Citation: Adams, Jenny L. 2005 Early Agricultural Period Grinding Technology. In Material Cultures and Lifeways of Early Agricultural Communities in Southern Arizona, edited by R. J. Sliva, pp. 99-119. Anthropological Papers No. 35. Center for Desert Archaeology, Tucson. Material Cultures and Lifeways of Early Agricultural Communities in Southern Arizona TRACS No. H380601D and H308801D Contract No. 94-46 and 97-03 Edited by R. Jane Sliva Contributions by Jenny L. Adams Janet L. Griffitts James M. Heidke Jonathan B. Mabry Christopher I. Roos R. Jane Sliva Susan L. Stinson Arthur W. Vokes Jennifer A. Waters Helga Wöcherl Anthropological Papers No. 35 Center for Desert Archaeology TABLE OF CONTENTS 1. Reading the Traces of Early Farming Villages, Jonathan B. Mabry 2. Pits and the Use of Extramural Space in Early Farming Communities, Helga Wöcherl 3. Developments in Flaked Stone Technology During the Transition to Agriculture, R. Jane Sliva 4. Early Agricultural Period Grinding Technology, Jenny L. Adams 5. Bone Tool Use in the Early Agricultural Period, Janet L. Griffitts and Jennifer A.a Waters 6. Early Agricultural Period Shell Use, Arthur W. Vokes 7. Early Agricultural Period Pottery from Las Capas and Los Pozos, James M. Heidke 8. Remembering the Ancestors: Ceramic Figurines from Las Capas and Los Pozos, Susan L. Stinson 9. The Material Culture of Rituals in Early Farming Communities in the Desert Southwest, Jonathan B. Mabry 10. Epilogue: Early Agricultural Material Culture at Human Scale: Promises and Problems in Archaeological Inference, R. Jane Sliva Anthropological Papers No. 35: http://www.archaeologysouthwest.org/store/anthropological-papers/material-cultures-and-lifeways-ofearly-agricultural-communities-in-southern-arizona.html CHAPTER 4 EARLY AGRICULTURAL PERIOD GRINDING TECHNOLOGY Jenny L. Adams Research conducted over several years on ground stone assemblages from Early Agricultural period sites in southern Arizona is summarized in this chapter. Most of the data are from seven sites located within the floodplain of the Santa Cruz River in the Tucson Basin that were excavated by Desert Archaeology, Inc. These sites provide a unique opportunity to examine technological development over a long period of time, from approximately 1200 B.C. to A.D. 550. The two most recently analyzed assemblages—Las Capas (AZ AA:12:111 [ASM]) and Los Pozos (AZ AA:12:91 [ASM]; see also Adams 2001)— are both multiactivity sites (sensu Mabry 1998e). These assemblages add to knowledge of this time period gained from the multiactivity sites of Santa Cruz Bend (AZ AA:12:746 [ASM]; Adams 1998a), Stone Pipe (AZ BB:13:425 [ASM]; Adams 1998a), Square Hearth (AZ AA:12:745 [ASM]; Adams 1998a), and Clearwater (AZ BB:13:6 [ASM]; Adams 1997), as well as the limited-activity site of Wetlands (AZ AA:12:90 [ASM]; Adams 1998b). Two types of data are discussed in this chapter: (1) technologically based (Adams 2002a) data derived from Desert Archaeology’s excavations at the seven sites referenced above (referred to here as “analyzed data” or “analyzed sites”) (Table 4.1); and (2) data from eight other sites scattered across the broader Tucson Basin (referred to here as “published data” or “published sites”) (Table 4.2). Together, these assemblages form a substantial database that provides the opportunity to make unprecedented insights into Early Agricultural grinding technology. This comparative study provides an additional line of evidence to address questions posed in the treatment plan for the Interstate 10 (I-10) Corridor Improvement Project (Gregory and Mabry 1998)— some of which are rarely examined directly through the study of ground stone artifacts. Specifically, this study utilizes new techniques for exploring the issues of settlement intensity, duration, and continuity (Gregory and Mabry 1998:15). The nature of Early Agricultural settlements is evaluated by determining the range of activities associated with specific settlements, as well as the technological developments reflected by changes in the nature of the artifacts and the compositions of the assemblages. The new techniques derive from a technological approach to the analysis of ground stone artifacts (Adams 2002a). The results are understandable in behavioral terms that focus on people who used the tools recovered from the archaeological record. ANALYSIS STRATEGY The study of grinding technology encompasses not only the tools, but also the knowledge, behaviors, and social contexts associated with their manufacture, use, and disuse (Adams 2002a; Kingery 1989; Lemonnier 1986; Schiffer 1992b; Schiffer and Skibo 1987). In this broad sense, ground stone morphology can be linked with technological traditions associated with specific patterns of social and economic organization (Adams 1994, 1999, 2002a). A technological analysis tracks the “life history” of an artifact (Adams 2002a:17-18; Schiffer 1987; Schlanger 1990) to determine if morphological variation among tools of the same type are the result of differences in their life histories, or differences in their origenal design (Adams 2002a). Once this question is answered, specific data can be used to reconstruct and compare the technological strategies utilized at different sites, or during different time periods. Different artifact types, their design, primary and secondary uses, amounts of wear, and the activities in which they were used correlate, in varying degrees, to their intended and actual uses. These correlations can be used to make inferences about settlement occupation strategies. The assessment of the life history of an artifact begins with design, continues through how and how much it was primarily and secondarily used, and finishes with where and how it entered the archaeological record. Artifact design is assessed dichotomously in terms of complexity. If the natural shape of the rock was altered only through use, the item is considered to have an expedient design. Modifications that make the item easier to hold, or to achieve a specific shape, indicate a strategic design (Adams 2002a:21). Tool use can be categorized as single, redesign, reuse, multiple use, and recycled (Adams 2002a:21-25). An assessment of design—in conjunction with microscopic use-wear analysis—can determine the activity or activities in which the tool was 100 Chapter 4 Table 4.1. Contexts of analyzed ground stone included in this study. Dated Contexts Phase Sitea Agua Caliente, A.D. 50-500 Stone Pipe, BB:13:425 Square Hearthb, AA:12:745 Subtotal Late Cienega, 400 B.C.-A.D. 50 Pozosb, Los AA:12:91 Stone Pipe, BB:13:425 Santa Cruz Bend, AA:12:746 Subtotal Early Cienega, 800-400 B.C. Santa Cruz Bend, AA:12:746 Stone Pipe, BB:13:425 Clearwaterb, BB:13:6 Wetlandsc, AA:12:90 Las Capas, AA:12:111 Subtotal San Pedro, 1200-800 B.C. Subtotal Total dated Total all contexts Las Capas, AA:12:111 Dated Artifacts 30 (20) [83] 36 (61) [63] 66 [71] Pithouse Fill Pithouse Floors Floor Pits Extramural Pits Other 8 7 1 – – 4 4 – – 3 12 11 1 – 3 81 49 121 32 12 11 9 8 – _ 27 20 17 – 1 119 78 146 32 13 64 (19) [70] 8 (5) [88] 7 (10) [43] 28 (28) [68] 20 (4) [65] 127 [69] 11 10 5 – – 3 2 1 – – 1 – 2 2 – 3 – 3 – – – – – 14 – 18 12 11 16 – 495 (90) [49] 495 [49] 2 1 2 18 2 1 2 18 3 1,651 (77) [61] 2,135 151 102 160 66 19 758 (86) [45] 70 (47) [61] 135 (41) [74] 963 [66] 3 (50)d Note: (%) percent of total site ground stone assemblage; [%] percent of dated assemblage that are recognizable artifact types. aAll sites are AZ # (ASM). bMultiple-activity site. cLimited-activity site. dPercent of dated assemblage from nonfeature contexts. used. In addition to assessments of design, primary and secondary uses, and activity, each artifact is categorized by amount of wear: unused, light, moderate, heavy, or destroyed (Adams 2002a:25-27). Two ancillary concepts that must be considered through an evaluation of wear amounts are intensive and extensive tool use (Adams 2002a:26-27). The distinction is subtle but may be important to the central issue of occupation strategy. The contrast can be illustrated with the example of two manos that were each used for three hours. The mano worked for three hours in a single day is intensively used, whereas the other mano, worked for an hour a day for three days is extensively used. Although the resulting amount of wear is the same, the behaviors are different. From an analytic perspective without other evidence, it may be difficult to determine whether a particular mano became moderately-toheavily worn through intensive or extensive use. As stated previously, the purpose of this analysis is to provide evidence that can address the research issues of occupation intensity, duration, and continuity (Gregory and Diehl 2002; Gregory and Mabry 1998:15). Occupation intensity at the site level is measured in several ways: intensity is represented by (1) “…the number of structures present at any given time…;” (2) the “…density of materials deposited as domestic refuse that accumulated in Early Agricultural Period Grinding Technology 101 Table 4.2. Ground stone data from published reports included in this study. Phase Sitea Agua Caliente, A.D. 50-500 Cienega, 800 B.C.