Between 3500 and 500 bc, the social landscape of ancient Mesoamerica was completely transformed. At the beginning of this period, the mobile lifeways of a sparse population were oriented toward hunting and gathering. Three millennia later, protourban communities teemed with people. These essays by leading Mesoamerican archaeologists examine developments of the era as they unfolded in the Soconusco region along the Pacific coast of Mexico and Guatemala, a region that has emerged as crucial for understanding the rise of ancient civilizations in Mesoamerica. The contributors explore topics including the gendered division of labor, changes in subsistence, the character of ceremonialism, the emergence of social inequality, and large-scale patterns of population distribution and social change. Together, they demonstrate the contribution of Soconusco to cultural evolution in Mesoamerica and challenge what we thought we knew about the path toward social complexity.
Early Mesoamerican Social Transformations Archaic and Formative Lifeways in the Soconusco Region
Early Social Transformations in Soconusco: An Introduction
Richard G. Lesure
The period 3500 to 500 B.C. was one of momentous change in Mesoamerica. At the beginning of that span, the region was sparsely occupied by low-level food producers whose rhythms of existence were dominated by the concerns of hunting and gathering. By 500 B.C., it was populated with settled agriculturalists in a landscape increasingly full of people. Proto-urban communities, laid out according to spatial schemes that would continue through the Spanish Conquest two millennia later, were foci of social and political life. Public rituals included worship of deities that were to persist into the Classic and Postclassic eras. Though there was a complex history still to come, Mesoamerica as a culture area or civilization is by that point recognizable.
Major developments between 3500 and 500 B.C. include the shift from Archaic to Formative after 1900 B.C., a far-reaching social transformation that involved the appearance of permanent villages, the introduction of pottery, refocus of subsistence organization towards agriculture, and a steep rise in population.1 The transition to the Formative was followed -- still in the second millennium B.C. and thus rapidly on archaeological time scales -- by the first indicators of inequality and sociopolitical complexity, including the emergence of settlement hierarchies, monumental sculpture with themes of rulership, and massive earthen platforms or modifications to the landscape. The period also includes, from 1400 B.C., a series of stylistic horizons involving the virtually pan-Mesoamerican dissemination of strikingly recognizable visual culture, including, most famously, Olmec art.
Despite the concentration of those three phenomena in the pivotal second millennium B.C., considerations of them as a package, in a unit of time that straddles the Archaic/Formative divide, is still relatively rare. One reason is the great disparity in quantity of available evidence on either side of the point at which pottery enters the archaeological record. But relevant as well is a tendency towards fragmentation of inquiry into multiple paths with different theoretical inspiration and little in the way of intersection or cross-communication between them.
In the present volume, the period 3500-500 B.C. is taken as crucial to understandings of the genesis of Mesoamerican civilization, but the focus is narrowed from Mesoamerica as a whole to one particular region: the Soconusco, on the Pacific coast of Chiapas, Mexico and adjacent Guatemala. A volume focusing on the early archaeology of the Soconusco but with an eye on the topic of early Mesoamerican social transformations more generally is appropriate for several reasons, but most obviously because recent fieldwork makes the region the best-known case from the lowlands of Mesoamerica for understanding social change through the entire span of 3500 through 500 B.C. The emerging picture contrasts in various ways with those for the Southern and Central Highlands, and claims that early developments in the Soconusco were precocious, distinctive, or even unique have entered larger interpretive discussions in Mesoamerican archaeology. Yet much of the evidence on which those claims are based remains only summarily published.
The time is ripe for a volume of papers showcasing both the diversity of current work on the early archaeology of the Soconusco and the growing opportunities for synergy as research in multiple subregions yields a richer understanding of the whole. My goal in this introduction is to set the topics and perspectives of the papers that follow into a series of larger contexts. Since an excellent recent review of southeastern Pacific coast archaeology is available (Love 2007), I have made no effort here to be comprehensive. Instead, I review the larger importance of work on early Soconusco, introduce the region and its prehistory, and provide a brief overview of the papers that follow.
