social units throughout the city. These delineated residential units, or solares, provide valuable information regarding the meaning of the association of particular structures beyond that based on the usual criteria of proximity and orientation. Other albarradas mark space that does not contain structures, including small enclosures or pens and possible field or orchard spaces that may reflect activities related to agriculture and livestock management.
Bullard noted that in addition to houselot albarrada walls that likely defined ancient solares or domestic yards, some albarradas formed lanes that permitted passage through the city (Figure 25). The best example was found by Bullard west of the Itzmal Ch’en group at the city’s east end. We extensively searched for additional lanes at Mayapán, with only limited results. As Bullard, and later, Clifford Brown, noted, the lanes are commonly formed by parallel, adjacent houselot boundary walls that allow passage through neighborhoods without entering specific solares – these are usually short. We have identified seven such parallel wall lanes at Mayapán – none of these were previously documented by Bullard or Brown. The average length is 33.1m, with length range extending from 5.9 to 96.8 m. This master map shows the location of all such parallel houselot lanes we have documented at the city (Figure 26). They do not appear to cluster in particular areas of the city, but they are not located in the densest area of occupation near the city center, nor do they appear in the northeast corner of the site around Itzmal Ch’en. Given their location and low frequency, they do not appear to serve the needs of pedestrian traffic beyond the neighborhoods where they are found.
We have also identified 16 lanes not seemingly formed by pairs of adjacent houselot walls (Figures 27, 28). – None of these were previously documented by Bullard or Brown. The average length is 17.21 m, with length range extending from 7.5 to 46.2. This master map shows the location of all double-albarrada lanes not formed by pairs of parallel houselot walls that we have documented at the city. They do not appear to cluster in particular areas of the city. Again, the general distribution of these features and their low frequency suggest that they only serve the needs of pedestrians moving through local neighborhoods and not longer-distance travel through the city.
The most notable pattern we have documented is that the size of houselots as determined by albarrada area is not statistically linked to four key variables: 1) the size of structures contained, 2) the number of structures contained, 3) proximity to the site center, or 4) proximity to the city wall. We used standard OLS-regression, as well as spatial regression using models for spatial lag and spatial error to test the relationship between houselot area, as the dependent variable, and the independent variables of residential structure area, the number of residential structures, the distance, from the city wall, and the distance from the city center. The distances from the city wall and city center produced no significant relationship. The number and area of residential structures produce weak, but significant results, with adjusted R-squared values of 0.28 and 0.29, respectively. Tests for spatial dependence indicate no apparent spatial patterning to the relationships among the variables tested. Simple explanations are thus refuted, and to the extent that albarradas might have delineated “property” of domestic groups, there is no simple correlation between amount of property and wealth or status.
Specifically, there is no reason to believe that high status families inhabiting larger structures exercised a privilege of controlling larger local and more spacious lots for infield orchards or other purposes. Nor is it plausible that density influenced houselot enclosures to be larger to accommodate greater numbers of structures generally present nearer the site center than the city wall. Finally, the idea that larger enclosures were linked to agricultural practices that would have been more feasible in the less densely inhabited areas near the city wall is also not supported. We are left with the impression that a complex array of processes formed the houselot albarrada configurations. Urban density, higher near the site center, may have contributed to fissioning of albarrada compounds, accounting for high numbers of small houselots (Figure 29). Conversely, houselots distant from the site center often have albarradas that more closely define structure clusters rather than incorporating larger expanses of domestic space (Figure 30). A pattern of ring albarradas is especially common outside the city wall, (Figure 31), in which domestic groups atop natural hill platforms have small rings of stone around the top or slopes of the hill platform itself. Many interpretations are possible. Perhaps smaller solares helped to pen in game, such as turkeys and dogs, as well as small children.
Some quantitative observations help to demonstrate this pattern (Figure 32). At milpas with the largest architecture near the city center, the solare area range is 215 to 3421.1 square meters and the mean is 1006 square meters. Solare areas of other milpas in squares within 500m from the site center range from 140 to 5238 square meters and have a mean of 1448.3 square meters. Solare areas in squares beyond 500m from the center but within the wall range from 77.2 to 3623.6 square meters and have a mean of 1101.1 square meters. Solare areas outside the city wall range from 166.4 to 2365.6 square meters and have a mean of 603.1 square meters.
We have identified two other types of albarrada enclosures, preliminarily termed pens and fields, although more work is needed to verify these functions. We hope that soil testing and excavations will assist in this effort. Two categories of small pens can be discerned, those that are self-contained, and those that use a portion of a houselot boundary wall for at least one side (Figure 33). A total of 105 pens have been identified in our milpa samples (Figure 34). Pens are found in all areas of the city as well as outside the walls.
These pens may have been used for storing agricultural products, or to contain game – we think both functions are likely. Three seasons of faunal analysis suggest that the city’s supply of whitetail and brocket deer and peccary derived in major proportions from tamed animals likely raised in Mayapán’s houselots, as along with dog, many of these animals were consumed upon reaching late adolescence or early adulthood – and their age at death profiles do not conform to expectations for a hunting pattern, although some older animals are present that may have been obtained in the wild. Turkeys were also a vital component of the city’s sustenance. While turkey and dog could have ranged free within houselot boundary walls, pens may have been necessary for deer or peccary.