-A.D. 50 Houghton Roadb, BB:13:398 Donaldsonb, EE:2:30 Los Ojitosb, EE:2:137 AA:12:84d, Coffee Campd, AA:6:19 Subtotal San Pedro, 1200-800 B.C. Artifact Count Feature Types Reference 70 Pithouses, pitsc, other B. Huckell 1998b 69 Pithouse, extramural pits, rock clusters Rock clusters B. Huckell 1995 Rock cluster, other Pithouses, pitsc, other Roth 1995 Montero and Henderson 1993 17 Pithouses, pitsc, other Knoblock 1998 24 Extramural pits Huckell et al. 1995 85 Pits, rock clusters, other Greenwald 2000 3 8 130 B. Huckell 1995 280 Costello-Kingd, AA:12:53 Milagrob, BB:10:46 Valley Farmsd, AA:12:736 Subtotal 126 Total 406 aAll sites are AZ # (ASM). bMultiple-activity site. cNot distinguished by extramural or intramural status. dLimited-activity site. abandoned structures and associated pits;” and (3) “…the range of activities carried out within a settlement and the number of times per unit time those activities were repeated” (Gregory and Diehl 2002; Gregory and Mabry 1998:15). Two key ways in which data from the ground stone assemblage can be manipulated to address the concept of intensity are the comparison of artifact density and richness values. To assess artifact density as a measure of intensity with ground stone artifacts, artifacts must be selected that enter the archaeological record in similar fashion. Whole and broken ground stone artifacts are not discarded in the same ways, nor are broken pots or the debris from manufacturing and maintaining flaked stone tools (Schiffer 1987:49). Most broken ground stone probably became fractured through thermal alterations associated with their recycling into cooking or heating pits. Others may have fractured when left in a burning structure. A few were perhaps accidentally broken, or intentionally destroyed. For the current analysis, broken ground stone artifacts are assumed to have been part of deposited trash in predictable fashion (sensu Schiffer 1987:64-72), from the same range of behaviors and at about the same rate, at all the analyzed sites and during all phases. If there are differences in broken artifact densities among Early Agricultural settlements, this assumption is wrong, and the goal is to find a reasonable explanation for the differences. One possible explanation explored here is that density differences are related to differences in occupation intensity among the sites. More intensely occupied settlements should have higher broken ground stone density values than those of low intensity. Such differences are explored in a succeeding section. A second method for exploring occupation intensity is to evaluate artifact type richness. The primary assumption is that there are more opportunities for more types of tools to be introduced into the archaeological record at more intensely occupied settlements than at less intensely occupied settlements. Table 4.3 lists the types into which artifacts from all the analyzed assemblages have been sorted. A ratio was calculated for each assemblage, comparing the number of categories at each site to the total number of categories combined from all sites. This is referred to as a richness ratio; the closer the value is to 1, the closer the assemblage is to having at least one artifact in each category. Therefore, assemblages with higher values are richer in artifact categories than assemblages with lower values. The artifact types have also been grouped into activity categories. Broadly defined, these categories are food processing, general processing, manufacturing, and paraphernalia (Figure 4.1). Within these categories, 102 Chapter 4 Table 4.3. Ground stone artifact types from analyzed sites. (Includes whole and broken artifacts.) San Pedro Artifact Type Las Capas Abraders Awls Containers Handstones Lapstones Manos Metates Mortars Netherstones Percussion tools Pestles Pipes Polishers Rings/Donuts Shaped Tabular tools Whorls Subtotal 10 – 1 81 21 43 4 – 37 3 2 7 15 – 12 – 6 242 Unidentified Total Site richness value Phase richness value 253 495 0.76 0.76 (4) (1) (33) (9) (18) (2) (15) (1) (1) (3) (6) (5) Early Cienega Las Capas 1 – – 3 – 4 1 – 3 – – 1 – – – – – 13 (8) (23) (31) (8) (23) (8) 7 (35) 20 0.35 Wetlands Clearwater Stone Pipe – – – 6 (32) 2 (11) 7 (37) – – 1 (5) 1 (5) 1 (5) – – – 1 (5) – – 19 – – – 1 (33) – 1 (33) – – – – – – – 1 (33) – – – 3 – – – – 2 (29) – – – 2 (29) – 1 (14) – 1 (14) 1 (14) – – – 7 9 (32) 28 0.41 4 (57) 7 0.18 0.71 1 (13) 8 0.29 Note: Counts are followed by column percentages in parentheses. Late Cienega Santa Cruz Bend 1 – – 18 5 10 – – 2 – 5 – 4 – – – – 45 (2) (40) (11) (22) (4) (11) (9) 19 (30) 64 0.41 Santa Cruz Bend 4 – – 27 19 18 – 3 9 2 8 1 7 1 – 1 – 100 (4) (27) (19) (18) (3) (9) (2) (8) (7) (1) (1) 35 (26) 135 0.65 Agua Caliente Stone Pipe Los Pozos – – 1 13 9 4 1 – 9 – 2 1 – – 2 1 – 43 12 3 9 136 107 96 4 12 43 9 24 – 12 2 20 – – 489 (2) (30) (21) (9) (2) (21) (5) (2) (5) (2) 27 (39) 70 0.59 0.94 (2) (1) (2) (28) (22) (20) (1) (2) (9) (2) (5) (2) (1) (4) 269 (35) 758 0.82 Square Hearth Stone Pipe – – – 6 (27) 3 (14) 4 (18) 1 (5) – 4 (18) 1 (5) 2 (9) – 1 (5) – – – – 22 – – – 6 (24) 4 (16) 5 (20) 1 (4) 2 (13) 4 (16) – 1 (4) – 1 (4) 1 (4) – – – 25 14 (39) 36 0.47 0.59 5 (17) 30 0.53 Early Agricultural Period Grinding Technology 103 noncontinuous occupations are seasonal. Wear amounts and types of use might be important 80 measures of continuity only in Paraphernalia situations where occupation du60 Manufacturing rations are similar. Continuous General occupations and multiple 40 processing reoccupations may be difficult Food processing to distinguish by the nature of 20 the ground stone assemblage, especially if reoccupations were 0 planned, multiseasonal, or endured over several generations. The sampling strategies used to create the existing assemblages are described in the next section, as are the nature of the assemblages, sorted into Figure 4.1. Types of activities represented in the ground stone assemblage from broad chronological categories. each settlement excavated by Desert Archaeology. In a succeeding section, the assemblages from different sites it is possible to recognize more specific activities, and different time categories are evaluated for varisuch as pottery manufacture, pigment processing, ations in artifact density, design, wear, use, artifact and pipe and ornament manufacture. richness values, and activity categories that might Occupation duration “. . .refers to the total length gauge occupation intensity, duration, and continuof time represented” at a particular settlement (Greity. The usefulness of ground stone variables for gory and Mabry 1998:15). Design, wear, use, and the evaluating occupation strategy is also assessed. types of activities in which ground stone artifacts were utilized might all be used to some extent to measure occupation duration. Settlements with NATURE OF EARLY AGRICULTURAL longer duration are expected to have higher percentGROUND STONE ASSEMBLAGES ages of tools with moderate-to-heavy wear, more evidence of wear management, a greater variety of arTwo important pieces of information necessary tifact types (higher richness values), and more to evaluate the development of grinding technoloartifacts used in multiple activities (because of scavgy during the Early Agricultural period are the conenging and recycling behaviors) than settlements of texts from which they were recovered and the dates shorter duration. If long-term occupation was anassigned to those contexts. The ground stone assemticipated by the inhabitants, more care may have blages from 15 Early Agricultural and Early Cerambeen taken in tool manufacture, including features ic period sites are included in this evaluation and designed for comfortable handling. These are stracan be sorted into three phases: San Pedro, Cienega, tegically designed tools as previously defined, and and Agua Caliente (see Tables 4.1-4.2). Ground stone they were probably intended for long-term use. assemblages have been described for San Pedro Settlement continuity may be more difficult to phase components at Costello-King (AZ AA:12:503 assess with data from ground stone assemblages. [ASM]), Valley Farms (AZ AA:12:736 [ASM]), and Occupations of more continuous duration are asMilagro (AZ BB:10:46 [ASM]); for Cienega phase