A Pueblo View of Abundance

We begin by exploring the concepts of abundance and scarcity from a modern but historically informed Pueblo perspective. For this we draw upon Naranjo’s knowledge as a Tewa¹ elder from Santa Clara Pueblo. His Pueblo perspective views these concepts—abundance and scarcity—as two parts of a greater whole: one cannot exist without the other, and they occur in cycles. This is similar to the observation made by Smith in her review of abundance as an economic principle and archaeological concept: she views the perception of something as scarce or abundant as situational and relative, with one defined in part by reference to the other (Smith 2012, this volume).

We believe the perspective that follows would apply broadly to all Pueblo groups, but Tito cautions that he is speaking specifically about the Tewa-Pueblo worldview he knows best. In Tito’s view, the concepts of scarcity and abundance are best understood in reference to deeper, more fundamental aspects of ideation that include Pueblo philosophical concepts and values. Perhaps the most central tenet of Tewa philosophy is a concept called “seeking life,” and abundance is best understood with reference to this concept. Seeking life refers to the Tewa belief that a life force is present in all aspects of creation—including matter that a scientific perspective would classify as both animate and inanimate—and that all of creation is interconnected through this life force. The Tewa word for this life force is p’o wa ha, which literally translates to “water, wind, breath.”

When Tewa people speak of seeking life, they are referring to seeking a connection to this life force, and they view this as the most important goal of one’s life. Tewa people use a variety of expressions to speak of searching for and finding the life force, with woatsi tuenji referring to seeking life, woatsi shaa as finding life, and woatsi maegi as giving life. According to Tewa understanding, the concepts of an interconnected life force and seeking life developed during the deep Pueblo past, including their time in the Mesa Verde region of southwestern Colorado (AD 600–1285).

A connection to the life force can only be achieved by living one’s life according to Tewa values. There are many core Tewa values. They include the importance of community, respecting elders and their knowledge of traditions, nurturing children, hard work, and displaying generosity by sharing with others. Tito understands abundance in terms of this formula: living one’s life in accordance with Tewa values brings one into contact with the life force, and contact with the life force ensures abundance. In this way, abundance is not a standalone concept in Tewa thought but is instead an integral component of Tewa values and connecting with the life force that underlies creation. In this way, for Tewa people, abundance is not only material but also spiritual.

Agricultural Productivity: Cycles of Scarcity and Abundance

Economists and other social scientists—especially those concerned with issues of social justice and sustainability—have criticized the central role scarcity plays in modern economic theory (Dugger and Peach 2009; Hoeschele 2010). They point out that rather than intrinsic to economic systems—especially capitalist economies—scarcity is often socially constructed to the advantage of some and the detriment of others. Socially constructed scarcity is an important aspect of theorizing economic systems, including pre-capitalist economies, but as Smith (this volume) notes, not all scarcity and abundance is socially constructed. True scarcity and true abundance exist and largely result from the inherent qualities of particular phenomena. We believe the inherent properties of ancestral Pueblo agricultural production did result in true scarcity and abundance, and so we begin our attempt to measure scarcity and abundance in Pueblo Indian society by estimating the agricultural productivity in the Mesa Verde region.

Despite the fact that Mesa Verde Pueblo people farmed a variety of crops, hunted wild animals, and collected wild plants (Adams and Bowyer 2002; Driver 2002), maize farming was by far the most important component of their subsistence economy. Bone chemistry studies that examine stable carbon and nitrogen isotopes have been used to reconstruct the reliance on maize and consumption of animal protein; these studies indicate that ancestral Pueblo people got at least 70 percent of their calories from maize (Chisholm and Matson 1994; Coltrain and Janetski 2013:4713; Coltrain, Janetski, and Carlyle 2006, 2007:314; Matson and Chisholm 1991:452–56). The results of these bone chemistry analyses are supported by analyses of coprolites, pollen, and plant macrofossils (Aasen 1984; Geib 2011:225–29; Lepofsky 1986) and by the analysis of groundstone (Hard, Mauldin, and Raymond 1996). Pueblo people relied on maize for the vast majority of their calories from the time of their first appearance in the Four Corners region at about 500 BC until they migrated from the area at the end of the thirteenth century.

