The Marana Community in Comparative Context

Suzanne K. Fish and Paul R. Fish

The foregoing chapters have summarized the present state of knowledge concerning one Hohokam community among the many distributed across the central portion of southern Arizona during the Classic period (Fig. 9.1). Description and interpretation of the Marana Community has been shaped by survey results and settlement pattern analysis, based on the most exhaustive data set of this kind yet obtained. The very uniqueness of these findings makes it difficult to place the Marana Community in a comparative context. Such perspective is necessary, however, for assessing the significance of Marana among like phenomena in time and space for formulating the most meaningful questions for future research.

Much previous study in Hohokam archaeology has been focused at investigative scales bracketing the community: excavation of individual sites on the one hand and supracommunity distributions of elements such as public architecture on the other. Thus, in order to view developments in the northern Tucson Basin against a backdrop of community variability within the Hohokam tradition, measures for relevant attributes must be assembled and synthesized. The majority of available information pertains to the irrigated valleys of the Salt and Gila rivers, where observations regarding the locations of settlements first gave rise to the concept of the Hohokam community.

Hohokam communities embody institutions that integrate both dispersed and concentrated settlement into bounded territorial units. A seminal paper by David Doyel (1974; see also 1980:31) defined irrigation communities as interrelated sites along a shared canal line, including smaller sites and at least one large site with ceremonial architecture. Wilcox (Wilcox and Sternberg 1983:195) emphasized the focal and integrative function of central sites with public architecture within their respective communities. Due to zones of dispersed but continuous settlement, relatively noncompact centers, and the critical role of intersite relationships, physical aspects of Hohokam territorial organization cannot be understood by studying individual sites. Structural changes in Hohokam society are also incompletely expressed at this scale of analysis. Both issues can be fully comprehended only at the level of the multisite community.

Scattered instances of large sites with ballcourts and a few examples of mound construction are known from the early Preclassic period by at least A.D. 750. Relatively standardized organizational modes focused on such centers with public architecture can be recognized by A.D. 1000, when sufficient data reveal regularized spacing of ballcourts in the Phoenix area and numerous occurrences elsewhere (Wilcox and Sternberg 1983). By that time, it appears that the multisite community had become the principle territorial and organizational unit of Hohokam occupations. In the early Classic period after about A.D. 1150, use of ballcourts ceased. Platform mounds, usually constructed in stages, were located in community centers during this time and continued into the late Classic period as the principle form of public architecture (P. Fish 1989; Wasley 1980; Wallace and Holmlund 1984; Wilcox and Sternberg 1983).


Community Spacing

Because sites with public architecture mark replicas of integrative nodes, community extents can be approximated by boundaries between consecutive centers of similar sate. In the area of large-scale irrigation in the Hohokam core, a majority of canal networks and sites with mounds and ballcourts were mapped in the earlier years of this century, prior to obliteration by urban sprawl. Distributions of centers, and thus communities, can be largely reconstructed where data are sufficient for precisely enumerating all constituent sites.

Figure 9.1. Distribution of Classic period mound sites in southern Arizona

Spacing of centers has suggested regularities in the size of integrated area. An average of 5.5 km between adjacent ballcourt sites along Phoenix canals was observed by Wilcox (Wilcox and Sternberg 1983:195) for predominantly late Preclassic ballcourts before A.D. 1150. Sites with more than one ballcourt may indicate a higher level of integration, reflecting coordination in rituals and related social interaction among several ballcourt communities.

The spacing of post-A.D. 1150 Classic period sites with platform mounds along the Salt and Gila river canal systems has been examined with the same methodology (Gregory and Nials 1985; Crown 1987). An average of about 5 km bewteen mound sites is closely similar to the 5.5 km distance for ballcourts (Table 9.1). These averages are derived from an only partially overlapping set of sites with both forms of public architecture. The convergence suggests continuity in basic size of integrated units between Late Preclassic and Classic periods. Classic mound spacing has been associated with distances suitable for regulation of canals and distribution of water along shared networks (Crown 1987: 155-158; Gregory and Nials 1985: 383). Regularities in linear distance between centers, along canals, may reflect optimal distances for agricultural travel and day-to-day communication within a single community or between adjacent community centers.

