THE EPICLASSIC PERIOD (a.d. 700–1000)

Prior to the Epiclassic, during the Early and Middle Classic (a.d. 200–600), the study area appears to have been nearly abandoned, a process that actually began in the Middle Formative period. Although the reasons for abandonment are poorly understood, the study area was part of a larger regional phenomenon that extended east to the lower Coatzacoalcos and Tonala drainages and Grijalva Delta (Rust and Leyden 1994; Rust and Sharer 1988; Sisson 1976; Symonds et al. 2002; von Nagy 1997). The pattern to the north, however, was quite the opposite. In the Tuxtla Mountains near Tres Zapotes, the population appears to have grown during the Early Classic (Pool 2000). East of Tres Zapotes near Lake Catemaco, the population decreased during the early Early Classic, but by the Middle Classic, with the establishment of Matacapan, the population exceeded that of the Formative period (Santley and Arnold 1996:233–234; Santley et al. 1997:186). A similar pattern followed in parts of the middle San Juan drainage immediately to the north of the study area, known as the Hueyapan region. There, populations may have decreased during the Early Classic (a.d. 200–400), but by the Middle Classic (a.d. 400–700) they had increased significantly (Killion and Urcid 2001). Still farther to the northwest, the growth of Early and Middle Classic populations has been recorded in the Mixtequilla (Stark and Curet 1994) and Cotaxtla drainage (Daneels 1997).

Thus, the middle San Juan drainage in general, and the study area in particular, appears to have represented a boundary between two different regional patterns of population development: growth to the north and northwest and abandonment to the east. It should be noted that the lack of Early and Middle Classic ceramic diagnostics in the study area may affect the recognition of sites. The use of comparative diagnostics from the Tuxtlas, Chiapas, central and northern Veracruz, central Mexico, Oaxaca, and the Maya region, however, did not reveal any known Early or Middle Classic material in the survey collections. Not even Thin Orange ware, the widely distributed Middle Classic marker from the Basin of Mexico, was recovered. Other Teotihuacan diagnostics successfully used in the adjacent Tuxtlas such as floreros, tripod vessels, candeleros, incense burners, and braseros were similarly absent. Likewise, excavations at eight different sites in the adjacent lower Coatzacoalcos drainage failed to identify any Early or Middle Classic occupations, even though each contained Formative and Epiclassic occupations (Symonds et al. 2002:96). While future excavations may alter the present suppositions, it must be tentatively concluded that the SLTLC area was abandoned for most of the Classic period.

This picture stands in stark contrast to developments that took place during the Epiclassic period, when site number and population exceeded the Olmec period. Settlement patterns for the Epiclassic are illustrated in Figure 2. In the 320 km² surveyed, 319 sites were recorded, with an estimated population between 6,500 and 14,200 persons (mean = 10,350 persons). The average mean density was approximately 32 persons/km². The dramatic rise in population suggests that immigration, as well as natural population growth, contributed to population levels in the region. A distinctive architectural style referred to as the “quadripartite complex” accompanied the influx and growth of population. This configuration consisted of two parallel elongated earthen mounds forming a rectangular plaza, with two earthen mounds enclosing each end, one larger than the other.

Epiclassic settlement patterns.

Figure 2 illustrates a number of Secondary and Primary Centers located within the study area. Each is surrounded by lesser administrative sites and rural homesteads. If circles of equal diameter are placed around each, as in Figure 3, it is evident that all are spaced at regular intervals of approximately 13.5 km, forming what appear to be distinct administrative districts. The regular spacing between capitals may relate to the minimization of travel and transportation costs as well as competition between centers (i.e., Inomata and Aoyama 1996; Renfrew 1975). District capitals include Laguna de los Cerros, Cuatotolapan (SLTLC-69), SLTLC-222, SLTLC-251, SLTLC-316, and Estero Rabón (SLTLC-5), occasionally referred to as San Isidro. Often spaced at the boundaries of administrative districts were sites containing quadripartite-complex architecture, only in miniature form. These sites, SLTLC-70, 212, 312, and 343 (Figure 3), also show equidistant spacing. Given the economic importance of religious and secular institutions as reflected by public buildings (i.e., Blanton et al. 1982:46), the presence of miniature replicated ceremonial complexes may represent the administrative and economic links between district capitals and/or capitals and their hinterlands (cf. Johnson 1975:299).