Fields are defined here as inter-residential spaces that do not contain domestic structures, and our working assumption is that they could have been used for infield agriculture (Figure 35). Such spaces are under milpa cultivation today, and were likely suitable for this purpose in the past. Fields, like lanes, are in some respects defined by configurations of houselot boundary walls of domestic clusters bordering the fields. As such, it is difficult to claim that their spatial patterns were consciously planned. In many cases fields were only partly enclosed, but we identified them as fields if they were bordered on at least three sides. This map shows the locations of all fields identified in our analysis thus far (Figure 36). Fields are broadly dispersed within the walls of the site, but are slightly more common in less densely inhabited areas. Field area is highly variable, ranging from 77.2 to 3275.3 square meters with a mean of 811.3 square meters. While the fields could have been attractive for agriculture, they may have also served as larger pens for game such as deer. Generally, fields are located in areas of low artifact density. Out of 127 mapped fields, only 11 contained any artifact concentrations. Hence, there is little evidence for deliberate fertilization through the distribution of midden materials.
A couple of other categories of space at Mayapán merit mentioning. Cenotes, as Brown initially noted, are not included within individual houselots, although they are sometimes associated with albarrada walls. His suggestion that water sources were shared features by the city’s neighborhoods is supported by our data.
Some spaces at Mayapán are open, and do not appear to be finite field features. The Itzmal Ch’en vicinity provides a good example of this phenomenon (Figure 37). Such open spaces could represent additional prime spots for cultivation, or they could also have served as pedestrian walkways through the city. Entering the city from East Gate H, for example, pedestrians could have walked through open space past the Itzmal Ch’en group to the lane just south of the cenote toward destinations further in the interior. We hope to experiment with such pathways in our future analysis of bounded and unbounded space at Mayapán.
In summary, Mayapán is rich in albarrada surface features and our efforts are directed at analyzing these toward a better understanding of the organization of the city’s diverse social and economic patterns. Our preliminary assessment of the range and distribution of houselot boundary wall patterns relative to proximity to the site center and city wall – which reveals no direct link to status or urban density. We have also identified additional types of enclosures that might be linked to agriculture or animal production activities, and considered features that served as pathways through the city’s maze of architectural clusters.
Architecture Analysis. Masson and Hare are currently analyzing the distribution of domestic structure types and group types across the city, using our 36 milpa maps. We had originally planned to do this using the Carnegie map (Jones 1962), which is arguably one of the best site maps ever created for a Maya city, but we now know that it is not suitable for detailed analyses of structure or group types. Hare’s comparisons of the accuracy of the Carnegie map to our own mapping data in the field reveal that the Carnegie map is not a consistently reliable source for such variables as: structure size, structure orientation, structure location, or the number and types of structures within a group. We have recorded the following attributes for each structure in our milpa database: building type and dwelling type, number of rooms, number of houses in group, number of structures in group, patio configuration type (L- shaped, enclosed, 3-sided, isolated residence, etc.), group cardinal orientation, number of benches, shape of benches, and status (based on size/elaboration) and analyzed these distributions across the site (Figure 38). Some preliminary conclusions include the following:
1) other types of commoner houses exist at Mayapán beyond the typical
one or two room rectangular house with benches defined by the Carnegie –
rectangular alignments with lateral or longitudinal room divisions and square,
elevated platforms are also likely residences based on analogous size and context
to that of “typical” houses (40 such structures were identified,
e.g., Figure
39),
2) specific “types” of elite houses can be identified across the
city (Figure
40),
3) large platforms may represent house forms of specific foreign ethnic groups
known to have resided at the city,
4) house group orientation of a given neighborhood tends to face the most prominent
elite/administrative nearby feature – such focal points within the city
tend to be the site’s monumental center and the outlying Itzmal Ch’en
group,
5) typical commoner (N-158) houses are located in all areas of the site, and
while they predominate in Mayapán milpas that are furthest from the site
center – they are also abundant in upper status inner city neighborhoods
where their residents likely performed supporting services,
6) square/rectangular and L-shaped benches (or combinations of these in a single
house) do not have completely distinctive distributions, however, frequencies
vary and there might have been some stylistic preference that is evident –
the square/rectangular form is more common and some lower status neighborhoods
exhibit these exclusively, perhaps adhering to a more traditional pattern,
7) domestic shrines are rare in Mayapán houselots – either internal
house shrines or external patio shrines – this observation contradicts
earlier assumptions that “household religious practice” was commonplace
at the city,
8) idiosyncratic house form or solare form variation is observed in specific
milpas (Figure
41) – Milpa 12 has several large square platform houses that are unique
for the city, Milpa 29 has several small altillo albarradas with separate, adjoined
field spaces formed by albarrada walls that is also unique at Mayapán,
Milpas 18 and 34 have many more sascaberas than is common in other parts of
the city and Milpa 34 has a large quantity of small external benches that are
not linked to a domestic houselot.
Analysis of these features is ongoing, these observations will be refined and disseminated throughout the 2006 cycle of publications. The next step in our analysis is to analyze artifact distributions by solare (albarrada houselot enclosure). All databases have been coded with solare classifications, based on field observations and followed up by detailed visual coding of feature clusters from our field maps.
Return to The Economic Foundations of Mayapán
Table of Contents
Return to the Main Page
© 2006 Institute for Mesoamerican Studies
Updated February 8, 2006