Sources: (a) Nials Gregory, and Graybill 1989; (b) Crown 1987; (c) Gregory 1987; (d) Castetter and Bell 1942:54; (e) Wilcox and Sternberg 1983; (f) Gregory and Nials 1985; (g) S. Fish and others 1989.

Community Area and Layout

Classic period mounds appear evenly spaced within the Phoenix area as a whole (Fig. 9.1), as well as linearly along individual canals. Average community territory can be calculated by generously outlining the extent of canal systems and dividing by the respective number of mound centers, if communities integrate all adjacent space whether irrigated or not. The average territory for each of the 23 Salt River mound communities would be roughly 40 square kilometers. Based on mound sites mapped by Patricia Crown (1987) along the Classic period canal systems on the Gila River, average territories here also approximate 40 square kilometers.

With 5-km spacing along canals and community territory averaging 40 square kilometers, the long axes of communities would tend to be across the canals. Gregory and Nials (1985:381) noted a linear, cross-canal arrangement of habitation features for major mound sites on the Las Colinas system that would have maximized land available along the canals for agricultural use. A cross-canal shape for communities as a whole, however, may also be a consequence of laterally integrated territory beyond the limits of irrigation. Nonirrigated land likely provided wild food resources, raw materials, fuel, and semicultivated desert species.

Irrigated Acreage

Another general trend for the Salt and Gila basin communities involves magnitudes of prehistorically irrigated acreage. Recent estimates place land irrigated from the Salt River at approximately 21,000 hectares (Nials and others 1989: 73-76), for an average of 935 hectares per platform mound community. Within the systems on the Gila River, recent estimates of 6250 irrigated hectares (Crown 1987) result in a comparable average of 1040 hectares for each Classic period community. Extents of irrigated acreage, like community size, may involve practical distances for regular communication and travel to fields (Crown 1987: 154; Gregory and Nials 1985: 383-384) or routine transport of agricultural products.

Although reasonably derived modalities can be cited for integrated area and irrigable acreage, these must be regarded as general averages incorporating a degree of community variation. For example, Crown (1987: 154-155) finds a correlation between the size of the mound site in Gila communities and mount of irrigable acreage. However, calculations for maximum acreage among Gila communities is no more than twice the average figure.

Population Size

Convergences in irrigated acreage may point to some optimal range for a more elusive and vital parameter of Classic communities: population. As a comparative exercise, estimates of supportable population in the Hohokam core can be made using amounts of cultivated land among late historic Pimans according to Castetter and Bell (1942: 54). A Piman family of five subsisted on 0.86 to 2.15 hectares. If plows, domestic animals, and commercial sale of harvests increased the acreage farmed by historic households, these figures might indicate the lower range for prehistoric populations that cultivated smaller plots. In an average prehistoric community containing about 1000 hectares of irrigated land, 2300 to 5800 persons could have been supported according to Piman analogy (Phoenix area totals would range from 53,000 to 133,000 persons). These figures overlap with previous estimates (Haury 1976: 356; Schroeder 1940: 20; Doyel 1991; P. Fish and S. Fish 1991). However, continuities in inferred Salt River community size between the late Prehistoric and Classic periods, based on similar spacing of earlier ballcourt and later mound centers, may or may not equate with a stability in population size determined by irrigated production. A greater inter-connectedness of Classic networks through canals linking major trunk lines, as posited by Nicholas and Feinman (1989), may indicate some degree of agricultural intensification. There is also general consensus that networks reached their maximum areal extent during this period (Masse 1981; Nicholas and Nietzel 1984; Nicholas and Feinman 1989; Plog 1980; Upham and Rice 1980).