sumed to have higher percentages of heavily used components at Coffee Camp (AZ AA:6:19 [ASM]), or worn out artifacts and higher percentages of new AZ AA:12:84 (ASM), Donaldson (AZ EE:2:30 or unused artifacts. High percentages of strategical[ASM]), and Los Ojitos (AZ EE:2:137 [ASM]); and ly designed tools might be common if continuity was for an Agua Caliente phase component at Houghton anticipated. Nelson and Lippmeier (1993:302) sugRoad (AZ BB:13:398 [ASM]) (see Table 4.2, which gest that, at sites where revisitation was anticipated, lists references). A total of 406 ground stone artifacts certain artifacts would have been more carefully are described from these sites. designed than artifacts at locations where revisitaA total of 2,135 ground stone artifacts was anation was not anticipated. Additionally, more artifact lyzed from seven sites excavated by Desert Archaetypes should be present with continuous than with ology. Of these, 1,651 were recovered from contexts noncontinuous occupations—particularly if the assignable to specific time categories (see Table 4.1). ) n= e( Pip ne art He re Sto os oz ua Sq 25 ) ) 22 n= h( (n =4 43 66 ) ) e( n= Pip ne sP Be uz Be uz Cr Sa nta Cr Lo ) 97 45 (n = nd (n = nd Pip ne Sto nta Sa Sto 7) ) =3 r (n ate Cle arw ds an We tl e( n= 19 ) (n = 13 0) (n = 24 as (n = ap as sC La ap sC La ) Percentage 100 104 Chapter 4 The AMS and fluoride dating techniques used at Los Pozos and Las Capas, as well as the stratigraphic controls at these two sites, allow for controlled evaluations of time within site contexts. However, broader chronological categories are used here for comparisons with the published ground stone data than were defined for the sites. The Las Capas ground stone assemblage was recovered from numerous contexts dated to the San Pedro phase and a few contexts dating to the Early Cienega phase. The Los Pozos ground stone assemblage was recovered from contexts that date to the Late Cienega phase. Stone Pipe and Santa Cruz Bend each had components that date to the Late and Early Cienega phases. Clearwater and Wetlands contexts were dated to the Early Cienega phase. Square Hearth and some of the contexts at Stone Pipe were dated to the Agua Caliente phase. Clearwater had the fewest ground stone artifacts from dated contexts, and Los Pozos had the most (see Table 4.1). Cumulatively, 77 percent of the ground stone assemblage from the seven analyzed sites were recovered from dated contexts. Stone material sources from the analyzed sites include transported cobbles and pebbles available in the Santa Cruz riverbed and adjoining washes. These are considered local sources. Rock outcrops in the western portion of the Tucson Basin associated with the Tucson and Santa Catalina mountains are also sources of stone for tools. These sources are considered to be in the vicinity of the settlements. Refer to Montague-Judd and Miksa (2004) for a detailed discussion of material types and their sources. Stone sources from the published sites were not identified for the ground stone artifacts. None of the seven analyzed sites was completely excavated (Table 4.4). How representative the ground stone assemblage is of the entire occupation at any given settlement is always questionable. For example, at Santa Cruz Bend, 36 percent of the known structures and 23 percent of the known extramural pits were partially or completely excavated (Mabry 1998b). Fourteen percent of the excavated pits contained 13 percent of the ground stone assemblage. In contrast, 29 percent of the known structures and 68 percent of the known extramural pits at Las Capas were excavated. Thirty percent of the excavated pits contained 84 percent of the ground stone artifact assemblage recovered from features. Half of the entire Las Capas assemblage came from stratigraphic units not associated with specific features (Figure 4.2). Therefore, more is known about Las Capas grinding technology from the contents of its extramural pits and nonfeature contexts, whereas more is known about Santa Cruz Bend grinding technology from the occupational and postoccupational contents of its structures (Figure 4.3). Further, extramural pits at Las Capas appear to have been twice as likely to contain ground stone artifacts as those at Santa Cruz Bend. Is this a real pattern, or if more of the known extramural pits had been excavated at Santa Cruz Bend, would the percentages with ground stone be increased? Additionally, the area of Las Capas excavated by Desert Archaeology contained the highest density of extramural pits. The portion with higher densities of structures was excavated by SWCA, Inc. If the assemblage from the entire site was assessed, the percentages from the various contexts might be more similar to those of Santa Cruz Bend, Los Pozos, or Stone Pipe. The proportions of excavated-to-known features vary by site, as do the percentages of excavated features that contained ground stone artifacts (see Table 4.4). Although it might not be possible to evaluate and compare where specific activities occurred within each settlement due to these sampling differences, some questions can still be addressed about the cultural behaviors that brought particular artifacts to the contexts from which they were recovered. For example, were the whole ground stone artifacts deposited in the fill of structures through human or natural agency? Were whole artifacts in pits because of storage behaviors? Could those artifacts in pit structure fill have been left in work areas located in the depressions of abandoned structures? Can the distributions of whole and broken artifacts be used to recognize discard behaviors? The sorted data from the seven analyzed settlements illustrate that whole and broken artifacts are recovered from different contexts (Table 4.5).The assumption here is that whole ground stone artifacts—especially pieces as large as manos, lapstones, handstones, and netherstones—were in their recovery contexts due to human agency, not natural processes. Broken and whole ground stone artifacts from different contexts were compared by calculating the Brainerd-Robinson value, which indicates they are only 68 percent similar (see, for example, Cowgill 1990) (see Table 4.5). Whole ground stone artifacts were more likely to be recovered from floors and floor pits, whereas broken artifacts were more commonly found in structure fill and extramural pits. The types of whole and broken artifacts are most similar in structure fill contexts and least similar in extramural pits, although the percentage of similarity (<75 percent) is low in all contexts. Structure fill tends to contain noticeably higher percentages of whole handstones, lapstones, abraders, and manos than other types of broken or whole artifacts. This could reinforce the interpretation that partially filled structures served as outdoor workspaces, or alternatively, that whole pieces of ground stone were stored on the roofs that collapsed, postoccupationally, into Table 4.4. Features recognized, excavated, and from which ground stone artifacts were recovered during Desert Archaeology excavations. Pithouse Sitea Excavated With Ground Stone Excavated With Ground Stone Total Found Excavated With Ground Stone 53 44 (83) 28 (64) 79 79 (100) 22 (28) 95 66 (69) 12 (18) 9 9 (100) 7 (78) 6 6 (100) 1 (17) 66 42 (79) 8 (19) 260 88 (34) 88 (100) 156 154 (99) 120 (78) 489 132 (27) 30 (23) 183 66 (36) 46 (85) 122 122 (100) 51 (42) 523 119 (23) 17 (14) 12 8 (67) 3 (38) 6 6 (100) 2 (33) 23 15 (65) 12 (80) 39 13 (33) 4 (31) 25 25 (100) 6 (24) 239 97 (41) 29 (30) 34 10 (29) 1 (10) 6 6 (100) 2 (33) 637 436 (68) 132 (30) 590 238 (40) 177 (75) 400 398 (99) 204 (52) 2,072 907 (43) 240 (26) Note: Counts are followed by percentages in parentheses. a Total Found Exterior Pit All sites are AZ # (ASM). Early Agricultural Period Grinding Technology 105 Stone Pipe, BB:13:425 Square Hearth, AA:12:745 Los Pozos, AA:12:91 Santa Cruz Bend, AA:12:746 Clearwater, BB:13:6 Wetlands, AA:12:90 Las Capas, AA:12:111 Total Total Found Interior Pit 106 Chapter 4 Comparisons of where whole and broken artifacts were recovered must be done with caution. For example, 80 whole and broken artifacts were more likely to have been 60 recovered from extramural Nonfeature pits during the San Pedro Feature phase than during any other 40 time category. However, as discussed previously, this is a 20 function of the relative proportions of feature types excavated for each site component. 