Focusing on maize farming as the starting point for our examination of scarcity and abundance makes even more sense when we consider the minimum precipitation and growing season requirements for Pueblo varieties of maize. A minimum threshold for precipitation is 30 cm, or about 12 inches, per year (Benson 2011a, 2011b; Benson and Berry 2009). By comparison, the average precipitation during the historic period for the town of Cortez, Colorado, which is in the center of our Mesa Verde region study area, is 12.6 inches, or 32 cm (Shaw, Sprague, and Dudley 1988). Pueblo maize also has a minimum temperature requirement; in technical studies this is measured as growing degree days (GDD), and the minimum threshold is 1,800°F GDD (Bellorado 2007; Benson and Berry 2009). Although not precisely equivalent to GDD, it is easier to understand the temperature requirement for maize when it is expressed as the length of the frost-free growing season. On average, varieties of maize indigenous to the greater Southwest require a 128-day growing season, with a range of 111 to 144 days (Adams et al. 2006:47–48). This compares to 125 days for the average growing season for Cortez. These data indicate that average conditions in our Mesa Verde study area were just above the minimum thresholds for growing maize, and annual fluctuations in precipitation and temperature could therefore result in either crop failure or abundant harvests. It is possible that socially constructed dimensions of maize production also affected the way Pueblo people obtained this crop, which was critical to their survival, but any socially constructed conditions would have derived from the yields determined by these inherent properties of maize farming. An understanding of these inherent properties and their effects on agricultural production provides an important foundation for examining scarcity and abundance in ancestral Pueblo society.

The research program known as the Village Ecodynamics Project uses computer modeling to estimate agricultural productivity in the Mesa Verde region (Kohler and Varien 2012). The VEP methods for modeling maize-farming yields are likely the most rigorous yet developed for reconstructing agricultural production in any ancient society, and they are broken down into steps and reported in detail (Kohler 2012:86–108; Varien et al. in press). The VEP model begins by using tree-ring analysis to get precipitation and temperature estimates for each year between AD 600 and the present. Then the study area is divided into 4-hectare cells—there are over 110,000 of these—and the elevation and the characteristics of the soils in each cell are measured, including the moisture-holding capacity of the soils. Precipitation and soil properties are combined to calculate the accumulated soil moisture for each cell during each year (Palmer 1965). The statistical relationship between this measure and crop yields is calculated for the historic period and subsequently retrodicted back to AD 600. These results were systematically modified to take into account changes in technology, differences between historic and indigenous maize varieties and planting strategies, the effects of elevation on production, and the possible effects of soil nutrient depletion. The result is an estimate of maize yields for every cell in the VEP study area for every year between AD 600 and 1300.

Figure 4.2 presents the results of the VEP analysis, showing annual variation in maize yields standardized as z scores. The horizontal black line at zero is the 700-year average production; the lighter, dashed horizontal black lines are 1 standard deviation above and below this long-term mean. The vertical black bars show individual years that are within 1 standard deviation of the mean, or years when yields were closest to the long-term average. These conditions prevailed during 493 years, or 70 percent of the time, of the 700-year Pueblo occupation of the region. These likely represent years when Pueblo farmers were able to obtain harvests, although figure 4.2 shows that yields would have fluctuated considerably even during those years. There were 230 years when the yields were above the long-term mean and 263 years when they were below average. Even during the times that were closest to the long-term average, Pueblo people would have experienced a constant swing between relatively abundant and relatively scarce yields.

Figure 4.2. Annual estimates of maize productivity for the VEP II study area, AD 600–1300 (standardized to z scores). Bars more than 1 standard deviation above the mean are unusually wet years with high productivity and more than 1 standard deviation below the mean are unusually dry years with low productivity. Courtesy, Crow Canyon Archaeological Center, Cortez, CO.

The gray vertical bars in figure 4.2 show years that are more than 1 standard deviation above or below the mean. Years below this 1 standard deviation threshold are times of exceptionally low precipitation, extreme cold, or both. These conditions produced extremely low yields or in some cases complete crop failure. There are 104 years—about 1 year in 7—when Pueblo farmers would have experienced extreme scarcity in food production. There are also 140 extremely wet years with adequate temperature and sufficient growing seasons that produced yields 1 standard deviation above the mean. These would have been years with unusually abundant harvests. These data illustrate that unusually abundant and scarce harvests occurred relatively regularly and in the same frequency, reinforcing Tito’s observation that a constant oscillation between abundance and scarcity did in fact characterize the lives of Pueblo people.

