Archaeological attention was drawn early on to the Alto Magdalena (Figure 21.1). The spectacular tombs and monumental sculpture of San Agustln were described in print 250 years ago (Santa Gertrudis 1970), and archaeologists were drawn to them repeatedly through the twentieth century (Preuss 1931; Perez de Barradas 1943; Duque 1964; Duque and Cubillos 1979, 1983, 1988, 1993; Llanos 1995b; and others). Consequently, the Alto Magdalena is perhaps southwestern Colombia’s most studied archaeological region (Dren-nan 2000). The famous tombs pertain mostly to the Regional Classic Period between about AD 1 and 900, and consist of earthen mounds up to some 4 m high and 40 m across over a principal burial chamber of large stone slabs, sometimes containing a wooden or stone sarcophagus. Although extremely acid soils make the recovery of skeletal material rare, it appears that a single individual was typically buried in this principal chamber. Smaller slab tombs and burial pits of various shapes and sizes might be included within the mound as well. Monumental sculpture was sometimes buried in the larger mounds and adorned the level plaza areas around the mounds, which seem suitable for ritual assembly of moderate numbers of spectators and/or participants (Figure 21.2). While a number of particularly large tombs and elaborate statues are found at various locations not far from the modern town of San Agustln, dozens of separate archaeological sites are known to contain such features (Sotomayor and Uribe 1987; Drennan 2000, 2005).
The iconography and symbolic significance of the Alto Magdalena statuary have been the subject of numerous analyses (Preuss 1931; Perez de Barradas 1943; Duque 1964; Reichel-Dolmatoff 1972; Hernandez de Alba 1979; Gamboa Hinestrosa 1982; Bruhns 1982; Velandia 1994; Llanos 1995a). While these analyses disagree on many points, the combinations of human and animal figures and characteristics (Figure 21.3) leave little doubt that the
Figure 21.2. Burial mounds and adjacent plaza with statues in the Alto Magdalena. (Robert Drennan)
Figure 21.3. Alto Magdalena sculpture showing supernatural combinations of human and animal figures and characteristics. (Robert Drennan)
Subject matter is supernatural. Some have argued, more specifically, that these are representations of the shamanic human-animal transformations known ethnographically for a number of indigenous belief systems in northern South America. The principal individuals, then, whose burials were accompanied by such statues, might have been shamans, or priests, or more secular leaders whose authority was backed up by supernatural power.
Although enough gold work exists to define a regional style, in the broader context of southwestern Colombia, gold in the Alto Magdalena stands out primarily for its scarcity (Duque 1982). The tombs, architecturally and sculpturally the most elaborate between the Maya area to the north and the Central Andes to the south, are very undistinguished in terms of offerings of gold, pottery, or other materials commonly associated with the burials of wealthy individuals. Most of the gold work from the Alto Magdalena was apparently recovered from tombs (Bray 2000; Duque 1964), where it does seem to have symbolized social status (Langebaek 2003; Rodriguez 2005). But gold work is a rare find in the Alto Magdalena, and actual offerings of any kind are just not generally abundant, even in
Architecturally elaborate tombs. This has been taken to indicate that leadership and social hierarchy in the Regional Classic societies of the Alto Magdalena were founded primarily on religion and ideology, as opposed to economic control or wealth accumulation (Drennan 1995b). Certainly the political economy was capable of mobilizing the resources necessary to construct monumental tombs and statues (Rodriguez 2005), but household archaeology has found only the slightest evidence of the larger or more substantial dwellings or more elaborate or costly possessions that might indicate wealthier households (Blick 1993; Jaramillo 1996; Gonzalez 1998; Drennan 2000; Quattrin 2001). And economic specialization and exchange, based either on craft production or vertical differentiation of agriculture, seem poorly developed up through the Regional Classic Period (Taft 1993; Gonzalez 1998; Quattrin 2001). Both these observations add further support to the conclusion that the social glue that held together the Regional Classic hierarchies of the Alto Magdalena derived more from the realm of beliefs than from the economy. This is consistent with the suggestion that monumental tombs and associated rituals were the principal currency of competition for succession at the death of a leader, in a society where leadership was incompletely institutionalized (Langebaek 2003).
