The real wealth of a planet is in its landscape, how we take part in that basic source of civilization—agriculture.
—Frank Herbert, Dune
The transition from foraging to agricultural-based economies was one of the most significant processes to occur in human history (Harris 1996; Matson 1991; Smith 1998). Yet it did not occur in all parts of the world. This was the case across the Borderlands between the United States and Mexico, an area stretching from the Gulf of California to the Gulf of Mexico. Although the development of Southwestern culture was based on a foundation of maize agriculture, nomadic foragers ranged the adjacent Tamaulipas regions of South Texas and northeastern Mexico. Recent discoveries of Late Archaic villages in southern Arizona and terrace hilltop communities in the deserts of Chihuahua, Mexico, have radically changed our view of this period. This volume explores these varied archaeological records and current research of the Borderlands Late Archaic. Understanding why foragers in South Texas failed to incorporate cultigens into their subsistence base may aid Southwestern researchers in understanding why agriculture became an important part of this desert economy.
The Borderlands cover a linear distance of approximately 1,600 km (1,000 miles) (Figure 1.1). On the west lie the arid creosote-covered lands of the Sonoran and Chihuahua Deserts, and on the east the brush country of the Tamaulipas. Separating these two areas is the Trans-Pecos transitional zone, with the Plateau and Plains to the north and subtropical regions to the south. The region is characterized by a general increase in effective moisture from west to east and decreases in seasonal temperatures with increasing seasonality from south to north. The rainfall regime changes dramatically, with summer monsoons in the west versus a bimodal pattern with a midsummer low in the east. The landscape also varies, with the broad river valleys of the Basin and Range in the west and a gently rolling topography dissected by stream channels in the east (Blair 1950; Brown 1994; Norwine 1995). The Borderlands therefore provide a setting within which the Late Archaic (ca. 3000 to 1500 BP) is characterized by a diverse set of agricultural and foraging strategies.
The concept of the Archaic is generally characterized as a post-Pleistocene mixed hunting and gathering economy, with the possible addition of cultigens during the Late Archaic. It is primarily differentiated from the earlier Paleoindian period by the presence of distinctly shouldered and notched dart points, more generalized retouched tools, one-hand manos, slab milling stones, and fire-cracked rock features. It is separated from later periods by the initial use of the bow and arrow and ceramics. It is these technological innovations that characterize the Southwestern Ceramic period and the Late Prehistoric in South Texas (Hester 1995; Huckell 1996a; Matson 1991; Vierra 1994c). Figure 1.2 illustrates the various sequences proposed for the Borderlands Archaic, as derived from Bruce Huckell (1996a); Robert Mallouf (1985, 1992, this volume); Solveig Turpin (1995); and Thomas R. Hester (1995, this volume). They all generally begin by about 8500 BP during the middle Holocene, although the Late Archaic appears to terminate later in the eastern Borderlands due to the absence of agriculture (ca. 1100 BP).
Early maize dates in the Borderlands and Greater Southwest cluster about 3000 BP, with several earlier dates mostly from Arizona (Gilpin 1994; Hard and Roney 1998, this volume; Huckell 1990; Huckell et al. 1999; Mabry 1999, this volume; Simmons 1982; Smiley 1994; Tagg 1996, 1999; Upham et al. 1987; Wills 1985). The earliest date for maize is 3690 BP from McEuen Cave in southeastern Arizona (Huckell et al. 1999; Shackley et al. 2001). Mabry (2002) notes in a recent study that there are twenty Southwestern maize dates older than 3000 BP, indicating the possible arrival of maize by ca. 3700 BP. The current evidence reflects that when maize entered the Southwest its use spread quite rapidly. It is debated, however, as to whether its spread was due to the northern movement of farmers (Berry 1982, 1999; Berry and Berry 1986), the integration of these cultigens into local hunting and gathering economies (Hogan 1994; Irwin-Williams 1973; Minnis 1992; Vierra 1994a, 1994b, 1996; Wills 1985, 1995), or a mixture of the two (Matson 1999, 2001, this volume). So the question is not so much when maize arrived, but rather how long it took for these farmers to move north or for these foragers to become dependent on agriculture. Importantly, the eastern boundary for agriculture appears to lie between the Chihuahuan Desert and Tamaulipas in the Trans-Pecos region, with no evidence for agriculture in South Texas. Directly dated maize specimens from in situ deposits are lacking for the eastern Borderland areas (Dering, this volume; Mallouf, this volume). The original source for maize agriculture is located to the south in Mexico, where recent studies indicate that maize represents a genetically altered form of a wild grass known as teosinte (Jaenicke-Després et al. 2003). Early maize specimens have been dated to 5420 BP at Guilá Naquitz, Oaxaca (Piperno and Flannery 2001), so it took about 1,500 years before this cultigen finally reached the Borderlands.