Accumulating Abundance: Pueblo Pottery Assemblages

Next, we examine abundance by focusing on the accumulation of artifacts. As Smith (2012, this volume) notes, as a measure of abundance, artifact accumulation includes per capita increases in the same type of artifact or increases in the diversity of objects. We investigate this form of abundance by examining the pottery assemblages of Mesa Verde Pueblo households and begin by focusing on the accumulation of bowls. We measure how the accumulation of bowls changed through time at thirty-two tree-ring-dated sites from the VEP study area.

A brief history of pottery use in the study area provides the basic context for this analysis. The earliest use of pottery by Pueblo groups in the northern Southwest dates to about AD 300 when a plain utilitarian ware, usually brownish in color, first appeared (Geib 2011:280). These early vessels were jars used primarily for cooking or storage. Pottery sherds from these early brown ware vessels were few in number, and in fact they were not present at many excavated habitation sites dating between AD 300 and the late AD 500s. Brown wares were largely replaced by gray wares and white wares in the late AD 500s, and pottery sherds became much more numerous and uniformly present at habitation sites by about AD 600, when the VEP study area was first settled by large numbers of Pueblo farm families.

There is a clear functional distinction between gray and white wares beginning with the first appearance of this pottery. Early gray wares were mostly cooking jars (Blinman 1988:458, 1993:18) but also included some storage vessels; gray wares become increasingly specialized as cooking pots over time. In contrast, most early white ware vessels were storage jars (Blinman 1993), but for the first time bowls used as serving vessels also occurred in low but consistent numbers at habitation sites. Red ware pottery that primarily included storage jars and bowls first appeared at about AD 780. Red ware vessels were produced in areas outside our Mesa Verde study area but were traded into southwestern Colorado in relatively small numbers and appeared at sites dating between AD 780 and 1300. In the analysis that follows, we focus on gray ware cooking jars, bowls that are predominantly white wares, and red ware bowls when present.

We measure changes in the abundance of bowls by calibrating them against the accumulation of cooking pot sherds. Elsewhere we have demonstrated that cooking pot sherds accumulated at a relatively constant rate (Varien 1999:73–80; Varien and Mills 1997; Varien and Ortman 2005; Varien and Potter 1997). The accumulation rate for cooking pots is largely a result of the regular use of these vessels, and the relationship among use, breakage, and uselife of cooking pots has been documented in experimental, archaeological, and ethnoarchaeological studies (Bronitsky 1986; Bronitsky and Hamer 1986; Nelson 1991; Pierce 1999, 2005; Rice 1987; Rye 1976; Schiffer et al. 1994; Steponaitis 1983, 1984; Tani 1994; Varien and Mills 1997; West 1992). The regular accumulation rate of cooking pots is further demonstrated by examining the relationship between cooking pot sherds and chipped stone debris from screened assemblages at sites dating between AD 600 and 1280 (figure 4.3). Even though their use, discard, and rate of accumulation were governed by entirely different behaviors, Pueblo people discarded chipped stone debris and broken cooking pots as a result of daily domestic activities. The strong correlation seen here could only result if these functionally unrelated artifact types accumulated at regular rates through time. The fact that we have a good understanding of the use, breakage, and discard of cooking pottery and can demonstrate that this pottery accumulated at a constant rate through time means that it can be used to examine how the accumulation rate of other artifact categories changed (or did not change) through time.

Figure 4.3. Scatterplot showing the relationship between discard of cooking pot sherds and chipped stone debris from screened assemblages at forty-five sites dating between AD 600 and 1300 in the central Mesa Verde region VEP II study area (Rsq 0.97). Sites were excavated by the Crow Canyon Archaeological Center or the Dolores Archaeological Program. Courtesy, Crow Canyon Archaeological Center, Cortez, CO.

Here, we used the constant accumulation rate of cooking jars to identify changes in the accumulation of pottery bowls. We examined the relationship between the accumulation of pottery sherds from cooking pots and bowls at thirty-two sites dating between AD 600 and 1280. These sites are located in the VEP study area and were selected because they are precisely dated with tree-ring analysis and they have screened deposits. This analysis shows that the accumulation of bowls increased exponentially over time, with an average growth rate of about 0.6 percent per year (figure 4.4). Clearly, household bowl inventories increased through time, and by the end of the sequence households had many more bowls in their pottery inventories, which serves as a direct measure of increasing abundance in the lives of Mesa Verde Pueblo people.