Processes of ongoing aggregation and agricultural intensification in the Phoenix Basin during the Classic period thus may have created higher population densities within community territories that were of similar size to those of the late Preclassic period. In spite of obliterated segments of settlement pattern and variable archaeological visibility by time period, numbers of sites over time provide one avenue for evaluating this issue. A compilation by Upham and Rice (1980) showed a clear increase in site numbers from the late Preclassic to Classic periods along the lower Salt River. Additional factors of site size and occupational density at large sites that are relevant to relative population magnitudes cannot be chronologically compared with any precision in the Hohokam core. Estimated boundaries for large Classic period sites commonly incorporate the overlapping areal extents of all earlier components.

Settlement Hierarchies

The existence of site hierarchies is basic to the recognition of Hohokam communities. As observed by Doyel (1974, 1980), common canal usage creates a basis for sociopolitical integration embodied at larger sites or centers with administrative functions. Many Phoenix area irrigation networks serve multiple ballcourt or platform mound communities. There is a general agreement (Upham and Rice 1980; Gregory and Nials 1985; Crown 1987) that by the Classic period after approximately A.D. 1350, a few very large sites such as Pueblo Grande, Mesa Grande, and Casa Grande contain more massive public architecture than previous or contemporary centers. These sites have been interpreted as representing a further level of hierarchy integrating a number of community untits (Howard 1987: 218-220; Corwn 1987: 157; Nicholas and Feinman 1989: 225).

A recent analysis of settlement along a major canal system with multiple communities considers attributes of hierarchy in addition to size, public architecture, and spacing of central sites. Howard (1987) documents differential access to high value items between the foremost site of Mesa Grande and lesser sites with and without public architecture along the Lehi canal system. Mesa Grande, the largest site, has the most massive mound, the only known burial offerings of unusual richness, and the greatest access to ceramic trade wares, axes, and turquoise. No artificial data are available for a second size category of sites with less massive public architecture. A third category of smaller sites with walled adobe compounds but without mounds or ballcourts has lower frequencies of high value goods.

Differentiated Land Use

On the broad, flat, basin floor of the lower Salt River encompassing modern Phoenix, relatively homogenous topography would have fostered similarities in land use among irrigated communities. However, modern urban and agricultural disturbance allows only partial reconstruction of intracommunity patterns for isolated residential, agricultural, and extractive sites. A gradation in land use diversity can be observed between these environmentally most homogenous Phoenix communities, communities within the adjoining but somewhat narrower basin of the middle Gila River, and those in other Hohokam regions without perennial rivers and large-scale irrigation.

Even in the Phoenix vicinity, some degree of differentiation in settlement and land use is correlated with features of the valley floor (Wilcox 1991; Wilcox and Sternberg 1983; Masse 1991; Mitchell 1988). For example, three environmental zones with differential settlement and production characteristics can be defined for segments of the Mesa Grande system at increasing distances from the river (Howard 1987). A low density of small dispersed sites on the first terrace of the Salt River is consistent with a predominantly agricultural land use. A diversity of site types occurs on the densely occupied bluff area of the second terrace. Here, Mesa Grande contains multiple compounds, the largest mound, and a second mound of lesser size. Inland on the second terrace, several sites with public architecture are near the ends of main canal branches.

Greater diversity of land use is exhibited within individual Classic period communities along the Gila River (Crown 1987: 149-153), although all communities are located adjacent to the river due to the lesser width of this basin. Irrigated portions of sequential communities along the Gila occupy the floodplain. Substantial habitation sites occur on adjacent higher terraces, as do complexes of agricultural features dependent on water from storm runoff. Locations of field houses also suggest floodwater farming from tributary drainages in upper terrace and lower bajada situations. This combination of agricultural technologies is generally replicated from community to community.