Current research has radically changed our perceptions of the Late Archaic across the Borderlands. This includes a diversity of scientific approaches ranging from culture-historical to evolutionary theory. The chapters in this book include both regional syntheses and specific problem orientations. With agriculturalists to the west and foragers to the east, the Borderlands provide a rare laboratory in which to study the question of why people did or did not shift to an agricultural-based economy. Scholars around the world are currently grappling with this problem, and this book provides them with a variety of perspectives and a new series of databases to use in addressing this significant research issue.
Research in the Sonoran Desert and Chihuahuan Desert regions of Sonora and Chihuahua, Mexico, and southern Arizona is illustrated in the chapters by Carpenter, Sánchez, and Villalpando C.; Mabry; and Hard and Roney. Carpenter et al.'s work at the extensive multicomponent site of La Playa, Sonora, has provided new perspectives on the Late Archaic of this region. Here they have identified an archaeological site of over 12 sq km. Much of this includes evidence of the Early Agricultural period, consisting of dense artifact scatters, thousands of roasting pits (some containing maize), possible agricultural features and canals, and cemeteries with several hundred human burials. Large villages have also been discovered in the Tucson Basin of southern Arizona. Mabry summarizes this recent research, including the large-scale excavation of domestic structures, storage pits, roasting pits, middens, cemeteries, large communal-ceremonial structures, and irrigation canals. As he points out, new evidence indicates that maize, squash, and beans may have been used as a suite of early cultigens. Finally, Hard and Roney present their findings of the Late Archaic trinchera site of Cerro Juanaqueña, Chihuahua. This site contains approximately five hundred terraces, a hundred rock rings, and midden deposits; however, like La Playa, the site includes only limited evidence for domestic structures. Nonetheless, the surface of the site is littered with chipped stone items and heavily worn basin metates. Not only is domesticated maize present, but domesticated amaranth has also been identified at the site. More importantly, Cerro Juanaqueña is not an isolated occurrence but one of several Late Archaic trinchera hilltop sites situated along the valley of the Río Casas Grandes. This new evidence changes our view of the people of the Late Archaic, from simple desert foragers to early farming communities.
Such large-scale systematic excavations are lacking from southern New Mexico and the Big Bend region of Texas. Current research indicates that although maize was present in southern New Mexico by ca. 3000 BP (Upham et al. 1987), the shift to an economy dependent on agriculture probably did not occur until quite late: that is, ca. AD 1200 during the El Paso phase (Hard et al. 1996). Doleman (this volume) explores the question of why a dependence on maize agriculture appears to have occurred much earlier in the Sonoran Desert of southern Arizona than in the Chihuahuan Desert of southern New Mexico. As he points out, the Sonoran Desert region provides a variety of resources within a more limited area, including broad perennial stream valleys conducive to floodplain agriculture. This environment would have reduced or eliminated seasonal resource scheduling conflicts experienced in other regions of the Southwest (also see Dering, this volume; Hard and Roney, this volume; Huckell 1996b; Stone and Bostwick 1995; Wills and Huckell 1994).
The eastern periphery of the Chihuahua Desert lies in the area of West Texas and northeastern Chihuahua, Mexico. Mallouf reviews the current archaeological evidence from this poorly understood region. This research has documented a marked increase in the presence of Late Archaic vs. Middle Archaic remains in the Big Bend area, something also suggested by Mabry (this volume) for southern Arizona (also see Waters 1986). This could represent the expansion of Archaic populations into the region during a period of more mesic conditions. Late Archaic campsites are distributed across a wide range of environmental settings. Mallouf (this volume) suggests that the limited evidence for maize indicates that it only represents a dietary supplement and that it was probably added to the subsistence base at the end of the Late Archaic during a period of more xeric conditions. The Trans-Pecos region therefore represents the eastern limits of maize agriculture.