Figure 4.4. Ratio of bowl-to-cooking jar pottery sherds from screened assemblages at thirty-two tree-ring–dated sites in the central Mesa Verde region VEP II study area (ratio calculated using sherd weights). Courtesy, Crow Canyon Archaeological Center, Cortez, CO.

Rim-arc measurements of bowls can be used to estimate the volumes of these vessels, and this analysis shows that two distinct sizes of bowls developed in the AD 1200s in the central Mesa Verde region (Ortman 2000). The volume of small bowls is about a liter, while large bowls hold about 5 liters (Ortman 2000:paragraph 45, figure 1). Small bowls were likely used by individuals to consume food, while large bowls were used to serve food to individuals; the difference in their size suggests that household groups numbered about five people. Several lines of data indicate that these bowls were produced in most, if not all, households by non-specialist potters (Pierce et al. 1999). A similar analysis also shows that household pottery assemblages came to include different-size cooking jars through time (Ortman 2000:paragraph 46, figure 2). These data show a more continuous distribution of vessel sizes, so rather than producing distinct size categories, potters in Mesa Verde households produced a range of vessels of different sizes to meet their needs.

The development of distinct size classes in both bowls and cooking jars is an example of household pottery assemblages becoming more diverse through time, and this diversity is another measure of increasing abundance in Pueblo society. Although few in number relative to bowls, households also added new forms of decorated white ware vessels through time, for example, ladles, mugs, and specialized seed jars known as kiva jars that also occur in different sizes. Here again, greater diversity is a measure of increasing abundance through time. By the AD 1200s, Pueblo families had numerous and diverse household inventories. This is illustrated in figure 4.5, which shows the vessels left behind when a house at Sand Canyon Pueblo was abandoned at about AD 1285.

Figure 4.5. Whole and partial pottery vessels left behind by the household that occupied Block 500 at Sand Canyon Pueblo. Photo by David Grimes. Courtesy, Crow Canyon Archaeological Center, Cortez, CO.

Ortman examined the frequencies of vessel size categories at both small and large sites that date to the AD 1200s. His analysis shows that large vessels—both cooking pots and bowls—occur in greater numbers at villages and are fewer in number at small farmsteads (Ortman 2000:paragraphs 43–58). He argues this may be a result of the fact that specific activities took place in villages that did not occur at smaller sites. This includes the use of larger vessels during feasts that accompanied communal ceremonies conducted at the large villages. We elaborate on this pattern in the section that follows.

Performing Abundance

The use of food and ritual to create relationships beyond the family—what we refer to as feasting—is well documented cross-culturally in small-scale societies (Dietler and Hayden 2001; Goody 1982; Gremillion 2011; Monaghan 1990; Weissner and Schiefenhovel 1996). In the Southwest, feasting is understood as one of the primary mechanisms whereby small-scale agriculturalists of the past increased the social, demographic, and political scale of their societies (Blinman 1989; Ford 1972; Graves and Spielmann 2000; Phillips and Sebastian 2004; Potter 2000; Potter and Ortman 2004; Spielmann 2002, 2004; Wills and Crown 2004). In contrast to the majority of small-scale societies around the world, though, communal feasting in Pueblo contexts was less overtly political and less lavish, and for the most part it involved the same suite of artifacts and food resources generally used for the preparation of domestic-style meals rather than rare or special items and foods. This has prompted Potter and Ortman (2004:175) to suggest that communal feasts at Pueblos are a metaphorical extension of the domestic meal. In this way, Pueblo feasts differ from many other feasting contexts around the world in which the ideological significance of communal feasting derives from its articulation with more authoritarian and sometimes masculine realms, for example, warfare, competitive politics, and hunting (de Garine 1996; Dietler 1996; Hayden 1995, 2001; Weissner 2001; Young 1971).

