1996; Cyphers et al. 2013: 56–59; Goman 1992; Ortiz and Rodríguez 2000; Pope et al. 2001; Rust and Leyden 1994; VanDerwarker 2006). Given the poor preservation in most Olmec sites, this undoubtedly is a partial list.
The list of animals is even more extensive, running to over 14 kinds (genus or species) of mammals, 8 kinds of birds, 8 of reptiles, 10 of fish, and 2 of mollusks (oyster and marsh clam) (Peres et al. 2013; Pope et al. 2001; Rust and Leyden 1994; VanDerwarker 2006; Wing 1980). At San Lorenzo, dog was so prevalent that Elizabeth Wing (1980) compared its consumption to that of beef in Medieval Europe. White-tailed deer, peccaries, and a variety of rodents (possibly obtained in garden hunting) commonly appear in Olmec sites as well. However, aquatic species, including waterbirds (principally ducks), turtles, and fish, account for a majority of the protein consumed by Olmecs at lowland sites such as San Lorenzo, San Andrés, and Tres Zapotes. The annual floods would have dispersed freshwater and estuarine fish over wide areas, making them harder to catch during high water, as Cyphers and colleagues (2013) point out, but as the waters receded fish would have become concentrated in oxbow lakes and backwater swamps (Coe and Diehl 1980; Loughlin and Pool 2017).
Settlement patterns also suggest particular strategies for exploiting aquatic resources across Olman. In the lowlands, the earliest and largest sites tend to concentrate along the edges of river floodplains and on levees and islands of high ground within them (including at San Lorenzo and La Venta). To the northwest of San Lorenzo, 62 low platforms locally called islotes dot the broad alluvial plain formed by the convergence of the Ríos Tatagapa and Chiquito. All but one seem to have been initially constructed in the pre-Olmec Ojochi phase (ca. 1750–1550 BCE) (Figure 2.3). Owing to the nearly 2 meters of alluvium covering their bases, many more may have existed in Olmec times. These islotes likely served as bases of operation for catching and preserving fish, exploiting other wild resources, and practicing recessional slough agriculture as floodwaters receded (Cyphers et al. 2013). If, as Cyphers and her colleagues argue, the ownership of these platforms remained in the hands of the founding pre-Olmec lineages, they could have provided them a source of control that promoted the emergence of Olmec elites. Much later, on the levees and flood plain of the Zapotal and San Antonio Rivers to the west of the Tuxtlas, borrow pits next to small platforms would have also served to trap fish brought in by flood waters, as they do today (Loughlin and Pool 2017).
Figure 2.3 Early Formative settlement patterns in three regions of Olman.
Middle Coatzacoalcos basin and Western Tabasco Plain adapted from Symonds et al. (2002: Figure 4.6) and Rust and Sharer (1988: Figure 1), respectively.
Mineral resources were an important component of the economic landscape, and each major Olmec center was in a position to control particular resources not shared evenly in other parts of the region. Tres Zapotes and Laguna de los Cerros lay close to sources of basalt stone used for monuments as well as axes and grinding stones so crucial in agricultural clearing and food preparation (Grove 1999: 228; Jaime-Riverón 2016) (Figure 2.1, Figure 2.2c). San Lorenzo’s location gave it access to salt and hematite (red ochre pigment) associated with salt domes; sandstone, used for some monuments and as abrasives for working stone; kaolin clay for pottery and bentonite used to pave floors; and a concentration of tar seeps that yielded bitumen for waterproofing boats and ceramic vessels as well as for painted designs and medicines (Cyphers 2012: 78–87; Grove 1999: 228; Wendt 2009; Wendt and Lu 2006). La Venta benefited from its proximity to similar sources of salt and bitumen on the western Tabasco Plain and also to coastal resources and transport routes (West et al. 1969). If emerging leaders at these sites were able to control cooperative exchanges with counterparts in other environmental zones, they would have been in a position to expand their own local authority and power (Grove 1999: 227–228). However, given the marked differences in the size and prominence of contemporaneous sites at different times in their settlement histories, as well as their geographical locations, the degree of engagement in such exchange networks could have varied profoundly. For example, San Lorenzo, Laguna de los Cerros, and La Venta likely would have been more engaged with one another toward the end of the Early Formative period than before or with the then-smaller site of Tres Zapotes on the region’s western margin (Pool et al. 2010a).