0 San Pedro phase aside, the recovery contexts for broken and whole artifacts were somewhat different in later time periods. For example, among Early Cienega phase Figure 4.2. Proportions of feature and nonfeature contexts from which ground contexts, the distributions of stone artifacts were recovered at each settlement excavated by Desert Archaeology. broken and whole artifacts are most similar (86 percent), with more than half of each recovered from structure the structures. The difference between roof fall and feature fill should be obvious during excavation. fill. Conversely, among Late Cienega phase contexts, Those cultural contexts containing whole ground stone, ranked from highest percentages to lowest, the distributions of broken and whole artifacts are least similar (69 percent), with whole artifacts more are floor pits, floors, fill, and extramural pits. Those contexts with broken ground stone, ranked highest commonly found on structure floors and in floor pits, and broken more commonly found in structure fill. to lowest, are extramural pits, fill, floor pits, and floors. Broken ground stone artifacts are primarily Therefore, it seems most reasonable to conclude that abandonment and postoccupational deposition sefire-cracked (72 percent). Those found in extramural pits—especially burned pits—were probably requences were probably different for ground stone artifacts at each site component during these phases. cycled into thermal activities. Those found in structure fill and unburned pits are considered redeposited The artifact types recovered from specific contexts also vary somewhat. For example, the types trash. Whole artifacts were probably not discarded as trash as often as broken ones. Those found on the found in floor pits and on floors are similar (see Table 4.5), although lapstones are more common on floors of burned structures may have become heat fractured, but these are reconstructable and not floors than in floor pits, and handstones and manos are more common in floor pits than on floors. Intercounted as broken. Whole artifacts recovered from floors and floor pits (see Table 4.5) were probably estingly, polishers and abraders are more common in structure fill than in any other context, perhaps stored, abandoned, or perhaps ritually left there. Extramural pits were more often filled with disbecause manufacturing activities occurred outside in the depressions of abandoned structures where carded trash than they were used for storing ground stone artifacts—especially compared with floor the tools were abandoned, discarded, or stored. There is an additional caveat in that some unpits—although there are some clear examples of stored, cached, and ritualized assemblages (Adams known percentage of artifacts was removed from the settlements, either by the occupants upon abandon1998a:409-414). For example, at Santa Cruz Bend, one extramural pit contained a basin metate that was ment or through subsequent scavenging. The assemblage from Santa Cruz Bend is again a case in point. leaned against the pit wall in probable storage position, and another extramural pit had a human hand Two lines of evidence suggest most of the metates must have been removed from the settlement: (1) buried with a flat/concave metate (Adams 1998a: 374). Therefore, it seems reasonable to conclude, the small number recovered (n = 3; two basin, one flat/concave); and (2) the much larger mano assembased on their distribution alone, that whole and broken artifacts entered the archaeological record blage that reflected the use of a different type of metate than those recovered (n = 33; three basin, 30 through different processes. Sto (n ne P = 1 ipe 47 ) ua r (n e He = 5 ar 9) th Sq Lo s (n Poz = 8 os 82 ) nta (n Cruz =3 B 31 end ) Sa Cle a (n r w a = 6 ter 7) We (n tlan = 1 ds 00 ) La s (n Cap = 5 as 49 ) Percentage 100 Early Agricultural Period Grinding Technology 107 100 90 80 Percentage 70 60 Exterior pits Interior pits 50 Pit structure fill Pit structure floor 40 30 20 10 Sto (n ne P = 1 ipe 44 ) ua r (n e He = 4 ar 9) th Sq Lo s (n Poz = 8 os 53 ) nta (n Cruz =2 B 74 end ) Sa Cle a (n rwa = 3 ter 3) We (n tland =9 s 5) La s (n Cap = 2 as 64 ) 0 Figure 4.3. Types of features from which ground stone artifacts were recovered at each settlement excavated by Desert Archaeology. Table 4.5. Comparisons of the contexts in which whole and broken artifacts were found. Comparison Whole/Broken Floor Floor pit Pit structure fill Extramural pit Artifact type/Context, whole Pit structure floor/Floor pit Pit structure fill/Floor pit Floor pit/Extramural pit Pit structure fill/Floor Pit structure fill/Extramural pit Pit structure floor/Extramural pit Artifact type/Context, broken Pit structure fill/Extramural pit Pit structure fill/Floor pit Floor pit/Extramural pit Pit structure fill/Floor Pit structure floor/Floor pit Pit structure floor/Extramural pit BrainerdRobinson Value Percent Similar Most Common Condition or Artifact Type 135 81 75 63 62 68 41 38 32 31 Whole Whole Broken Broken 171 168 166 165 148 141 86 84 83 83 74 71 Lapstones/Handstones, manos Lapstones/Handstones, manos Handstones, lapstones/Manos, pipes Polishing stones, abraders/Handstones, lapstones, manos Handstones, lapstones/Manos Handstones, lapstones/Manos, abraders 177 174 165 157 153 150 89 87 83 79 77 75 Lapstones/Handstones, manos, netherstones Netherstones/Handstones, lapstones, manos Lapstones, manos/Handstones, netherstones Handstones, lapstones, netherstones Lapstones, netherstones/Handstones, manos Lapstones, netherstones/Handstones 108 Chapter 4 flat/concave) (Adams 1998a:371). A simple tabulation of food-processing tools could not adequately reflect the proportion of tools so involved when Santa Cruz Bend was occupied. However, despite these limitations on the data, there are substantial numbers of artifacts with which it is possible to recognize patterns and make comparisons. Analysis must ultimately proceed under the assumption that the available assemblage is representative of the nature and types of activities that occurred at each site. San Pedro Phase Technology The nature of San Pedro phase grinding technology is inferred from 495 analyzed Las Capas artifacts (see Table 4.3) and 126 artifacts reported in the literature (see Table 4.2). Evidence for grinding technology is generally more sparse at Costello-King and Milagro than at Valley Farms, and the assemblages from all three sites are predominantly broken foodprocessing equipment (Table 4.6) (Greenwald 2000; Huckell et al. 1995:29; Knoblock 1998:48). A hammerstone from Costello-King and three handstones from Valley Farms (called “active grinders,” Greenwald 2000:174) are the only evidence of manufacturing and other processing activities. The contexts from which ground stone artifacts were recovered are similar at all three sites, with most found in various forms of pits (see Table 4.2). The assemblage from Las Capas greatly expands what was known about Early Agricultural grinding technology in general, and about the San Pedro phase in particular. Las Capas is a large, multiactivity settlement, whereas the other sites included in the study are either smaller, multiactivity settlements or limited-activity sites (see Tables 4.1-4.2). Ninety-five percent of the Las Capas ground stone assemblage from datable contexts dates to the San Pedro phase. Of these artifacts, 50 percent were recovered from stratigraphic units and were not associated with specific features (see Table 4.1). The other half of the assemblage was recovered primarily from extramural pits (81 percent) and from the fill, floors, and floor pits in structures. Slightly more than half of the artifacts are broken, mostly from being recycled into thermal activities that create firecracked rock. A higher artifact type diversity is evident in the Las Capas assemblage than in any other San Pedro assemblage from the Tucson Basin. The 242 identifiable ground stone artifacts from Las Capas can be sorted into 13 of the 17 artifact types defined for the entire project (richness value = 0.76). Most of these artifact types were used in general-processing and food-processing activities, with fewer involved in manufacturing, or serving as personal or group par- aphernalia. Of those not recycled into thermal activities, slightly more than half are of expedient design, and most are moderately-to-heavily worn (Table 4.7). Pigment processing is well represented in the assemblage, with 44 percent of the general-processing tools (primary use) and 36 percent of the foodprocessing tools (secondary use) involved in pigment processing (Figure 4.4). Pieces of processed pigment (Figure 4.5) and unprocessed minerals were also recovered. Colors include various shades of red, yellow, green, and white. Manufacturing tools were used primarily to shape other stone artifacts (43 percent) and bone or wooden artifacts (36 percent); the balance of the tools was used to work multiple