Another unique aspect of Pueblo feasts is that they are generally financed by the participating families rather than a single “big man” or political leader (Potter 2000). Thus, feasts have been described more as potluck than potlatch events (Blinman 1989). Food is prepared in homes and then brought to the communal gathering place, often a large structure or plaza, and shared among the participants. Households that have fared better provide greater quantities of food, and this is distributed—often by masked and anonymous katchinas—to less fortunate families. These events ensure that household scarcity and abundance are evened out (to the extent that they can be) and that household abundance is translated to communal abundance or at least sufficiency. As discussed above, food resources in the Southwest are episodically scarce when annual precipitation was too low or the growing season too short to produce maize. Hunted game, particularly deer, became scarce as a result of sedentism, population growth, and resultant overhunting (Driver 2002). Yet even during times of scarcity, the performance of the feast and associated ceremony celebrated and acknowledged abundance through the active display and exchange of food and gifts among community members.

Evidence indicates that these performances occurred at villages (i.e., settlements containing nine or more aggregated households [Wilshusen and Potter 2010]) rather than at smaller sites such as single-family farmsteads or small hamlets. This is exemplified by four cases presented below—the Dillard site, Sacred Ridge, McPhee Village, and Sand Canyon Pueblo—all of which are villages surrounded by smaller contemporaneous settlements. In addition to being associated with larger settlements, feasting is also typically associated with specific types of buildings and open space at these villages: great kivas, oversized pit structures, multi-wall structures, and plazas.

We examine the evidence for ritual and feasting at several sites in our study area at which this practice has been documented in association with these various types of architecture: the Dillard site during the AD 600s, Sacred Ridge during the late 700s and early 800s, McPhee Village in the late 800s, and Sand Canyon Pueblo in the late 1200s. Our study shows that the scale of these rituals varied greatly but in general increased through time, which may be considered another aspect of increasing abundance in Pueblo communities.

The Dillard Site

The earliest evidence for public ceremony and communal feasting in the Mesa Verde region appears at the Dillard site, a village of the Basketmaker III period located just west of Cortez, Colorado, and dating to the AD 600s. Investigations at this site are ongoing by the Crow Canyon Archaeological Center. The site is centered on a great kiva that is the earliest and only known great kiva in the Mesa Verde region. The Dillard site is unusual because most of the sites dating to this time period are single-household farmsteads; Dillard, by contrast, is a village with at least a dozen pit structures concentrated in two clusters, one north and one south of the great kiva. Kari Schleher and others (2013) have documented relatively high proportions of serving bowl sherds in association with the great kiva at this site, presumably because of the communal serving and consumption of food during ceremonies performed in and around this structure. Data on the types of foods served have not yet been recovered, but presumably stews and boiled foods prepared in cooking pots and in domestic contexts were served in these bowls.

The great kiva is a very large communal structure that served the ritual needs of the village and perhaps a larger, more dispersed Basketmaker III population. This marks the initiation of a long tradition of the use of great kivas as communal structures and a context for feasting, one that began in the northern Southwest during the late AD 500s or early 600s and continued for the next 700 years. Although some archaeologists argue that great kivas disappeared from the Pueblo record at about AD 1300, we view great kivas as the best analogy for the kivas that continue to be used in Pueblo communities today.

Sacred Ridge

Located on a tributary of the Animas River, the early Pueblo I (AD 700–825) settlement Sacred Ridge occupied a large knoll at the west end of Ridges Basin. The village contained twenty-two pit structures, dozens of burials, and a complex of ritual features at the apex of the knoll consisting of a two-story surface structure, a large circular surface room, and a palisade enclosure. Five of the pit structures were unusually large, or “oversized,” and contained features similar to those found in smaller pit structures, such as a hearth and a wing wall. But they also contained additional ritual floor features and immense wraparound benches. Ritual floor features include small circular holes analogous to sipapus directly behind the central hearth and large conical pits offset from the hearth, all of which were plastered and, when closed, filled with clean sand and capped with a layer of clay. Averaging 8 m in diameter (50 m² in roofed area), these oversized pit structures were two to five times the size of the smaller, household-level pit structures. Feasting has been argued to have occurred in these structures based on their large capacity, the disproportionately high numbers of serving bowl sherds, and the relatively high frequencies of artiodactyl (deer) and lagomorph (rabbit) remains associated with them (Allison 2010; Potter and Edwards 2008). Thus, stews and roasted game appear to have been served and consumed in these contexts.