District capitals and other sites with quadripartite complex architecture.

Each district capital possessed a similar architectural grammar—that is, massive quadripartite-complex architecture, a probable ballcourt, and, except for Cuatotolapan, a possible palace structure. Only a few complexes, however, have similar orientations. Cuatotolapan, SLTLC-222, and Estero Rabón, for example, are all oriented east–west, though the principle pyramid at Cuatotolapan sits on the east end and the one at Estero Rabón sits at the west end, and SLTLC-222 contains a double quadripartite complex, one possessing a principal mound on the west and the other on the east. Of the remaining district capitals, Laguna de los Cerros and RLSTLC-251 share an orientation of 10° west of north, and RLSTLC-316 possesses a north–south orientation. All district capitals except Laguna de los Cerros and, to a lesser extent, Estero Rabón appear equal in size. As discussed in greater detail later, Laguna de los Cerros was both qualitatively and quantitatively different.

In addition to district capitals, most secondary and tertiary administrative sites possessed quadripartite-complex architecture (Figure 3). The orientations of lesser administrative sites do not appear to correlate with either political allegiance or noticeable features of the landscape. In this respect, the Cuatotolapan administrative district is illustrative. Within the area surveyed, eight sites were recorded with quadripartite-complex architecture, only two of which have the same orientation. None, moreover, possesses the same orientation as the district capital, Cuatotolapan. In short, the sharing of architectural canons across administrative districts seems to indicate regional coherency, although the variation in quadripartite complex orientations, both across and within administrative districts, suggests a certain degree of political autonomy.

QUADRIPARTITE COMPLEXES

Unlike the Olmec, who left behind little evidence of surface architecture (Cyphers 1997), the inhabitants of southern Veracruz during the Epiclassic period were prodigious builders; their structures ranged from small single mounds to massive quadripartite complexes (Figure 4). A less common variant consisted of a single elongated mound. Twenty-nine sites with quadripartite-complex architecture were recorded in the study area, covering different environmental zones and including all district capitals and other high-ranking sites.

Quadripartite complexes from three sites in the middle San Juan Drainage, Veracruz, Mexico.

Although the form is repeated from site to site, quadripartite complexes vary in size, orientation, complexity, and number of additional structures. With respect to size, sites with quadripartite-complex architecture in the study area range from 1,223 cm³ to larger than 300,000 cm³. District capitals represent the largest sites. Some of the smallest, already mentioned, are those located at the boundaries of administrative districts. Orientations also vary. Although a north–south, followed by an east–west, orientation is most common, five other orientations have been recorded. This does not take into account the end on which the principal mound is located. That is, two sites with a similar orientation might have their principal mounds located at opposite ends. Finally, variation exists in the complexity of quadripartite complexes. Some sites include double and triple quadripartite complexes, either built side by side or linearly arranged, as in Figure 5. It is currently difficult to account for such variations, although some may be the result of different patterns of growth. While some centers may have grown by increasing the size of extant structures, others may have expanded horizontally through the construction of entirely new buildings.

Three linearly arranged quadripartite complexes. Two share an end mound.

Sites with quadripartite-complex architecture also vary in the number of additional structures. Many of the largest sites possess an elongated parallel mound on the exterior of each of the two principal elongated mounds. Together they may represent a ballcourt (Figure 5). Less common are probable ballcourts independent of the two principal elongated mounds. Large quadrilateral platforms, sometimes supporting low mounds in an “L” or “U” configuration, and a central sunken patio are also common at the largest sites (sites SLTLC-284 and 212 in Figures 4 and 5). None of these platforms has been excavated, and their function remains unclear. Based on analogies with the Maya Lowlands and other regions, such platforms may represent the substructures for elite residences or palaces (Andrews 1975:43; Stark 1999:209).