Regional settlement study has been carried out in four separate tracts totaling 850 km2 in the Alto Magdalena. These are spread from the San Agustln zone along the upper course of the Magdalena River northeastward for some 100 km along the eastern slopes of the Central Cordillera to the Valle de la Plata (Drennan 2000, 2005). Ceramics throughout this area are easily classified into the same set of types, and excavations at several sites demonstrate that these same ceramics are found all along the adjacent floor of the Magdalena Valley (Llanos 1988, 1990, 1993, 1999; Moreno 1991, 1995; Sanchez 1991). Sites with monumental tombs and sculpture, however, are restricted to the higher flanks of the Central Cordillera between about 1,400 and 2,000 masl. It is this same zone of fertile soils, abundant rainfall, and very steep slopes (Herrera et al. 1989), where the density of Regional Classic occupation is by far the highest, reaching levels estimated between 22 and 44 persons per km2—somewhat in excess of modern population density in the same zone (Drennan 2005). Even at such high population density, substantial areas of uncleared forest, as well as under-exploited zones of very fertile soils, have been seen as evidence that Regional Classic population was still well below carrying capacity under the existing agricultural technology (Drennan and Quattrin 1995). Cold, wet conditions verging on paramo set the upper limit to substantial pre-Hispanic human occupation at about 2,600 masl. Occupation was spread in very dispersed fashion through the landscape, occurring largely in the form of individual farmsteads or small groups of houses interspersed with cultivated and uncultivated land (Figure 21.4). Excavated houses in the Alto Magdalena are round and small—less than 20 m2 until after the Regional Classic (Blick 1993; Jaramillo 1996; Drennan 2000; Quattrin 2001)—and are usually taken to be the residences of nuclear family households. There is no spatial indication that interaction between these households was structured into the kind of small local communities, or villages, often assumed to characterize simple agrarian societies (Peterson and Drennan 2005; Drennan 2005; Drennan and Peterson 2005). The lack of correspondence between the distribution of occupation or of elite centers and the distribution of prime agricultural resources has been seen as further evidence that control over basic resources was not at the heart of social hierarchy in the Regional Classic (Drennan and Quattrin 1995; Gonzalez 1998; Drennan 2005).
Although dispersed, settlement distribution was not uniform. Rather, it formed clear concentrations at a regional scale—zones where residence, while still interspersed with
Figure 21.4. The distribution of Regional Classic Period occupation in the western survey zone of the Valle de la Plata, in the Alto Magdalena. a. map of occupied areas; b. occupational density represented as a surface, showing a jumble of dispersed occupation not forming clear nucleated local communities; c. occupational density as a mathematically smoothed surface, showing higher-order communities, each with a central funerary complex (Peterson and Drennan 2005; Drennan and Peterson 2005; Drennan, ed., 2005)
Ample amounts of unoccupied land, was denser than in other zones that, thus, separated the concentrations (Figure 21.4). In one surveyed area in the Valle de la Plata, these concentrations, or higher-order communities, have been interpreted as small chiefly polities, each focused around a ritual center with monumental tombs and sculpture (Drennan and Quat-trin 1995; Peterson and Drennan 2005; Drennan 2005; Drennan and Peterson 2005). The burials of deceased leaders, and the rituals that may have been carried out for some time afterward at the permanent monuments that marked their places of interment, seem likely to have been an essential element in the centripetal forces that produced the demographic centralization evident in the settlement patterns. The settlement concentrations comprised populations of perhaps as many as 5,000 inhabitants in an area no more than 10 or 15 km across. The tomb-and-statue complexes in the western survey zone of the Valle de la Plata, shown in Figure 21.4, are not the most elaborate of the Alto Magdalena, but complete results of settlement study around the most impressive centers near San Agustln (Drennan 2000) have not yet been presented.
The Formative Period, which precedes the Regional Classic in the Alto Magdalena, occupies the first millennium BC, and has been divided into three roughly equal parts. Population levels were much lower than those of the Regional Classic, but occupation was scattered broadly throughout the region from very soon after the establishment of cultivation and sedentary living around 1000 BC. Regional-scale settlement concentrations, foreshadowing those of the Regional Classic, are detectable during the latter part of the Formative, although the tendency toward concentration is not as strong as it was later on. A few dates for monumental tombs suggest that their construction began in the last few centuries BC, but the practice became considerably more common and widespread at about AD 1, coinciding with dramatic regional population growth.
After about AD 900, in the Recent Period, large earthen burial mounds and monumental statues were no longer made. Nor are other monumental remains in evidence. This was once taken to indicate an abandonment of the zone most densely occupied during the Regional Classic, but it is now clear that regional population actually grew, and the settlement concentrations of the Regional Classic persisted and even intensified. This suggests considerable continuity of occupation, but raises the possibility of change in the basis of sociopolitical organization, which seems no less centralized than before, with integration at a similar spatial scale, but lacking the conspicuous symbolic foci provided by the tomb-and-statue complexes. The possibility has been raised that social hierarchy and political centralization were increasingly underwritten by a degree of economic control that rendered continued construction of burial monuments unnecessary (Drennan 1995b). Most houses continue to be small, but one, variously identified as a public structure or a chief’s house, reaches nearly 60 m2 (Duque and Cubillos 1981). If a residence, its larger size could indicate a household of substantially greater wealth. In the arena of craft production, there is evidence of more specialization and possibly greater control by elites (Taft 1993). Greater investment in intensifying agricultural production through drainage works is also in evidence by the Recent Period, raising yet another familiar possibility of resource mobilization by elites (Sanchez 2000, 2005). Some evidence suggests a major population decline in the zone at 1,400-2,000 masl shortly before the Spanish Conquest, while lower elevations continued to be occupied (Drennan 2005). These lower zones are much more conspicuous in sixteenth-century Spanish accounts of indigenous occupation, and yield abundant late pre-Hispanic ceramic types not commonly found higher up. Cultural relationships within the Alto Magdalena and with neighboring regions may well have been in flux in the fifteenth and sixteenth centuries.
World History