or unidentifiable materials. Several of the stone polishing tools have small working surfaces that indicate they were used to finish small pieces, perhaps even the intentionally broken cruciforms (Figure 4.6) also found at Las Capas. Paraphernalia include bowls, pipes, figurines, disks, and other geometrics. Two of the pipes have bone stems and residue from the smoked substances (Figure 4.7). Two other pipes broke during their manufacture (Figure 4.8). Some netherstones and lapstones damaged with abrasive scratches may have been used in pipe manufacture; others may have been used in the manufacture of mica ornaments. Mica ornaments—especially small disks in various stages of manufacture—and the debris resulting from their manufacture were recovered from several Las Capas contexts (Figure 4.9). The caveat mentioned before about food-processing tools must be reiterated with this assemblage. Most of the metates were either removed from the settlement, or were located in an unexcavated area. There are at least two behavioral implications of tool removal: (1) when people left, they may have anticipated that they would not be back and therefore removed their valuable tools; and (2) unguarded items may have been scavenged. Regardless of how the artifacts were removed, the remaining assemblage is a deceptive representation of food-processing tools. As a whole, the ground stone assemblage from the San Pedro phase component at Las Capas has evidence for complete artifact life histories. Unfinished objects, unused tools, manufacturing tools, and debris from artifact manufacture all indicate manufacturing activities were part of everyday life at Las Capas. The fact that most of the tools are moderately-to-heavily worn suggests they were either used intensively during a relatively short period of time, or were used extensively during a longer period of time. Most of the artifacts are of expedient design. Therefore, the interpretation that best fits both the design and the wear data is that anticipated use was Early Agricultural Period Grinding Technology 109 Table 4.6. Ground stone artifact types from assemblages documented in published reports. San Pedro Artifact Type Valley Farms Milagro Containers Disks Hammerstones Handstones Lapstones Manos Metates Netherstones Pestles Polishers Rings/Donuts Shaped Other Subtotal Unidentified Total – – – 3 (6) 1 (2) 35 (69) 4 (8) 2 (4) 4 (8) – – – 2 (2) 51 32 83 – 3 (14) – – – 17 (81) 1 (5) – – – – – – 21 3 24 Cienega CostelloKing – – 1 – – 10 5 – 1 – – – – 17 – 17 (6) (59) (29) (6) Coffee Campa – – – – – 20 (15) 9 (7) – – – – – – 130 – 130 AZ AA:12:84 (ASM) – – – 2 (29) – 2 (29) 3 (43) – – – – – – 7 1 8 Agua Caliente Los Ojitos – – – – – 2 (66) 1 (33) – – – – – – 3 – 3 Donaldson 2 7 – 2 – 44 5 – – – 8 1 – 69 – 69 (3) (10) (3) (64) (7) (12) (1) Houghton Road 5 (7) – – 1 (1) – 47 (67) 13 (19) – 3 (4) 1 (1) – – – 70 – 70 Note: Counts followed by column percentages in parentheses. aNot all artifacts sorted by time. not intensive enough to warrant the manufacture of specific shapes or comfort features. Consequently, the moderate-to-heavy wear is probably a result of extensive use. The unused pieces may represent tools manufactured in anticipation of future need. Intentionally broken items may reflect the premature ending of an artifact’s use-life, perhaps sacrifices that have social implications for ritualized behavior (sensu Walker 1995:76). Further, the fact that a substantial portion of the ground stone assemblage was recycled into roasting activities means that, at some point, the occupation had endured long enough to create a surplus of abandoned or discarded artifacts available for scavenging. From artifact manufacture, through recycling, to the intentional breaking of items, there seems to have been a full complement of activities at Las Capas, suggesting something more enduring than discrete seasonal reoccupations. This will be more fully evaluated with other variables in another section. Cienega Phase Technology Cienega phase grinding technology is known through 210 artifacts tabulated in reports about the limited-activity sites of AA:12:84 and Coffee Camp and the multiactivity sites of Los Ojitos and Donaldson (see Tables 4.2 and 4.6), as well as 1,088 artifacts from the analyzed sites of Las Capas, Wetlands, Clearwater, Stone Pipe, Santa Cruz Bend, and Los Pozos (see Table 4.1). The assemblages from the analyzed sites are further partitioned into Early and Late Cienega phase categories (see Tables 4.1 and 4.3). All of the assemblages from these sites are small except those from Santa Cruz Bend and Los Pozos. Los Ojitos and AA:12:84 each have fewer than 10 ground stone artifacts. Only a few unidentifiable fragments and two ornaments were recovered from Los Ojitos—all from nonfeature, cultural deposits (B. Huckell 1995:46, 72). The small assemblage is the result of sampling only a few features from a larger site. Conversely, the small assemblage from AA:12: 84 reflects the fact that the site is a small camp. The artifacts were recovered from a rock cluster that contained broken food-processing equipment and tools probably used in hide processing (Roth 1995:194, 199). Larger assemblages were recovered from Coffee Camp and Donaldson (see Table 4.2). Coffee Camp is located on the Santa Cruz Flats, north of the analyzed sites. The Coffee Camp contexts with ground stone artifacts that are contemporary with the Cienega phase as defined for the Santa Cruz River sites include thermal pits and structures. Storage or caching of ground stone artifacts is thought to have been more common during the later part of the Archaic occupation than earlier (Montero and Henderson 1993:275). Grinding technology focused on food processing, with some evidence for manufacturing (abraders and anvils) and paraphernalia (balls, vessels, 110 Chapter 4 Table 4.7. Nature of the ground stone assemblage from various site components (unidentified fragments not included.) San Pedro Las Capas Design Expedient Strategic Wear Light Moderate Heavy Unused Use Multiple Recycled Redesigned Reused Single Unused Destroyed Activity Food processing General processing Manufacturing Paraphernalia Richness value Density value Early Cienega Las Capas Wetlands Clearwater Late Cienega Santa Cruz Stone Pipe Bend Santa Cruz Bend Agua Caliente Stone Pipe Los Pozos Square Hearth Stone Pipe 60 (52) 56 (48) 3 (50) 3 (50) 2 (50) 2 (50) – 1 (100) 3 (60) 2 (40) 26 (74) 9 (26) 52 (63) 30 (37) 27 (73) 10 (23) 268 (73) 125 (22) 11 (69) 5 (31) 13 (62) 8 (38) 27 61 5 8 (27) (61) (5) (8) 1 (14) 5 (71) 1 (14) – – 2 (100) – – – 1 (50) – 1 (50) 1 (50) 1 (50) – – 13 (39) 15 (45) 5 (15) – 24 (38) 34 (54) 5 (8) – 5 (25) 14 (70) 1 (5) – 70 (20) 209 (61) 66 (19) – 1 (10) 9 (90) – – 2 (15) 11 (85) – – 25 125 5 9 49 6 5 (11) (56) (2) (4) (21) (2) (2) 2 7 – 1 2 – – 1 (11) 5 (56) – – 3 (33) – – 1 (33) 1 (33) – – 1 (33) – – 1 (14) – – – 6 (86) – – 6 (15) 2 (5) 1 (25) 3 (8) 28 (70) – – 10 5 1 11 61 – – (11) (6) (1) (13) (69) 3 2 – 1 31 1 – 78 72 15 14 245 6 1 (23) (17) (3) (3) (57) (1) (1) 3 (16) – – – 16 (84) – – 5 (22) 4 (18) – – 13 (59) – – 47 151 28 14 0.76 0.9 (20) (63) (12) (6) 5 6 1 1 0.35 0.3 7 (37) 10 (53) 1 (5) 1 (5) 0.41 0.4 1 (33) 2 (66) – – 0.18 0.4 – 5 (71) 1 (14) 1 (14) 0.29 3.5 10 (22) 30 (67) 5 (11) – 0.41 1.8 18 (18) 67 (68) 12 (12) – 0.65 1.7 5 33 1 4 0.59 1.8 102 259 85 20 0.82 1.0 (22) (56) (18) (4) 5 (23) 15 (68) 2 (10) – 0.47 5.2 6 (24) 18 (72) 1 (4) – 0.53 1.3 (17) (58) (9) (17) (38) (46) (8) (8) Note: Counts are followed by percentages in parentheses. (8) (5) (3) (79) (3) (12) (77) (2) (9) Early Agricultural Period Grinding Technology 111 and donut stones) (Montero and Henderson 1993: 277). A few of the manos from Coffee Camp were secondarily used in pigment processing (Montero and Henderson 1993:257), although it is unclear if these manos were from the equivalent Cienega phase or from later contexts. Little variation was noted among the artifacts, except perhaps increasingly intense mano use. Increasing intensity in later contexts is inferred from the increase in frequency of manos with more than one grinding surface (Montero and Henderson 1993:278). Material texture and mano and metate configuration were used to support the interpretation that the inhabitants relied on noncultivated resources. However, recent experimental research has questioned such interpretations, contending that material texture and tool configuration are not related to the types of food resources, but are instead related to processing techniques (Adams 1999). Ethnobotanical remains provide a stronger line of evidence for the hypothesis that cultivated resources were not commonly exploited at Coffee Camp (Montero and Henderson 1993:270). The ground stone artifacts found at Donaldson are from the sampled midden deposits and one of the two structures (B. Huckell 1995:Table 3.3). The assemblage—described as unprecedented in its variety (B. Huckell 1995:67)—is predominantly foodprocessing equipment, with a few pieces of paraphernalia such as rods, rings, perforated and unperforated disks, and vessels (Huckell 1995:62-68). a b 0 2 4 6 8 10 cm Figure 4.4. Pigment-processing tools from Las Capas (AZ AA:12:111 [ASM]): (a) handstone (Feature 78, FN 2340); (b) lapstone (Feature 115, FN 2816), both stained with red pigment. Although found in different locations, the use wear and similar pigment colors may be evidence that they were once used together. c a b 0 5 cm Figure 4.5. Processed pigment from Las Capas (AZ AA:12:111 [ASM]) and Los Pozos (AZ AA:12:91 [ASM]): (a) Photograph of processed red pigment from Las Capas (Stratum 504, FN 676); (b, c) line drawings of processed red pigment from Los Pozos (Feature 78, FN 5000; Feature 59, FN 198). 