Sacred Ridge was one of a number of contemporaneously occupied early Pueblo I sites in Ridges Basin, thirty-four of which were excavated as part of the Animas–La Plata Reservoir project from 2002 to 2005 (Potter 2010). These sites were much smaller than Sacred Ridge, consisting of one to four pit structures. As indicated, though, Sacred Ridge was not only larger than these sites, but it also contained (1) the only communal architecture in the community (in the form of oversized pit structures); (2) the only architectural complex comprising unique architectural forms, such as a two-story tower and a circular surface room, all enclosed by a palisade; (3) the only evidence of feasting in the community; and (4) the highest relative frequency of serving bowl sherds and artiodactyl remains in the community. It is doubtful that feasts at Sacred Ridge were truly communal in scale and as inclusive as those associated with great kivas; this interpretation is based on the large size of the community (i.e., over fifty households at peak population levels), the comparatively small size of oversized pit structures, and the direct association of the oversized pit structures with particular households.

McPhee Village

McPhee Village was an aggregated community in the Dolores River Valley that was occupied during AD 850–900; an earlier component exists but is not part of an aggregated community (Kane and Robinson 1988). It was composed of twenty-one spatially discrete but contemporaneous roomblocks, each with associated pit structures. Like Sacred Ridge, the site lacked a great kiva but contained a number of oversized pit structures that have been interpreted as communal ritual structures. Three oversized pit structures associated with three different roomblocks (5MT4475, 5MT4477, and 5MT5107) in this community had roofed areas greater than 30 m² (67 m², 64 m², and 37 m², respectively) and contained an array of floor features comparable to historic “chief kivas” (Mindeleff 1891:134), including complex sipapus, lateral vaults, sand-filled pits as sockets to support altars, possible prayer-stick holes in structure floors, and floor vaults with wooden covers interpreted as foot drums (Wilshusen 1989). These structures differ markedly from those at Sacred Ridge in the number and complexity of floor features associated with them that were likely related to the performance of ritual.

Another distinction at McPhee is that the oversized pit structures are associated with U-shaped roomblocks. The construction of U-shaped roomblocks created small, semi-enclosed spaces that enveloped the oversized pit structures and formed some of the earliest known architecturally bounded plazas. In at least one case, a wall closed off the open end of the surrounding U-shaped roomblock, restricting access to the oversized pit structure and surrounding plaza to an even greater degree (Brisbin, Kane, and Morris 1988:234–35). This suggests that feasts held in these plazas and the rituals conducted in these structures were even more exclusive than were the feasts at Sacred Ridge.

Eric Blinman (1989) has demonstrated a significant correlation at McPhee Village between roomblocks associated with oversized pit structures and high relative frequencies of red ware bowls, which were imported into the area during the ninth century AD. He suggests that the communal ceremonial behaviors associated with these structures account for the inflated frequencies of red ware bowls, both because of their value as imported commodities and because of their use in serving and consuming food during feasts. Likewise, James Potter (1997) noted the high proportion of rabbit and deer remains associated with oversized pit structures within the community, as well as inordinate frequencies of wild and carnivore remains, arguing that these patterns are a result of the performance of communal feasting and ritual. This is supported by Timothy Kohler and Charles Reed’s (2011) analysis that indicates that McPhee Village and other large settlements in the Dolores River Valley emphasized deer hunting and located their villages to ensure successful hunts. Again, data suggest that roasted game, in addition to stewed and boiled foods, were served and consumed in these contexts and that participation (both physical and visible) was relatively restricted within the community.

Sand Canyon Pueblo

The late Pueblo III site Sand Canyon Pueblo dates from about AD 1250 to 1285; with 400 to 600 residents, it was one of the largest villages inhabited during the final decades of Pueblo occupation in the Mesa Verde region (Kuckelman 2007). Prior inhabitants in the area had reduced local game populations; the developed sociopolitical landscape also made it difficult for people to range widely in search of game (Driver 1993, 2002; Schollmeyer and Driver 2011, 2013; Varien 1999). Intensified ritual and ceremonial feasting at the site is suggested by the presence of a variety of communal architectural spaces: a plaza, an unroofed great kiva, and a D-shaped bi-wall structure.