The Soconusco region is a narrow strip of the Pacific coast of Chiapas, Mexico and adjacent Guatemala (Figures 1.1 and 1.2). Sharply delimited inland by the rise of the Sierra Madre Mountains, it extends from around the modern town of Mapastepec, Chiapas, southeast to the Tilapa River in Guatemala (Lowe, Lee, and Martínez 1982:55-62). High rainfall feeds numerous rivers that cross a short coastal plain to feed freshwater marshes and brackish estuaries and lagoons. To the northwest, the Chiapas coast is drier, and indeed there is variation within the Soconusco itself. As one moves to the southeast, the coast widens and rainfall increases. In Guatemala, the southeastern boundary of the Soconusco as a geographical region is marked by an inland extension of the coastal plain, but early settlement related to that of Soconusco extended throughout Pacific-coastal Guatemala into El Salvador. One paper included here (Chapter 9) extends coverage of the volume into this related area.
The Soconusco and adjacent coastal Guatemala are rich in natural resources. Since biotic communities tend to run in narrow strips parallel to the ocean, a range of wild resources would have been readily accessible to ancient inhabitants. Indeed, the sheer richness and diversity of aquatic resources have figured prominently in claims that subsistence well into the Formative was focused more on hunting, gathering, and fishing than on agriculture.
The Larger Import of Early Soconusco
The period 3500 through 500 B.C. is of interest for study of the genesis of Mesoamerican civilization because it unites a set of deeply transformative developments -- but also because scholarly work on those developments ranges across the full theoretical diversity of contemporary anthropological archaeology. Understandings of the transition to the Formative, the emergence of sociopolitical complexity, and the pan-Mesoamerican occurrence of Olmec art increasingly drawn in evidence from the Soconusco.
Changing Subsistence from Archaic to Formative
The subsistence systems of ancient urban Mesoamerica were the products of a complex history. Of the three key crops -- maize, beans, and squash -- the staple maize was particularly deeply interwoven into the economic, social, and symbolic fabric of life (Taube 2000). There were, though, other domesticated plants, including avocado, maguey, and amaranth (McClung de Tapia and Zurita Noguera 2000). The Soconusco, in the 15th century A.D., was important as a source of cacao for the distant Aztec capital, and cacao residue has been identified on a pottery sherd of the period interest here (Powis et al. 2008). There has been on-going speculation that root crops such as manioc were important Formative-period crops in the Soconusco and on the Gulf Coast (Pool 2007:75; Clark, Pye, and Gosser 2007:28-29). Wild animals everywhere provided sources of protein, but in certain contexts domesticated animals formed an important component of the diet. Dogs were eaten at Early Formative San Lorenzo on the Gulf Coast (Wing 1981) and at Middle Formative La Blanca in the Soconusco (Wake and Harrington 2002).
The full story of agricultural origins in Mesoamerica is thus long and complicated. Still, despite some recent efforts to displace it in favor of a subsistence transformation (to maize as a staple) at 1000 B.C., the transition to the Formative stands out as a pivotal in terms of demography, subsistence, and lifeways. Debate over the causes of the transition has centered around climatic fluctuations, population growth, and the systems perspective developed by Kent Flannery. In Flannery's model for the Valley of Oaxaca, the growing productivity of maize in relation to that of wild mesquite eventually triggered removal of mesquite trees to clear the way for farming on river banks, the most desirable lands of the valley for agriculture (Flannery 1986a:26-27; 1986b:504-506). Such a shift in priorities among subsistence options could have generated a cascade of other changes, including greater sedentism and population growth.
Recent work has tended to add to rather than winnow down the list of potentially relevant variables. Now, to a set of possibilities that included climatic fluctuations, the changing productivity of maize, and localized population dynamics, we might add status competition (Blake et al. 1992a, Clark and Gosser 1995), differences between highland and lowland environments (Piperno and Pearsall 1998), population histories at vast scales (Bellwood 2005), and specific technological innovations (Neff, Pearsall et al. 2006).
More than one of these factors may be relevant for explaining the Formative transition, but it is unlikely that all are equally relevant. One way to weigh their respective importance would be to trace their articulation from Archaic to Formative in different contexts and compare the results. Such an effort is hampered by the chronic low visibility of preceramic sites in Mesoamerica and the consequent rarity of sustained work on Archaic adaptations. Path breaking efforts in the semiarid highland settings of Tehuacán (MacNeish 1964, 1981) and Oaxaca (Flannery 1986c) sparked relatively few comparable efforts in other regions. Voorhies's (1976, 2004) work over three decades therefore places Soconusco alongside Tehuacán and Oaxaca as one of a handful of Mesoamerican regions in which a social understanding of Archaic adaptations is possible. Voorhies and colleagues suggest that Archaic inhabitants of Soconusco were mobile low-level-food-producers and hunter-gatherer-fishers who visited the estuary to process marsh clams for transport to inland base camps (Kennett et al. 2006, Michaels and Voorhies 1999, Voorhies 2004). A recent addition to this work is a better understanding of how Archaic adaptations changed in the centuries leading up to the transition to the Formative (Chapter 2, this volume).