Quadripartite-complex architecture, as described here, was not limited to the SLTLC area and has been noted by scholars in nearby regions. On both sides of the study area—to the north in the Hueyapan area (Killion and Urcid 2001; see also Ceja Tenorio 1997) and east in the lower Coatzacoalcos drainage near San Lorenzo (Coe and Diehl 1980; Symonds et al. 2002:97–117)—the number of sites with this architectural form appears as numerous as in the SLTLC area. Farther north in the Tuxtlas, the form is noticeably absent, save for a few sites (Santley 1994:Figure 9). West of the Tuxtlas, sites with similar plans are present in the Mixtequilla area near Cerro de las Mesas at the Chivo, Ojochal, and Tio Primo complexes (Stark 1999) and into the Jamapa-Cotaxtla drainage in central Veracruz (Daneels 1997). To the south, the form is present at Medias Aguas (Gómez Rueda 1996:103), Las Limas (Gómez Rueda 1996), and Jonotal (Thomas Killion, personal communication 2002). Farther east, La Venta (Drucker et al. 1959) and Encrucijada 4^a (Gómez Rueda 1996:110) contain similar architecture.

Traditionally, sites with quadripartite-complex architecture were either assigned to the Olmec period or were believed to be Epiclassic in date but derived from an original Olmec pattern. It is likely that confusion has arisen because Epiclassic people consistently occupied areas of earlier Olmec sites, thus obscuring the ascription of structures to either period. Also, few quadripartite complexes have actually been excavated; consequently, their chronological placement in the Olmec period is often assumed rather than proved. Such an assumption may stem from the belief that the Olmec were the progenitors of mound and plaza architecture in Mesoamerica. Limited excavations at Tenochtitlán (Coe and Diehl 1980:38–39), San Lorenzo (Coe and Diehl 1980:50–55), Laguna de los Cerros (Cyphers 2001), and Las Limas (Gómez Rueda 1996:64–65), however, suggest that the quadripartite complex—at least, at those sites—is Epiclassic in date.³

Taking a larger regional perspective, regions to the north and northwest, though more peripheral to the Olmec florescence, show greater cultural continuity back to the Formative period. Such is the case for the Cotaxtla drainage, the Mixtequilla, and the Hueyapan area. Many scholars have also noted architectural similarities in the formal layout of La Venta, which is most notable for its Olmec occupation. It should be noted, however, that while the principal conical mound at the site, C-1, is believed to date to the Olmec period, it has not been excavated save for a small trench in the apron upon which the pyramid rests (Drucker et al. 1959:119). The carbon sample taken was from fill and does not necessarily date to the period that the mound was constructed (Coe and Stuckenrath 1964:16). Likewise, as noted by Richard Diehl (1981:77–78), it is difficult to assign a date to one of the excavated elongated parallel mounds, A-5, and the other elongated parallel mound, A-4, has not been excavated. In short, the inception and origin of this architectural pattern await further excavations.

POLITICAL ORGANIZATION DURING THE EPICLASSIC PERIOD

As mentioned, the sharing of architectural canons across administrative districts and the variation in the orientations of quadripartite complexes indicate regional coherence but also a degree of political autonomy. At present, it is not possible to account for the type of political organization that existed in the region during the Epiclassic. A refinement of the ceramic chronology and excavations of quadripartite complexes are necessary to assess issues of function and temporality. Nevertheless, as a first attempt, it is instructive to examine two models of political organization located at the polar ends of a continuum. A segmental model of political organization is based on a decentralized, replicated, and unintegrated structure. In contrast, the unitary model is based on a centralized, differentiated, and integrated political structure. Segmental versus unitary political systems are commonly debated in the Maya area (Fox et al. 1996); both models appear germane to southern Veracruz, as well.

A. W. Southall (1956) described segmentary polities as semiautonomous districts that represent political duplications of one another. As mentioned, they have low degrees of centralization, differentiation, and integration characterized by both political affinity and autonomy. Within the administrative hierarchy, similar political functions are repeated at each level, only on a decreasing scale. That is, provincial centers mirror the capital but in reduced form. The recognizable powers of each subcenter make heterarchical relationships important. Economically, the basic organizational unit, the segmentary lineage, possessed its own property composed of elite estates and corporate holdings that insured economic independence from other lineages. Each, moreover, maintained its own positions of leadership and self-rule. Lineages were ranked by the degree to which they were descended from a common ancestor. Thus, the overarching cohesiveness of many segmental polities was through ritual and kinship (Fox et al. 1996:795).