Unlike the perennially flowing Salt and Gila rivers of the Hohokam core, desert rivers in the other basins of this cultural tradition are intermittent, lacking year-round flow over much of their courses. Without equivalent scales of riverine irrigation, spatial and organizational characteristics of communities in these regions cannot be related to shared systems, nor are settlement expressions nececessarily identical. Systematic records of central sites were not made by early observers, as along the Salt and Gila, and such distributional data for community identification are absent or incomplete. The delimitation of sites within community boundaries is also more difficult without clear spatial linkages among sites afforded by common canals, as in the Phoenix core. Although currently there are few descriptions of noncore communities, greater potential exists for understanding the role of dispersed populations in these areas where modern development has been less pervasive.

Community Spacing and Size

Ballcourts and even mounds are not yet fully tabulated outside the Salt and Gila core. Recent investigations have added significantly to the known Tucson instances of public architecture. Central sites of the Preclassic and Classic periods occur both along the Santa Cruz River and on the bajadas, reflecting varied and land-extensive agricultural bases. In keeping with this contrast between Phoenix and Tucson subsistence and settlement patterns, spacing between central sites in most sectors of the Tucson Basin is greater and more variable than in the core area. Settlement data revealing details of community size and layout are available only in a few cases.

In an exception to more distant Tucson spacing, William Doelle and others (1987) describe the distribution of primary or larger-sized Preclassic period villages in the southern basin along a stretch of the Santa Cruz with persistent surface flow. Most of these villages contain ballcourts. Spacing is approximately every 3 km along the river, a shorter distance than the 5.5 km between the Preclassic ballcourts along canals in the Phoenix area. Although riverine irrigation is probable on this part of the Santa Cruz, Tucson topography and hydrology would have restricted the size of the systems. Closer spacing of central sites than in the core area is not associated with denser populations, since territory surrounding these Tucson primary villages includes only an average of two small habitation sites.

Continuities in average spacing between centers of earlier ballcourt and later mound communities in the Phoenix Basin are not duplicated near Tucson. Sociopolitical reorganization throughout the Hohokam tradition occured at the end of the Preclassic period, at which time some Phoenix ballcourt sites were abandoned. In other Phoenix ballcourt sites, patterned arrangements have been identified between ballcourts and the mounds built in the earliest part of the Classic period (Gregory 1987), but use of Hohokam ballcourts appears to have ceased before the late Classic period. A disjuncture between Preclassic and Classic period centers is more strongly expressed by changing site primacy and other community dynamics outside the core area, where locational strictures of preexisting canal networks did not apply.

In the northern Tucson Basin, settlement clusters are concentrated along the Santa Cruz River and along mountain flanks at the basin edge (S. Fish and others 1989, Chapter 3 in this volume). Spatially separated Preclassic communities with ballcourt centers in each of these locations, at 70 and 57 square kilometers, are larger than the 40-square-kilometer average for core communities of the Classic period. These two earlier communities merged into the subsequent Marana Community. Sites appeared in the formerly intervening area, including the newly settled central mound site. The resulting early Classic period community spanned the basin from river to mountains, encompassing 146 square kilometers.

Settlement Hierarchies

Hierarchy in Tucson Preclassic period communities has been approached largely through site size and presence of ballcourts. A three-tiered settlement hierarchy in the Marana Community of the Classic period is based on site size, architecture, and ceramics. The four largest sites are distinguished by low frequencies of imported ceramics, higher consumption of decorated pottery, walled adobe compounds, and, in one case, the mound (see Chapter 3). The mound site is geographically central within the community, but in a location of secondary subsistence potential, suggesting an important role for intracommunity exchange. Intermediate-sized sites contain cobble outlines of adobe surface structures. More than three-fourths of the smallest residential sites, presumably with pit houses or less substantial surface dwellings, have no visible architectural remains. Contemporary sites fitting these Marana categories have been recorded widely in the Tucson Basin (Doelle 1985b; Wallace and Holmlund 1984), likely reflecting similar hierarchies in other Classic period communities.