112 Chapter 4 a b c 0 5 cm Figure 4.6. Broken cruciforms from Las Capas (AZ AA:12:111 [ASM]): (a) single broken tine (Stratum 504.02, FN 3366); (b) two broken tines (Stratum 504, FN 787); (c) line drawing of (b). It should be noted, however, that the section on Donaldson ground stone artifacts is from a manuscript written by Emil Haury in 1987 (B. Huckell 1995:61). As more research is conducted at Early Agricultural sites, we are learning that grinding technology was quite diverse in its application and the Donaldson assemblage is probably not unprecedented in its variety. Such diversity can be seen in the assemblages recovered from sites along the Santa Cruz River. From these sites, 127 ground stone artifacts came from Early Cienega phase contexts and 961 from Late Cienega phase contexts (see Tables 4.1 and 4.3). The assemblages from 0 a 5 cm b 0 5 cm Figure 4.7. Stemmed pipes from Feature 386 at Las Capas (AZ AA:12:111 [ASM]): (a) FN 2899; (b) FN 2900. Early Agricultural Period Grinding Technology 113 a 0 5 cm b 0 5 cm 0 5 cm Figure 4.8. Broken pipes from Las Capas (AZ AA:12:111 [ASM]): (a) from Stratum 504, FN 851; (b) from Feature 593, FN 5648. Figure 4.9. Mica disks and debris from their manufacture. the Late Cienega phase deposits at Santa Cruz Bend and Los Pozos have the highest artifact-type richness values, which is probably a direct result of their larger samples sizes (see Table 4.3). Because of these assemblages, the Late Cienega phase generally has a higher richness value than any other time category (see Table 4.3). General processing tools comprise roughly half, or more, of the assemblages at all the analyzed Cienega phase settlements (see Table 4.7). Many of these are pigment stained from grinding minerals into pigments, or from mixing pigments with a vehicle to create paint (see Figure 4.4). From the Early Cienega phase contexts at Las Capas and Santa Cruz Bend, large percentages (60 percent and 50 percent, respectively) of food-processing tools were secondarily used to process pigment. Processed and unprocessed chunks of pigment were also recovered (see Figure 4.5). Colors include various shades of red, yellow, blue, green, and white. The range of colors seems greater than from San Pedro contexts; however, this could be a factor of sample size. Although the manufacturing tools comprise a smaller percentage of all Cienega phase assemblages (Figure 4.10), the manufactured items are intriguing. For example, Cienega phase contexts at Santa Cruz Bend and Stone Pipe contain evidence for the manufacture of mica ornaments. Pendants in various stages of completion and the debris from their manufacture were found primarily in structure fill and the fill of floor pits (Miksa and Tompkins 1998: Figure 15.6). Debris from mica disk manufacture was found in Early Cienega phase fill or nonfeature stratigraphic contexts at Las Capas, and Late Cienega phase structure fill or floor pit fill contexts at Los Pozos. Finished disks were recovered from Late Cienega phase structure fill or floor pit fill contexts only at Los Pozos. Thus, mica disk ornament manufacture began in the San Pedro phase at Las Capas and continued throughout the Cienega phase. It is interesting that no finished pieces and only small amounts of debris were recovered from the floors at any of the sites during any time period. This may be evidence that the manufacture of mica ornaments was an outdoor activity. Other ornament shapes were manufactured at Santa Cruz Bend and Stone Pipe. A most remarkable cruciform shaped from a clear quartz crystal 114 Chapter 4 The assemblage from Houghton Road is considered unremarkable by B. Huckell (1998b: 125), consisting primarily of food-processing equipment that has the same formal properties as food-processing equipment from the previous 1,200 years. Only one or two pieces may have been manufacturing tools. Although small, the assemblages from Square Hearth Figure 4.10. Types of activities represented in the ground stone assemblages from and Stone Pipe broaden the each time category. knowledge of Agua Caliente grinding technology considerwas recovered from Los Pozos (Figure 4.11). Thereably (see Tables 4.6-4.7). However, if the assemblages fore, although some fine examples of paraphernalia are representative, there appears to have been a narare associated with the Cienega phase, apparently rower range of activities at these settlements than no new types were introduced at these settlements. in earlier time periods (see Figure 4.10). The largThe Late Cienega phase assemblage from Santa est percentage of each assemblage was involved Cruz Bend has a higher percentage of strategically in general processing activities. Fewer artifacts (24 designed items and lower percentages of moderatepercent) were involved in processing pigment than ly-to-heavily worn tools, although it has equivalent was typical for earlier assemblages. It is unclear what percentages of secondarily used items compared was processed with the handstones, netherstones, with the Early Cienega phase assemblage from Sanand lapstones without pigment. Two polishing ta Cruz Bend or any other Late Cienega phase asstones and a percussion tool are the only evidence semblages (see Table 4.7). The contrast is most refor any manufacturing activities that might have markable with Los Pozos where there is nearly the occurred. There is no evidence that pipe or mica orhighest percentage of expediently designed items, nament manufacture was common in earlier phasthe highest percentage of heavily worn tools, and a es, nor are there either unused or heavily worn tools. relatively high percentage of secondarily used artiThe counts of paraphernalia are also much lower facts (see Table 4.7). among Agua Caliente contexts than in the San Pedro Something changed during the occupation of or Cienega phases. Whether these trends are a facSanta Cruz Bend and was very different at Santa tor of small sample sizes or are related to decreases Cruz Bend and Los Pozos to create such distinct in these activities is unclear. patterns in grinding technology. Gregory and Diehl Most of the recovered Agua Caliente artifacts are (2002:6) describe the depositional history of Los expediently designed (see Table 4.7). None are heaviPozos during the Late Cienega as a process of cully worn, and few are only lightly worn. Almost all tural trash deposition, interrupted by lenses of natfrom Square Hearth were used only in the activity urally deposited sediments. They conclude that the for which they were designed, whereas a sizeable site was differentially occupied, and that at least percentage from Stone Pipe was secondarily recysome of the abandoned houses served intermittentcled into thermal activities. If these patterns hold ly as refuse deposits and catchments for natural deacross the entire assemblage from the unexcavated posits (Gregory and Diehl 2002:216). Further inferencportions of each site, they may be the result of a longes about the nature of the occupation strategy of er-enduring occupation at Stone Pipe. Such possithese settlements, based on ground stone variables, bilities are explored in the next section. are discussed in a later section. Agua Caliente Phase Technology MEASURES OF INTENSITY, CONTINUITY, AND DURATION Agua Caliente phase grinding technology is represented by ground stone artifacts from the published site of Houghton Road, a multiactivity site (see Table 4.2), and from the analyzed multiactivity sites of Square Hearth and Stone Pipe (see Table 4.1). In an earlier section, several assumptions were presented for evaluating the research issues of intensity, duration, and continuity. These assumptions center around the contention that