This architectural evidence is complemented by the differences in the sizes and decoration of cooking pots and serving bowls discussed above (Ortman 2000; Ortman and Bradley 2002). A higher proportion of very large cooking pots was deposited at Sand Canyon Pueblo than at earlier and contemporaneous smaller sites around the village. These vessels are larger than needed for family meals, and Scott Ortman and Bruce Bradley (2002) argue that the larger cooking pots were used to prepare food for communal meals and that the meals were consumed in public view around the plaza and in the great kiva. Large serving bowls were also present in greater numbers than those found at smaller sites. In addition, a higher percentage of the larger serving vessels at Sand Canyon were decorated on their exteriors (Ortman 2000:paragraphs 59–61). The exterior decoration on these vessels has been interpreted as evidence of intentional display of social identity in public contexts, as these intricate decorations would have been visible even when the bowls were full of food (Ortman and Bradley 2002:68; Robinson 2005). This strongly suggests that communicating social identity was important in these contexts and that these events were likely multi-community in scale.

The depleted nature of the local environment, a high frequency of grinding implements, and a low proportion of faunal remains, especially deer, compared to Pueblo I contexts (Potter and Ortman 2004:table 10.1) all suggest that cuisine at Sand Canyon Pueblo was dominated by agricultural products such as corn and beans rather than by hunted game. By extension, serving and consuming prepared agriculturally based cooked food was likely emphasized in communal feasting contexts, and roasted and stewed fauna deemphasized. Moreover, the high proportion of cooking pots and serving bowls (particularly compared to Pueblo I villages) indicates that boiling was the preferred preparation method (rather than roasting, for example).

Feasting and Abundance: A Summary

Communal ritual accompanied by food sharing is a practice that extends deep into the Pueblo past. By sharing food surpluses at public ceremonies, community members performed and experienced abundance, perhaps even in the face of production shortfalls and scarcity. This performed abundance provided food to needy families, but these rituals also produced long-term social relations of reciprocity, debt, and obligation.

The scale of these events, and the network of social relations facilitated by them, varied but tended to increase through time. Two very different communal ritual traditions operated in the northern Southwest, one that was truly communal—the great kiva tradition—and one that was restricted to certain households within the community—the oversized pit structure tradition. The great kiva was the longest-lived tradition. It began in the Basketmaker III period at the Dillard site and other early villages (e.g., Cummings 1953; Gilpin and Benallie 2000; Haury 1928; Roberts 1929; Wills and Windes 1989) and continued through the sequence until 1285, as exemplified by Sand Canyon Pueblo. Even during the Pueblo I period, when some villages adopted the oversized pit structure ritual tradition, great kivas were still in use in some communities. A great kiva at Grass Mesa Village in the Dolores River Valley, for example, was in use between AD 760 and 810, the same time as the oversized pit structures at Sacred Ridge (Lightfoot 1988; Lightfoot, Emerson, and Blinman 1988). This great kiva measured 22.6 m in diameter and had a floor area of approximately 400 m², making it 16 times the size of the contemporaneous structures that surrounded it.

By contrast, the oversized pit structure tradition was relatively short-lived. There are examples of oversized Basketmaker III pit houses (Hurst 2011; Lux-Harriman 1982), but it is not clear if they are analogous to the later oversized pit structures discussed above, with these examples limited to the period AD 750–900. In addition, communities with oversized pit structures were short-lived and volatile (Wilshusen and Potter 2010). Occupation at Sacred Ridge, for example, ended in a community-wide massacre involving at least thirty-five victims (Potter and Chuipka 2010). We suggest that this was in part the result of a contradiction between the communal practice of potluck feasts that occurred in the plazas and open spaces at these villages and the more exclusive ritual practices that occurred in the oversized pit structures. The smaller, more exclusive structures may have been less effective at integrating culturally diverse village populations than were the more communal and inclusive great kiva feasts. Not only do great kivas (and the rituals associated with them) persist after 900, but more inclusive and visible contexts for ritual performance—the plaza and the unroofed great kiva—become more common as well.

We also see a shift from primarily hunted resources being shared at feasts in earlier villages to agricultural products in later contexts (Kohler and Reed 2011; Potter 1997; Potter 2000). Unlike hunted meat, corn and beans can be stored for long periods of time. Thus, the timing of corn-based feasts is not dependent on the timing of resource acquisition—these feasts can occur, for example, during times of the year in which household stores are low and consequently can facilitate the redistribution of food surpluses across households, as Richard Ford (1972) has documented for the historic Tewa. Maize-based feasts would not have been as dependent on the timing of resource acquisition and could have occurred during times of the year when individual household stores were low, redistributing food across households.