Large portions of most Phoenix core communities have been destroyed and smaller elements of settlement pattern cannot be systematically recovered. However, sets of interrelated communities along the same network are demarcated by well-mapped canals. Relationships among the better preserved but more diffuse and widely spaced Tucson communities, on the other hand, could be similarly well established only through prohibitively broad retrieval of settlement data are sufficiently divergent between the Phoenix and Tucson basins to confuse comparison of hierarchical arrangements within and among communities. For example, it is unclear whether the previously described site hierarchy within the 146-square-kilometer Marana Community is organizationally more equivalent to a single mound community covering about 40 square kilometers along a major Phoenix canal or to the set of communities sharing that network.

Within proposed hierarchies involving Classic period communities along shared canal lines in the Salt and Gila core, developmental histories of the communities and their centers are incompletely known. It is generally unclear whether rankings based on size and magnitude of public architecture in central sites pertain equally to the early division of the Classic period or only to late Classic times. The Marana Community and contemporary entities are separated by vacant areas or attenuated settlement. Hierarchical relationships involving the Marana Community and early Classic counterparts in the northern Tucson Basin are not apparent, although merger of the two prior Preclassic communities within its boundaries seems to reflect a consolodation of territorial sway at the beginning of the Classic period. The developmental trajectories of the other early Classic communities are more poorly understood, including chronological details concerning centers and their mounds.

In the late Classic period, hierarchical relationships are possible among the set of communities located in the southern Tucson Basin and among those near the Picacho Mountains to the north. If so, a common irrigation network cannot have provided a relational basis, nor have preeminent centers and communities been identified. North of the Picacho Mountains along the Gila River, Classic mound centers sharing the Casa Grande canal system have been considered hierarchically related to the large and elaborate site of the Casa Grande (Crown 1987). Interestingly, the mound site of the Tom Mix Community near the Picacho Mountains is no farther from Casa Grande than the more distant centers on that common canal system.

Differentiated Land Use

Internal differentiation in settlement and subsistence activities by topographic zones seems to characterize communities outside the core area (Wood and McAllister 1980, 1984; Dolle 1985b; G. Rice 1987b) to a greater degree than ones within it (Crown 1987; Cable and Mitchell 1987; Howard 1987). Noncore communities frequently include segments along major drainages that could support irrigation or modest scale compared to core systems. Irrigated acreage may have contributed disproportionately to total production of annual crops, but substantial remainders of community land were farmed with alternative technologies.

In the Northern Tucson Basin Marana Community, six zones of functionally and topographically differentiated settlement cover almost the full range of basin environments (Chapters 3 and 4). Inhabitants of individual sites in the community may have pursued subsistence activities in one or more adjacent zones, but distances and environmental diversity necessarily entailed variable participation in agricultural technologies among the population as a whole. The geographic concentration of rockpile fields also demonstrates an appreciable degree of agricultural specialization (S. Fish and others 1989).

Irrigation networks, series of adjacent floodwater fields, and the large rockpile complexes all involved common interests and efforts for subsets of community members. Households and villages liekly participated in more than one agricultural association or water users group that coordinated joint tasks, scheduling, or water allocation. Relations based on consensus and cooperation in these contexts would have crosscut and sequentially interlinked members across the community. Such networks of interrelationship may have served as a social and economic cement across productively diversified communities in a manner resembling the interdependence created by shared use of massive canal systems in the Hohokam core.


Even with the best distributional data, population in noncore regions can be approached only through the broadest comparative estimates. Agricultural acreage cannot be approximated in a manner similar to the extent of irrigable land in the core as a basis for calculating totals. However, without the same degree of settlement continuity imposed by canals in the Phoenix Basin, superpositioning of successive occupations over hundreds of years occurs in fewer sites. It is therefore less difficult to assess individual components in settlement and regional trends over time. Indeed, population shifts and rearrangements can be recognized in the developmental histories of these areas.