A segmentary model seems at least partially supported in the study area. Most obvious is that the study area is composed of administrative districts of equal size, each with a similar settlement system made up of a capital with quadripartite architecture, lesser administrative sites, and a rural hinterland. The sharing of architectural canons across administrative districts indicates regional coherence, but the variation in quadripartite-complex orientations suggests overall political autonomy. Within administrative districts, the replication of quadripartite complexes at differing hierarchical scales may represent the scaling down of identical political functions. Quadripartite complexes may have been the markers for elite estates bound together through kinship, ideology, political, and, possibly, economic functions.

Data also partially support a unitary model, however. Recall that the capital of Laguna de los Cerros was both qualitatively and quantitatively larger than other district capitals. Two streams bound this site on the west and east, the Zanja Grande and Zanja Prieta. Based on excavations and surface reconnaissance, Laguna de los Cerros has tentatively been divided into a main core containing the largest civic-ceremonial architecture and two subgroups to the north and south, each with their own internal organization. Site SLTLC-133, for example, centrally dominated Subgroup 2 and contained quadripartite-complex architecture (Figure 6). The subgroups are connected to the main core by an intervening area of light but continuous occupation. On the sides opposite the two streams appear separate architectural complexes, the Grande complex and Prieta complex, each composed of a cluster of sites. Each has its own internal organization and is dominated by a centrally positioned large site (Figure 6). The estimated population for Laguna de los Cerros and its two complexes is between 3,465 and 7,515 persons (mean = 5,490 persons). That is, within the entire survey area, more than half the population resided around Laguna de los Cerros.

Settlement pattern of Laguna de los Cerros, including Grande Complex and Prieta Complex.

In many ways, the internal hierarchy that characterizes Laguna de los Cerros approximates what Olivier de Montmollin (1989) has described for the capital site of Tenam Rosario in Chiapas. There, de Montmollin (1989:153) depicts the capital as a political “microcosm” of its hinterland, with an internal hierarchy of sites and complexes representing the levels of hierarchy found outside the capital. Extending this model to Laguna de los Cerros, there are two ways to relate the internal organization to the larger regional political hierarchy. One is to view all internal political levels as hierarchically above all Secondary Centers in the hinterland. The other is to view only the topmost level—that is, the main core—as hierarchically above other Secondary Centers in the hinterland. In this way, Secondary Centers within Laguna de los Cerros would be equivalent to Secondary Centers in the hinterland (de Montmollin 1989:165, 227).

Related to the idea of a microcosm is Barbara Stark's (1999) concept of a “capital zone.” In the Mixtequilla area of south-central Veracruz, Stark reports a pattern of closely spaced architectural complexes of differing size. She posits two possible explanations with respect to political organization. One is that closely spaced architectural complexes represent secondary and tertiary centers within a regional hierarchy. The other is that distinct but closely spaced complexes correspond to a core zone that formed a superordinate administrative capital, or a capital zone. Both the microcosm and capital-zone models merit further investigation with respect to Epiclassic political organization at Laguna de los Cerros.

RESUMEN

Las investigaciones arqueológicas en el sur de Veracruz, México, se han centrado tradicionalmente en los logros impresionantes del Los Olmecas formativo temprano, una de las sociedades complejas más tempranas de Mesoamerica. Menos estudios han investigado los desarrollos culturales desde los Olmecas. Investigaciónes en partes del los drenajes San Juan y Coatzacoalcos del sur de Veracruz proporciona nuevos datos en el período de epiclassico, cuando los niveles de la población excedieron el período olmeca. Examinos los patrones de asentamiento regionales y complejos arquitectónicos para interpretar la organización política existente durante el epiclassico. Se examinan dos modelos políticos en gran parte polarizados, uno segmentado y el otro centralizado. Quizás no sorprendentemente, los datos son consistentes con aspectos de ambos modelos. Ultimamente, los datos son comparados con otras regiones cercanas para colocar el área del estudio en un contexto regional más grande. Concebido tradicionalmente como región monolítica, las tierras bajas del Golfo son caracterizadas mejor por la variación cultural, discontinuidades de la población, y la interacción intra- e inter-regional.