Despite the greater archaeological visibility of sites from the ceramic era, the continued evolution of subsistence systems during the Formative also needs further work. An important issue is again the degree of variability between regions, particularly during the Early Formative (the second millennium B.C.). The case of the Soconusco has figured prominently in arguments supporting a high degree of variability, with discussion centering on three topics: manioc, aquatic estuary fauna, and maize.
Lowe (1975:10-14) suggested that root crops, specifically manioc, might have been a significant food source in Early Formative Soconusco. The idea was based on a perceived lack (at Altamira, Chiapas) of manos and metates suitable for the grinding of maize and an abundance, instead, of small obsidian chips seemingly suitable for the grater boards used in the processing of manioc. Direct evidence of manioc continues to be rare or nonexistent at lowland Formative sites (Pool 2007:75), and Clark's (1981, 1994) use-wear analysis of Early Formative obsidian indicates that the chips were used for a variety of domestic tasks that did not include grating. Clark, Pye, and Gosser (2007:28-29, 31) nevertheless favor keeping open the possibility of manioc as food source in the earlier Formative.
A second suggestion -- championed particularly by Blake et al. (1992a, 1992b) as part of a larger argument on maize that I will come to a moment -- is that wild aquatic fauna of the Soconusco estuaries were an important focus of Early Formative subsistence. Neff, Pearsall, et al. (2006) pick up on that suggestion in their efforts to explain the Formative transition in coastal Guatemala, but there have been relatively few empirical contributions to this topic in the last 15 years (though see Wake 2004a, 2004b; Wake and Steadman 2009; Lesure, Wake, and Steadman 2009). Wake and I review the available data from the Mazatán zone in Chapter 4.
The third topic is the role of maize in the diets of Formative villagers during the second millennium B.C. Isotopic studies of human bone provide an important source of evidence. Certain aspects of diet can be investigated by measuring ratios of stable isotopes of carbon and nitrogen. Maize in particular has an unusually high 13C/12C ratio compared to other common plant foods in the Americas, and it is the only cultivated C4 plant. As a result, if maize was a dietary staple, it should have left an isotopic signature in human skeletons. Blake et al. (1992a) found no such signature in the Soconusco during the second millennium B.C. and suggested that agriculture was not particularly important in the Formative transition in Soconusco (see also Ambrose and Norr 1992; Blake et al. 1992b; Chisholm and Blake 2006; Chisholm, Blake, and Love 1993; Clark and Gosser 1995). Blake and his colleagues, inspired by Hayden's (1990) ideas on the role of plentiful environments in the emergence of social inequality and food production, have suggested that greater sedentism, demographic expansion, and the adoption of pottery in Soconusco was the result of fundamentally social processes, such as spiraling status competition. Maize might have been consumed as beer at feasts rather than as a dietary staple.
This work in the Soconusco helped bring diversity and variation to the foreground in studies of Formative-period subsistence. Temporal variation is one factor. If maize was originally cultivated for its stalk sugar (Smalley and Blake 2003), then it becomes important to determine the point at which the focus of selection changed from stalk to grain (Webster in press). There is some evidence to suggest that the shift to maize as a staple may have occurred around 1000 B.C. in multiple parts of Mesoamerica (Smalley and Blake 2003; Rosenswig 2006; Clark, Pye, and Gosser 2007, Webster in press). Still, regional diversity is an important topic. Was the initial-Formative subsistence system in the Soconusco unique, or did it fit into some larger pattern of variation? For instance, was it typical of early adaptations in lowlands tropical forests but different from those of the semi-arid highlands (Piperno and Pearsall 1998)? If there were such differences between highlands and lowlands, then why was the transition from Archaic to Formative essentially simultaneous in the Gulf Coast, the Valley of Oaxaca, Morelos, and the southern Basin of Mexico (Lesure 2008)? Evidence on interregional diversity should prove important for evaluating a more sophisticated generation of explanations for the Archaic-Formative transition.