But storable foods can also encourage surplus accumulation among households or communal storage of surpluses (Gremillion 2011; Twiss and Bogaard, this volume). Surplus accumulation is one of the principal factors cited in evolutionary models of the development of political leadership and social inequality in middle-range societies (Blitz 1993; Feinman 2000; Smith 2012:30–31), and it appears that surplus accumulation was an important dimension of the social dynamic at Sand Canyon Pueblo (Lipe 2002). One of the buildings in the village appears to have been a communal storage facility located in the plaza, and adjacent to this structure was a D-shaped building that appears to have functioned as both a high-status residence and a setting for secretive ritual (Lipe 2002:225; Ortman and Bradley 2002:67). This raises the possibility that community surpluses were controlled by the community’s political leadership. Community leaders at Sand Canyon Pueblo appear to have had greater access to larger quantities of stored food resources such as corn, and redistributing those resources in the context of public rituals would have solidified their position as leaders—a scenario more in line with the potlatch model of feasting than the potluck model.

As noted, large bowls of a standardized size were added to the pottery assemblage in the thirteenth century, and at late villages with public architecture a greater percentage of these bowls have exterior decoration, serving as statements of identity in public contexts. The need to convey identity in this manner suggests that people from beyond the immediate community attended these events; this would further extend the network of social relations at play during communal rituals at these late villages. Thus, feasts grew in both scale and political import as surpluses—the consequences of periodic abundance—were publicly shared. The shift to corn-based feasts not only allowed for increased abundance but also facilitated greater wealth distinctions among households and opportunities to display that variation in a public context.

Conclusion

It seems clear that ancestral Pueblo people understood and sought out abundance. This is indicated by years when agricultural harvests were unusually bountiful, by the increasing accumulation of pottery bowls and other vessel forms through time, and by communal ceremony and feasts where abundance was performed. We have also shown how these three aspects of abundance were woven together: harvested maize was cooked and consumed in pottery vessels, bountiful harvests were the basis for ceremonial feasts, and special oversized cooking pots and large bowls with exterior decoration were manufactured for use in these feasting events. But this expression of Pueblo abundance was framed by an equally intimate understanding of scarcity, structured in part by a heavy dependence on maize agriculture in circumstances where episodic drought and cold temperatures resulted in years when maize harvests were extremely scarce or entirely absent (Bocinsky 2014; Bocinsky and Kohler 2014).

Tito notes that there is no Tewa word for feast or feasting; however, there is a Tewa word for starvation: haa’chu, which translates as “life-breath death.” For Tito, this indicates that Tewa people have a profound understanding of scarcity, one derived from direct experience and a social memory rooted deep in the Pueblo past. Tito views the dual Pueblo understanding of abundance and scarcity as central to modern Tewa public ceremonies in which community members and costumed dancers exchange food. This performed ritual of abundance reinforces the fundamental Tewa values of community and sharing and the establishment of an extended web of social relations—one’s maa tú—upon which Tewa people depend. Ford (2014) translates maa tú literally as “next of kin,” but he notes that this can extend beyond one’s biological relatives who are reckoned bilaterally to include a wider network of relationships. Similarly, Tito views maa tú as referring to both one’s blood relatives and the larger social network created by events such as the participation in performed ceremonies of abundance noted above. Ford (2014) notes that one’s maa tú is named, and individuals use this name to connect with individuals in distant villages who are part of a similarly named maa tú, establishing fictive kin relations with socially distant people.

Pueblo communal feasts and ceremonies, as social and performed expressions of abundance, differ from similar practices in small-scale societies elsewhere in the world. It seems likely that these practices were connected to the development of social inequality in ancestral Pueblo society, but here, too, the Pueblo case seems more nuanced when compared with the more straightforward relationship among feasts, ceremony, and inequality in societies elsewhere in the world. This is especially true when we compare the evidence for these practices in ancestral Pueblo societies with the relationship among feasting, reciprocity, ritual, and inequality in modern Pueblo societies. There appears to be a transformation in Pueblo society that occurred during the late thirteenth and fourteenth centuries AD, the time when vast areas of the ancestral Pueblo world, including those in the Mesa Verde region, were depopulated and the modern Pueblos began to take shape. This transformation produced a modern Pueblo world in which this relationship among feasting, ceremony, and inequality is even more nuanced and muted. For modern Pueblo people this nexus of practices, so interesting from an anthropological perspective, is less important than the relationship among the fundamental concept of seeking life, the values of community and reciprocity, and the ability to gain a life of abundance.