Unlike West Texas, extensive excavations have been conducted in the stratified rockshelters of the Lower Pecos River region (Turpin 1991, 1995). This area of the Borderlands is situated along the transition from Chihuahuan Desert to the Tamaulipas brushland. The Late Archaic diet contained a wide variety of plant and animal resources, including lechuguilla and sotol. Dering's study (this volume) indicates that the return rates for these items are similar to those of other low-ranked species like grass seeds and roots. The resource homogeneity, the lack of broad floodplains, and a spring/fall rainfall regime may have contributed to the absence of agriculture in this region (see the discussions of Doleman as well as Hard and Roney in this volume). Bison hunting, rather than agriculture, occurred during a period of more mesic conditions ca. 2500 BP (also see Dillehay 1974; Mallouf, this volume; Turpin 1995:548).
The archaeological record of South Texas also contains no evidence for agriculture. As described by Hester (this volume), the Late Archaic archaeological record is poorly documented in the region but generally reflects short-term campsites that were commonly situated along stream channels. The people exploited a variety of plant and animal resources, including both riverine species and land snails. Nonetheless, cemeteries and middens are present and presumably reflect the repeated reuse of specific resource patches and not sedentism (e.g., Taylor and Highley 1995).
Ogilvie's chapter and my chapter represent specialized research projects along the Borderlands that involve understanding the effects of agriculture on human biology and stone tool technology, respectively. Ogilvie's human biology study of forager, early agricultural, and Pueblo groups is very informative. Specifically, her analysis of the Late Archaic Tucson Basin population indicates that the males resemble foragers but the females resemble agriculturists. This implies that important changes in sexual division of labor were beginning at this time. Ogilvie's study also has implications for my research. My preliminary results from the analysis of chipped stone items from Cerro Juanaqueña indicate a mixed core reduction/biface production assemblage similar to those of other Late Archaic habitation sites. This contrasts with an emphasis on biface production at Late Archaic campsites and core reduction at Ceramic period sites. Barbara Roth (1992, 1998) discusses similar patterns for the Tucson Basin region. The chipped vs. ground stone assemblages at Cerro Juanaqueña appear to indicate conflicting evidence for residential stability and economy at the site (also see Roney and Hard 2002). This could reflect important changes in division of labor related to an increasing dependence on maize agriculture and early village formation.
Finally, Matson and Smith provide reviews of the research presented in this volume. Matson discusses each of the chapters from a Southwestern perspective, whereas Smith offers a global perspective on the Borderlands. As he so aptly points out, many of these chapters describe the Late Archaic as representing that "middle ground" between foragers and agriculturalists, being characterized by "a rich variety of low-level food-producing societies."
In conclusion, if we are going to understand the origins of agriculture and village formation along the western Borderlands, it would help to know why the foragers in the eastern Borderlands failed to incorporate cultigens into their diet. The archaeology of South Texas indicates that cemeteries and middens can occur in a nonagricultural setting. This should be a cautionary note to those in the western Borderlands who suggest that these features are solely characteristic of sedentary agricultural communities. Nonetheless, the chapters in this volume identify several factors that could be contributing to this process, including differences in resource structure (e.g., distance between resource patches), landscape (e.g., broad alluvial valleys), rainfall regime (e.g., summer monsoons), population expansion (e.g., hunter-gatherers or farmers), and seasonality (e.g., the need for storage to solve the over-wintering problem). Finally, the diet-breadth model as derived from optimal foraging theory is discussed by Hard and Roney and by Dering. Various species can be ranked based on a cost-benefit analysis of foraging return rates (Kelly 1995:78-90; MacArthur and Pianka 1966; Stephens and Krebs 1986:17-24). Large game is generally considered to have the greatest return; succulents and grasses, the lowest. Floodplain farming is also seen as a low-investment and high-return strategy (Barlow 1997, 2002; Dering 1999; Simms 1987). It is noteworthy that during more mesic conditions along the Borderlands (ca. 2500 to 3000 BP) floodplain farming was initially used in the west, and a switch to bison hunting occurred in the east. These changing environmental conditions modified the cost-benefit relationship of these resources, thereby allowing for increasing return rates for these regionally divergent Late Archaic subsistence tactics.