A second major social transformation in the period of interest here is the emergence of sociopolitical complexity. With the initial Early Formative (1900-1400 B.C.) poorly known in many areas, the sequences of Tehuacán and Oaxaca have again proven influential as rough models of the trajectory that might have been followed in other areas. The sequences in those cases begin early, are relatively complete, and suggest a developmental sequence of gradually increasing complexity that satisfies common sense. Agricultural villages of the initial Early Formative were egalitarian, though in the Valley of Oaxaca San José Mogote was already by far the largest center (Marcus and Flannery 1996:78). In the later Early Formative, a continuum of status differences, increasing investment in public architecture, and a two-tiered settlement hierarchy suggest the emergence of chiefdoms (Marcus and Flannery 1996:96-110). The subsequent Middle Formative (1000-400 B.C.) witnessed consolidation of the hierarchical system, with population expansion, greater disparities in the distribution of high status goods, truly monumental public architecture and sculpture, and emergence of three-tiered settlement hierarchies (Marcus and Flannery 1996:111-134).
In Morelos, the development of settlement patterns around Chalcatzingo from the Amate phase suggests a basically similar trajectory of increasing complexity (Hirth 1987), but the initial Early Formative (before 1400 B.C.) remains poorly understood both there and in the southern Basin of Mexico. By the later Early Formative, there were significant concentrations of people at a scattering of regional centers -- some with platform architecture -- in both areas, including Tlatilco, Tlapacoya, Chalcatzingo, and San Pablo (Grove 2007, Niederberger 2000). The Tlatilco burials reflect village social relations of considerable complexity (García Moll el al. 1991, Joyce 1999, Tolstoy 1989). The Middle Formative is characterized by rapid population expansion, the colonization of new areas, and the replication of hierarchical political units everywhere (Grove 2007; Lesure et al. 2006; Sanders, Parsons, and Santley 1979:95-97; Serra Puche et al. 2004).
On the Southern Gulf Coast, the period 1400-1000 B.C. witnessed the spectacular fluorescence of San Lorenzo (Cyphers 1999; Cyphers and Di Castro 2009; Symonds, Cyphers, and Lunagómez 2002). The site extended across 700 ha and its sustaining area appears to have been integrated into a well-developed settlement hierarchy. Dozens of sculptures incorporate a rich and varied iconography (Coe and Diehl 1980, Cyphers 2004). Arguably, the later Early Formative sociopolitical system here was organized at a qualitatively different scale from coeval developments in the highlands (e.g., Clark 2007), particularly if these patterns coalesced earlier (towards 1400 B.C.) rather than later (towards 1000 B.C.).
Given all the above, the trajectory of social and political developments on the Gulf Coast during the initial Early Formative is of particular importance, but on that issue the data are not yet in. Coe and Diehl (1980) defined three phases prior to the San Lorenzo phase (Ojochi, Bajío, and Chicharras), but Symonds, Cyphers, and Lunagómez (2002) were only able to distinguish between "Ojochi-Bajío" and "San Lorenzo" in their regional survey. There are earthen constructions pre-dating the San Lorenzo phase (Coe and Diehl 1980:106-109; Cyphers and Di Castro 2009:23), but they are deeply buried. Hopefully, it will eventually be possible to trace the trajectory of change in sociopolitical institutions and practices on the Gulf Coast throughout the Early Formative -- we already have a glimpse of a developing ritual tradition at El Manatí (Ortiz and Rodríguez 2000) -- but we are not yet at that point.
In this context, the Soconusco Formative sequence has larger significance simply because its completeness rivals that of the Valley of Oaxaca. In particular, despite a variety of unresolved issues, it is possible to trace a sequence of changing social and political institutions over the entire Early and Middle Formative beginning from 1900 B.C. Other observations underscore the importance of the Soconusco case. First are the unexpected finds. A two-tiered hierarchy of site sizes appears surprisingly early -- in the initial Early Formative in the Mazatán area -- and there is supporting evidence for the sort of hierarchy of functions that would indicate a network of chiefdoms by 1650 B.C. (Clark 1991, 1994, 2004). The site of Paso de la Amada has yielded two dramatic discoveries, both also dating to the initial Early Formative. There is an earthen ballcourt of an impressive size by later Mesoamerican standards (Hill 1999) and, in Mound 6, the 22 m long apsidal Structure 4, with its inset porches and low clay benches or walls (Chapter 5; also Blake 1991; Blake et al. 2006). Such finds support claims of rapid -- even "precocious" -- emergence of small-scale hierarchical societies in the Soconusco (Blake and Clark 1999, Hill and Clark 2001, Clark 2004), but they raise other questions as well. Could the large buildings indicate residential patterns and social organization fundamentally different from the nuclear-family focus suggested by the residential architecture of Early and Middle Formative sites in the highlands? This new, relatively complete Early-Middle Formative sequence therefore again raises the issue of variation across Mesoamerica, here in trajectories of emerging sociopolitical complexity. Were patterns of emerging complexity in the Soconusco unique -- or perhaps characteristic of lowland or coastal areas?