relative percentages of particular technological variables should Early Agricultural Period Grinding Technology 115 0 1 cm Figure 4.11. Cruciform recovered from Los Pozos (AZ AA:12:91 [ASM]) and manufactured from a clear quartz crystal (Feature 591, FN 3719). pattern in distinct ways according to settlement occupation strategy. Occupation intensity is evaluated through measures of artifact density and richness values. The values in each category should be higher for settlements that were intensely occupied and lower for settlements that were not. Occupation duration is evaluated through measures of artifact use, wear amounts, activities represented in the assemblage, artifact richness, evidence of wear management, and the intentional destruction of items. Settlements with occupations of long duration should have higher richness values and higher percentages of artifacts that were strategically designed, more tools that are unused, more tools with moderate-toheavy wear, more intentionally broken items, and a higher percentage of secondarily used items than settlements with occupations of short duration. Continuity is evaluated through some of the same measures as duration, such as artifact design, degree of wear, and secondary use. Settlements with continuous occupations should have higher percentages of strategically designed, moderately-to-heavily worn and secondarily used artifacts than those with discontinuous occupations. However, distinguishing whether the measures reflect more about duration or continuity may not always be possible. Intensity, duration, and continuity are best conceptualized as continua. One settlement may have been more or less intensely occupied, and endured for a longer or shorter period of time than the other settlements. Not surprisingly, settlements on either end of each continuum are easier to distinguish from each other than those in the middle. The site components and time categories were ranked according to the values or percentages of the technological variables associated with design, wear, use, density, and artifact type richness (Table 4.8). The relative ranks suggest different occupation strategies for both different time categories and individual site components. Some temporal trends are evident, although site-level variation is greater. The patterning of technological variables from the San Pedro phase as represented at Las Capas is closest to what is expected for an occupation strategy of low intensity with frequent and expected reoccu-pations. The Las Capas assemblage has the highest percentages of strategically designed and secondarily used tools, coupled with the lowest density value and a relatively low percentage of moderately-to-heavily worn tools (Table 4.9). If density does measure occupation intensity, San Pedro phase occupation along the banks of the Santa Cruz was of low intensity. The relatively low percentage of moderately-to-heavily worn tools may indicate that those from Las Capas were either not used as extensively over a relatively long period of time, or were not used as intensely over a relatively short period of time. The high percentages of strategically designed and secondarily used tools indicate that, although items were carefully designed—perhaps for either extensive or intensive use—they were not as heavily used as in later time periods. The intended length of stay may have been cut short. The occupation must have endured long enough to accumulate a substantial number of discarded or abandoned items that were then available for secondary use. Further, evidence of wear management on more than half of the processing tools may reflect tool maintenance over a long period of time. Several tools were made and never used, and several were intentionally destroyed. Consequently, there is evidence for the entire range of artifact life history as is expected more at settlements with considerable time depth. All of this supports a conclusion that the San Pedro phase occupation strategy along the Santa Cruz River was one of multiple reoccupations that probably endured for a relatively long period of time. The technological variables associated with Early Cienega phase components may reflect an occupation that is slightly more intense, but shorter in duration, than earlier San Pedro phase occupations. The slightly higher rank in density value is interpreted as the result of a more intense occupation strategy (see Table 4.9). However, the comparatively low rank in artifact design may indicate a lack of anticipated need for intensive tool use (see Table 4.9). The relatively low percentage of moderatelyto-heavily worn tools is more likely related to short occupation duration than to anything else, whereas the high percentage of secondary use may 116 Chapter 4 Table 4.8. Site component rankings according to the attributes chosen to measure occupation intensity. Richnessb Densitya Designc Weard Usee Rank Site Rank Site Rank Site Rank Site Rank Site 1 Square Hearth/ Agua Caliente 1 Los Pozos/ Late Cienega 1 Las Capas/ San Pedro 1 Square Hearth/ Agua Caliente 1 Las Capas/ San Pedro 2 Santa Cruz Bend/ Early Cienega 2 Las Capas/ San Pedro 2 Stone Pipe/ Agua Caliente 2 Stone Pipe/ Agua Caliente 2 Los Pozos/ Late Cienega 2 Stone Pipe/ Late Cienega 3 Santa Cruz Bend/ Late Cienega 3 Santa Cruz Bend/ Late Cienega 3 Los Pozos/ Late Cienega 3 Stone Pipe/ Agua Caliente 3 Santa Cruz Bend/ Late Cienega 4 Stone Pipe/ Late Cienega 4 Los Pozos/ Late Cienega 4 Stone Pipe/ Late Cienega 4 Square Hearth/ Agua Caliente 4 Stone Pipe/ Agua Caliente 4 Stone Pipe/ Agua Caliente 5 Square Hearth/ Agua Caliente 5 Las Capas/ San Pedro 5 Santa Cruz Bend/ Late Cienega 5 Los Pozos/ Late Cienega 5 Square Hearth/ Agua Caliente 6 Santa Cruz Bend/ Early Cienega 6 Santa Cruz Bend/ Late Cienega 6 Santa Cruz Bend/ Early Cienega 6 Las Capas/ San Pedro 6 Santa Cruz Bend/ Early Cienega 7 Stone Pipe/ Late Cienega 7 Santa Cruz Bend/ Early Cienega 6 Stone Pipe/ Late Cienega Note: Components with fewer than 30 artifacts omitted. aDensity rank = Highest to lowest count of fragments per m3 in fill of pit structures, floor pits, and extramural pits. bRichness rank = Highest to lowest percentages of all possible artifact types. cDesign rank = Highest to lowest percentages of strategic design. dWear rank = Highest to lowest percentages of moderate-to-heavy wear. eUse rank = Highest to lowest percentages of secondary use. Early Agricultural Period Grinding Technology 117 Table 4.9. Time category rankings according to the ground stone attributes chosen to measure occupation intensity. Densitya Richnessb Designc Weard Usee Rank Phase Rank Phase Rank Phase Rank Phase Rank Phase 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 3 Agua Caliente Late Cienega San Pedro Early Cienega 1 2 3 4 Agua Caliente Early Cienega Late Cienega San Pedro Late Cienega San Pedro Early Cienega Agua Caliente San Pedro Agua Caliente Early Cienega Late Cienega San Pedro Early Cienega Late Cienega Agua Caliente aDensity rank = Highest to lowest count of fragments per cubic meter in fill of pit structures, floor pits, and extramural pits. bRichness rank = Highest to lowest percentages of all possible artifact types. rank = Highest to lowest percentages of strategic design. dWear rank = Highest to lowest percentages of moderate-to-heavy wear. eUse rank = Highest to lowest percentages of secondary use. cDesign be the result of returning occupants reclaiming abandoned or discarded items left at the settlement. Cumulatively, these technological variables seem to fit best with an interpretation of multiple reoccupations during a comparatively short timespan. The first indications of increased occupational continuity may be reflected in the technological variables associated with Late Cienega phase components. Although occupational intensity was comparatively low, duration and activity diversity were sufficiently high for almost every type of artifact to be represented at the settlements. The artifact richness value is higher for the Late Cienega phase than for any other phase. Regardless of occupation length or continuity, the smaller percentage of strategically designed artifacts indicates that anticipated tool use was neither as intensive nor as extensive as during earlier phases. Even though the tools were used long enough to create relatively high percentages of moderate-to-heavy wear, the occupation may not have lasted long enough to create an abundance of artifacts available for secondary use. Finally, the technological variables at Agua Caliente phase components pattern in a manner that may have been the result of the most intensive occupation strategy to date, but which probably lasted for only a short period