In the northern Tucson Basin, population is a variable in community dynamics. Densities in the early Classic period Marana Community seem only partially accounted for by inhabitants of the two preceding communities within its territory. Habitation site area in the Preclassic period communities totals about 2,000,000 square meters compared to just under 6,000,000 square meters in the early Classic period community, which represents a shorter time span. Although local population growth must have contributed to this increase, it is probable that some members of the Marana Community were newly arrived. Habitation site area averages 16,000 square meters for each kilometer of the 127 square kilometers covered by the two Preclassic entities compared to 40,000 square meters for each kilometer in the 146-square-kilometer Classic period community.

Population dynamics further accelerate during the following Hohokam sequence in the Tucson Basin. The Marana Community and an adjacent early Classic community in the northern basin are abandoned by the beginning of the late Classic interval marked by Salado polychromes after A.D. 1350. Nearly 1300 square kilometers (500 square miles) surrounding these communities lack habitation sites of that time. Rather than wholesale population loss, these dramatic shifts in settlement are probably related to processes of aggregation.

On either side of this abandoned area, late Classic communities cluster in the southern Tucson Basin and near the Picacho Mountains to the north. In some cases it appears that mounds were first constructed at the central sites during the late Classic period. As with the largest sites of that date in the Hohokam core, late Classic mound sites in both the northern and southern clusters of communities exhibit the densest architectural and artifactual remains for any period, suggesting that population densities within central sites reached a peak. These late Classic communities near Tucson also coincide with hydrological situations most suited to agricultural intensification through irrigation and other means.

Population magnitudes of the late Classic period McClellan Community, surveyed by the same full-coverage method as the Marana study area (Fig. 1.8), are greater than those of the earlier Marana Community. The McClellan Community shares the developmental pattern of an incorporation of two earlier Preclassic period settlement concentrations into a single Classic period unit. Population then peaks in Late Classic settlement covering 136 square kilometers or 10 square kilometers less than the Marana Community. Total habitation site area for the McClellan Community of this date is about 16,000,000 square meters. Increased population density in the later community is registered by an average of 118,000 square meters of habitation site per square kilometer compared to the 40,000 square meters per square kilometer at Marana


Although locational and organizational imperatives of massive irrigation may have been critical along the Salt and Gila rivers, shared canals were not the impetus for Hohokam communities of the Tucson Basin and other regions. Cooperation, coordination, and any central decision making could not have been shaped by such interaction. Yet similar community patterns began in the Preclassic, and by the Classic period included settlement hierarchies and mound precincts. Risk sharing and subsistence exchange for larger populations in regions lacking dependable irrigation are among probable community functions in these cases. A concept of integrated communities transcending irrigation or other locationally specific needs must have existed and been transmutable to a variety of environmental situations. This basic organizational structure provided the integrative framework for settlement and society throughout the Hohokam tradition.

In worldwide developmental sequences, there is a repetitive correlation between integrative structure and public architecture. The relationship between these phenomena may be particularly significant for the Hohokam. Mounds are diagnostic elements by which archaeologists define Classic period communities, and they may have been similarly perceived by their builders. Frequently identified with public ceremony and leadership roles for the individuals using or inhabiting their precincts, mounds are also the most imposing and visible structures of the Hohokam community. Construction episodes, which in at least some cases were periodic and enlarged mound size by stages, must have involved participation and logistical support from many social groups. Mounds likely embodied symbols of community identity, cohesiveness, and differentiation from other such entities in surrounding areas.

The most massive instances of Hohokam public architecture appear in the late Classic period, in sites often regarded as representing an additional level of hierarchy linking several lesser mound communities. The erection of mounds may have served in the expression of inter-community relations involving both hierarchy and competition. One aspect of such "peer polity interaction" (Price 1977; Renfrew 1986: 1-8) in the Tucson Basin may have been competition for population. Both early and late Classic communities of this region apparently grew their ultimate extents through increments of population, drawn from preceding or contemporary communities. Agricultural intensification and a general increase in productive specialization may have provided incentives for more concentrated settlement in late prehistoric time.