The larger import of the archaeology of early complexity in the Soconusco actually extends beyond the evidence itself. Theoretical understandings of the emergence of social inequality and political complexity in Mesoamerica have been directly influenced by the Soconusco case over the last 20 years, particularly through the work of John Clark and Michael Blake. In a seminal 1994 paper on the "power of prestige," they proposed a general model for the emergence of hereditary social inequality as the result of competition for prestige among political actors ("aggrandizers") in zones of resource plenty. They illustrated the model with data from their work at Paso de la Amada and other sites in the Mazatán area of the Soconusco. The impact of that model is evident in recent synthetic treatments of emerging complexity in Mesoamerica (e.g., Joyce 2000: Chapter 2) and by inclusion of the paper in two collections intended for classroom use (Preucel and Hodder 1996, Smith and Masson 2000).
The empirical status of the aggrandizer model in the Soconusco itself is thus of wider interest. Most of us working on Formative topics in the area have shared, absorbed, or responded to Clark and Blake's emphasis on the unintended consequences of internal political competition. Clark and several colleagues have elaborated and refined the model, adding, among other things, consideration of collectivities to an original version which overemphasized the individual (Blake and Clark 1999, Clark and Gosser 1995, Clark and Pye 2000, Hill and Clark 2001, Clark 2004). Still, puzzles remain. At Paso de la Amada itself, construction of platform-top residences has a complicated history (Chapter 6; also Lesure 1997). Further, despite dramatic differences in residential architecture, there is no convincing artifactual evidence for economic differences between households in such categories as decoration or size of serving vessels, access to imported obsidian and exotics, and the manufacture of greenstone ornaments (Lesure and Blake 2002). There are hints of differential mortuary treatment, but the assemblage of excavated burials is quite limited (Clark 1991, Love 2007:285). While acknowledging the settlement hierarchy and earthen constructions that would have required coordination of labor, Love (2007:284-286) emphasizes the need for debate concerning how well a model of competitive aggrandizers actually characterizes Soconusco data.
The Historical Specificity of Olmec
A third major phenomenon of the period of interest -- wide dissemination of stylistic horizons 1400 B.C. -- invites inquiry into the specific historicity of early Mesoamerican civilization. Perhaps it is not so surprising that, despite recurring theoretical tensions between (for instance) agency and evolutionary perspectives, the deepest passions generated by investigations of the period 3500-500 B.C. surround culture-historical issues raised by the pan-Mesoamerican sharing of Olmec art (e.g. Clark 2007, Flannery and Marcus 2000; Neff, Blomster, et al. 2006a, 2006b; Sharer et al. 2006). The term "Olmec" can by itself trigger intense reactions. Grove (1989) noted problems in the common dual usage of this term as the name for an art style with characteristic subject matter and for the culture of the Early- and Middle-Formative inhabitants of the Southern Gulf Coast (or the inhabitants themselves -- "the Olmecs"). Ongoing definitional discussion has yielded a variety of more or less nuanced positions, with no version fully laying to rest the tenacious problems identified by Grove (see Clark and Pye 2000; de la Fuente 2008; Flannery and Marcus 1994, 2000; Love and Guernsey 2008; Pohorilenko 2008; Stark 2007). In this volume, Gutiérrez (Chapter 7) uses the term "Olmec" sparingly to refer to Early Formative people of the Gulf Coast, while Morgan (Chapter 9) uses it equally sparingly to refer to a style of material expressions. Rosenswig (Chapter 11), like Morgan, uses it to refer to the style, while Love and Guernsey (Chapter 8), following Clark and Pye (2000:217-218), emphasize instead the network of cultural interaction that they envision having generated the stylistic expressions. Indeed, they would identify multiple "Olmec" styles (Love and Guernsey 2008). Pye, Hodgson, and Clark (Chapter 10) seem to again blur style, culture, and people, but they embrace the notion of multiplicity, so that it becomes possible to identify distinctive "Gulf Olmec" and "Highland Olmec" expressions.