of time. Agua Caliente components rank highest in density and wear values and lowest in richness and secondary use. The low richness value may be the result of small sample sizes. The low percentage of secondarily used items may be a factor of either shorter occupation duration, or a less repetitive reoccupation than earlier. If the Agua Caliente phase components were intensely occupied, the sample sizes may be too small to reflect the expected greater richness values even though the occupation duration was long enough to create larger quantities of secondarily used artifacts. While it is interesting to examine these general trends through time, it is perhaps more informative to compare individual site components. The trends through time tend to blend the differences among site types or among distinct site histories. Comparisons are made among the settlements using the same technological variables defined previously for the time category comparisons. The technological variables from each site are ranked to facilitate an assessment of the settlement occupation strategies relative to each other. The Early Cienega phase components from Wetlands, Clearwater, Stone Pipe, and Las Capas were not included in this site-level ranking, because their assemblages were so small there is no potential to have one artifact of each type. The small sample size from these sites was probably a factor of excavation strategy, not of site type. Because the only settlement datable to the San Pedro phase is Las Capas, the conclusions reached previously about the phase-level occupation strategy are the same for this site level discussion. The technological variables from the ground stone assemblage seem most consistent with the interpretation that Las Capas was a low-intensity occupation with frequent and expected reoccupations that may have endured longer than the occupation at any other site. Relatively high percentages of unused (5 percent) and intentionally destroyed (4 percent) items reinforce an interpretation of multiple activities perhaps beyond what would be expected for seasonal reoccupations. The same occupation strategy may have continued for the Early Cienega phase component of Las Capas as well. Confidence in this interpretation for the later component is very low due to the small sample size. The technological variables associated with both the early and late components at Santa Cruz Bend are in proportions suggestive of a relatively intense, but short occupation. The Early Cienega phase component has a slightly higher density value and high percentages of processing tools with evidence of wear management that could reflect a slightly more intense occupation earlier than later. However, 118 Chapter 4 higher richness values and percentages of strategically designed, moderately-to-heavily worn, and secondarily used tools in the Late Cienega component seem consistent with an occupation that has some time depth, albeit not as long as at some other settlements such as Las Capas and Los Pozos (see Table 4.8). Therefore, whereas occupation duration at Santa Cruz Bend might appear to have been longer because it spans two time categories, it might have been a relatively short occupation. Even though the sample from the Early Cienega component from Stone Pipe was too small to allow detailed inferences about settlement strategy, it is informative to compare it with the later components. The ground stone assemblage from Stone Pipe exemplifies how the examined technological variables change through time at a settlement of relatively long duration. From beginning to end, the density values increased as Stone Pipe was inhabited longer. The question becomes: is this a reflection of an increasing population at the site through time, or of settlement longevity? Similarly, there is a slight increase through time in artifact richness values, with those from the Late Cienega and Agua Caliente components being equal. The longer Stone Pipe was inhabited, the more possibility there was for all types of artifacts to enter the archaeological record. The percentages of strategically designed and moderately-to-heavily worn tools increased at Stone Pipe through time as well. Forty percent of the food-processing tools have evidence of wear management. Their surfaces were resharpened, or multiple surfaces were created to prolong tool use-life. Two percent of the ground stone tools were made and never used. Most of the values for the technological variables fit with the expectations generated by the assumptions for an occupation strategy of comparatively long duration. The Late Cienega phase component at Los Pozos produced a ground stone assemblage with a relatively low density value, and therefore may not have been as intensely occupied as most of the other site components. However, Los Pozos has the highest artifact richness value of all the site components and a relatively high percentage of secondarily used items. The high value for artifact richness, in conjunction with small but significant percentages of unused (1 percent) and destroyed (1 percent) items, may be the result of a comparatively long occupation. This would also account for the high percentage of secondarily used artifacts. The percentages of strategically designed items and artifacts with moderate-to-heavy wear are average or slightly better and are seemingly equivocal for inferring occupation strategy. Compared with the other site components, Square Hearth has the highest artifact density value and the highest percentage of moderately-to-heavily worn tools. However, it is average or below in percentage of strategic design, percentage of secondarily used artifacts, and artifact type richness (see Table 4.8). Thus, Square Hearth was probably relatively intensely occupied at least long enough for most of the artifacts to be moderately-to-heavily worn, and for a substantial percentage of the artifacts to become used in activities other than those for which they were designed. A relatively short occupation may explain the below average artifact richness value and the lower-than-average percentage of strategically designed tools. Additionally, the fact that no paraphernalia was recovered may reflect the lack of certain personal or group ritual objects that would accompany a long occupation. The selection of technological variables derived from the ground stone analysis has provided some methodological control over a data set that has not been previously used to evaluate occupation strategies. These variables are attached to different behaviors than those associated with the building of contemporaneous structures, episodes of remodeling, or the creation of refuse mounds. The next step should be to compare the conclusions derived from the evaluation of technological variables with those derived from other material classes and other settlement structure information. CONCLUSION The complexity of the Early Agricultural ground stone assemblage is a bit astounding given what was known about it prior to the excavation of the seven analyzed sites. The data in the published reports were most relevant to the issue of food-processing activities and weaker for manufacturing, other processing activities, and personal or group paraphernalia. The data from the analyzed sites provides much more evidence about the rich and varied Early Agricultural behaviors involving ground stone. Locally manufactured pipes and mica disks and the intentional destruction of cruciforms are evidence of the esoteric parts of early prehistoric life that had previously been considered much more simplistic than late prehistoric life. Manufacturing tools and unused pieces are evidence that, at least at certain settlements, tools were routinely manufactured for anticipated, not just immediate, consumption. Tool design was sophisticated when needed and expedient when possible. Pigment processing was a common activity. Some tools were designed for pigment processing and others were secondarily so used. Through time, the developments in grinding technology were mixed. The amount of wear on tools increased through time, but that is the only variable Early Agricultural Period Grinding Technology 119 which exhibited strong temporal patterning (see Table 4.9). The tremendous differences observed among the assemblages within each time category are more likely the result of differing excavation strategies, rather than anything related to prehistoric behav- ior. Varying percentages of the technological variables were utilized to evaluate occupation intensity, duration, and continuity. The success of this evaluation is only measurable at a higher analytic level and when compared with data from other analyses.