The Marana Community represents no superlatives in the Hohokam world. Its developmental history begins with preceramic agriculturalists, but ends while other communities continue to grow. The mound at its center is small compared to most Classic period edifices. The areal extent of the Marana Community is greater than the average for territories surrounding mound centers in the Hohokam core, but size information is inadequate for noncore comparisons. Although population estimates derived from irrigated acreage in the core cannot be directly compared to measures of habitation area in Marana sites, it is doubtful that Marana and other outlying communities attained Phoenix Basin densities. According to site area measures, a population of greater magnitude lived within the late Classic period McClellan Community of the same noncore region.

Delineation of the Marana Community has unfolded in a context of settlement pattern unparalleled in Hohokam archaeology. Survey in the northern Tucson Basin is the first instance of full-coverage examination of similarly large contiguous blocks and systematic coverage at a truly regional scale. Although survey results have provided the primary source of interpretations concerning the Marana Community to date, the major advantage of a regional backdrop is an increased depth for inter-connected inquiries at multiple scales. The most powerful insights have arisen through an interplay among outcomes from several investigative levels, studies of broad and narrow focus, and findings from survey and excavation. This interplay is perhaps best illustrated by the topic of agave cultivation. Fine-scale studies of features, artifacts, and biotic remains at individual sites interdigitate with regional distributions of cultural and environmental variables to provide a basis for economic interpretation. The role of the mound center in community organization and dynamics is another example of such evidence merger. As future excavations are undertaken, the ability to place site-specific studies in a community framework will simultaneously sharpen understanding at smaller and larger scales.

Quantification presents an ongoing challenge in Hohokam studies. The benefits of a large data set for deriving relative magnitudes where more precise estimates cannot be achieved is illustrated by comparative measures for the Marana Community. Habitation area in community sites as a proxy for population is a prime example. Trends derived from community-wide distributions are clear and convincing, as in the tripling of Preclassic habitation area in the Marana Community during the Early Classic period. Increasing community densities over time are strongly indicated by averages for site habitation area per square kilometer. The Preclassic Marana level of 16,000 square meters per square kilometer increases to 40,000 square meters in the early Classic period, and rises again to 118,000 square meters per square kilometer in the McClellan Community of the late Classic period. These quantitative data may be subject to substantial future refinement, but reversal of magnitudes is highly improbable.

If not distinctive among its prehistoric counterparts, the Marana Community is nevertheless archaeologically unique in several respects. Precisely because there has been relatively minor use of community territory from the beginning of the Late Classic period until the present, the record of growth and change from the earliest ceramic occupations through the Early Classic period is particularly clear. Marana studies provide the first comprehensive view of community parameters in a Hohokam region outside the valley of the Salt and Gila rivers. Nowhere else has such a detailed and complete record yet been obtained of the settlement components of a Hohokam community, their spatial relationships to one another, and to features of the desert environment.

Investigations in the Marana Community have revealed a social and territorial configuration that incorporated broad environmental diversity, but constituted a well-integrated subsistence system. Agricultural technologies were distinctive from zone to zone, varying with topography and sources of water. Because of localized risks among zones, it is unlikely that individual community segments could have sustained Classic period population densities in isolation; reciprocity and circulation of subsistence items within the community undoubtedly contributed to the long-term welfare of all members. For inhabitants in the desert basins of intermittent rivers, such communities encompassed the basic requirements for population sustenance and reproduction and provided the means for equal participation in the Hohokam tradition.

At the end of the Hohokam sequence sometime after A.D. 1400, a long-term developmental trajectory ceased and community organization disappeared. Less aggregated and smaller successor populations in southern Arizona echoed many aspects of Hohokam lifeways, including the eventual Pima resurrection of the large-scale canals on the Gila River in the mid-nineteenth century, but the previous integrative principles and their expression in public architecture never reappeared. The level of integration embodied in communities is a hallmark of Hohokam society and distinguishes their tradition from all others in the southern deserts of the southwestern United States.

Next: References | Previous: Chapter 8

UAP Online Books  |   UAP Home Page

The University of Arizona Press, 2/10/01 5:04PM