The Olmecs also positioned prominent sites at key localities on intraregional transportation routes (Figure 2.3). As described by Ann Cyphers and her colleagues (e.g., Cyphers 2012; Cyphers and Hirth 2016; Symonds et al. 2002; see also Coe and Diehl 1980), the plateau occupied and expanded by San Lorenzo was located with access to a maximum set of control points in the lower Coatzacalcos river system, while secondary centers and facilities were placed at confluences, fords, and intersections of terrestrial and riverine routes. Landscape modifications on the floodplain near San Lorenzo and at the southern end of the adjacent Loma del Zapote plateau have been interpreted as facilities for canoe traffic (Cyphers and Hirth 2016; Cyphers et al. 2013: 122–123). Elsewhere, the locations of the large Early Formative center at Laguna de los Cerros and the nearby basalt workshop at Llano del Jícaro probably took advantage of their location some 5 km from the Río Negro (an old course of the Río San Juan) to export basalt and acquire other products (Figure 2.1). La Venta was situated similarly to San Lorenzo on a salt dome in the swamps amid the Tonalá River and its tributaries, with access to the Gulf of Mexico (Rust 2008) (Figure 2.3). At the opposite end of Olman, Formative settlements concentrated along the arroyos that flowed out of the Tuxtlas to the Zapotal/San Antonio rivers and thence to the Papaloapan (Figure 2.3). The largest of these early settlements, at Tres Zapotes and El Mesón, lay near the fall line between the Tuxtlas foothills and the alluvial plain, suggesting they too were positioned strategically at matrix control nodes in local transportation routes (Pool and Loughlin 2016; see also Hirth 1996).
The Olmec economic landscape articulated with other Mesoamerican landscapes through a network of pathways along which traveled several important resources. Those recovered archaeologically include obsidian from sources in central Mexico and Guatemala, greenstone that included jade and serpentine from Guatemala and serpentine from the Mexican states of Puebla, Oaxaca, and Guerrero, iron ore from Oaxaca and Chiapas, marine shells and stingray spines from the Gulf and Pacific coasts, and pottery and ceramic figurines from the Gulf Coast (see, e.g., Blomster et al. 2005; Garber et al. 1993; Hirth et al. 2013; Jaime Riverón 2003; Pires-Ferreira 1976a, 1976b; Pool et al. 2014) (Figure 2.4). The volume of exchange in different goods and particular sources of goods varied greatly across space and time in Olman, as it did throughout Mesoamerica. For example, the acquisition of greenstone had a long history in Olman, as evidenced in Initial Formative offerings at El Manatí (see discussion that follows and Ortiz Ceballos and Rodríguez 2000), but it reached extraordinary heights in the Middle Formative at La Venta (Drucker et al. 1959) and declined in the Late Formative period at Tres Zapotes (Pool and Loughlin 2016). Furthermore, interregional exchange was not confined to prestige goods like greenstone. Flakes of obsidian, which does not occur naturally in Olman, constituted a common item in household toolkits there and across much of Mesoamerica. Obsidian blades appear to have been a more specialized product, but the degree to which their production was controlled by elites is uncertain and likely varied from region to region (Cyphers and Hirth 2016; cf. Clark 1987). Recent studies of obsidian assemblages at San Lorenzo (Cobean et al. 1991; Hirth et al. 2013), La Venta and surrounding sites (Doering 2002; González personal communication 2005; Stokes 1999), Tres Zapotes (Pool et al. 2014), and sites in the Tuxtla mountains (Santley et al. 2001) indicate that Olmec obsidian assemblages varied greatly in the variety and relative frequency of different sources across space and time, even at contemporaneous sites within a few kilometers of one another, such as La Venta, San Andrés, and Isla Alor. Consequently, it is unlikely that any single site controlled interregional exchange for Olman at any time in its history.
Figure 2.4 Map of Mesoamerica, showing exchange of products to and from Gulf Olmec sites. Heavy line indicates area of pedestrian