A larger theoretical issue here is how far an explanation of the rise of Mesoamerican civilization (through 500 B.C.) should venture from generalizing analysis to historical particularism. Should we attribute key innovations to the peoples of specific regions or even particular sites? What degree of impact should we accord to specific events or actions -- and at what spatiotemporal scales? Recent work in the Soconusco again makes that area an important point of reference in larger discussions. A major issue is the nature of interrelations between the Soconusco and the Southern Gulf Coast during the later Early Formative.
Clark and Blake (1989) envisioned two stages of interaction, beginning with local emulation of Gulf Coast practices, in the Cherla phase (1400-1300 B.C.), and culminating in political domination in the subsequent Cuadros phase (1300-1200 B.C.). Clark has since elaborated on that suggestion and worked to expand the evidential base from the crucial Soconusco sites of Cantón Corralito and Ojo de Agua. He argues for early (c. 1400 B.C.) state or even empire status for the site of San Lorenzo and profound "Olmec" (Gulf Coast) influence throughout Mesoamerica (Clark 1997, 2007; Clark and Pye 2000). Cheetham's (2009, 2010) careful comparisons of material culture between Cantón Corralito and San Lorenzo help to make what originally seemed like inflated claims from Clark -- a Gulf Coast colony at the Soconusco site -- increasingly convincing. There follow a host of fascinating questions concerning the nature of the immigrant group, its relations with local populations in the Soconusco, and the degree of continuing contact with its homeland. Any such discussion would revolve around particular events and actions with impact at large spatial scales.
In sum, major topics in the archaeology of the period 3500 through 500 B.C. include adaptive transformations, the rise of social inequality and political complexity, and the widespread sharing of artistic expressions from 1400 B.C. All three topics are important for any larger investigation of the genesis of Mesoamerican civilization, and in each case evidence from the Soconusco is part of larger debates. In each topic, as well, larger synthetic efforts point towards the importance of variation between different parts of Mesoamerica. The completeness of the Soconusco sequence is, by itself, enough to establish the importance of this area for larger efforts to characterize variation and assess its implications.
Yet those points do not touch on interrelations between the three topics -- or, rather, the chronic problem of the lack of such interrelations. Indeed, research in each tends to proceed largely independently of research in the others. Studies of adaptive change ignore agency and politics. Research on sociopolitical complexity becomes snobbish about politics and ignores subsistence, with its wiff of vulgar materialism. Finally, generalizing work on emergent complexity is impatient with or even embarrassed by the complications of historical specificity, while the latter becomes so self-absorbed that it ignores larger anthropological relevance or so devoted to squabbles over an Olmec "mother culture" that interpretation chases itself around in circles.
Papers in this volume chip away at those divides: Voorhies and Kennett (Chapter 2) bring gender into the study of Archaic adaptations; Lesure and Wake (Chapter 4) review the first systematic comparison of Formative Soconusco faunal assemblages; Gutiérrez (Chapter 7) explores a new dimension of historicity in the disastrous tropical storms to which the region is subject; and several papers grapple with shared style horizons at multiple scales (Chapters 10 and 11). Still, a more extended consideration of the relations between different theoretical perspectives on early Mesoamerica will help to further situate this volume with respect to a larger, integrated research effort.
Scale, Rhetoric, and the Analysis of Macro-Regions
In the eighties and nineties, anthropological archaeologists tended to treat the relations between diverse theoretical perspectives as at best competitive and at worst antagonistic. These days there is instead considerable enthusiasm for the idea that multiple perspectives might yield insight into topics as diffuse as social transformations in early Mesoamerica or the genesis of Mesoamerican civilization. The idea would not be to banish theoretical competition completely but to shift it to another level. In its place, front and center, we would place an effort to elicit synergistic contributions from seemingly disparate theoretical programs. This is not the place for an extended justification of that idea, which admittedly involves numerous challenges. I confine my attention to two important arenas for integrative work, rhetoric and scale. In each case, it becomes clear that theoretical integration will require a suite of complementary research endeavors -- endeavors with respect to which the present volume may be situated as an important category of empirical work.
Different perspectives on early Mesoamerica may be theoretically and philosophically antagonistic, but one area of common ground is the shared goal of tracing and explaining ancient social change. Rhetoric -- in this case, the shared prospect of narrating and explaining -- therefore emerges as an arena in which interconnections between theories may be explored. To narrate the rise of Mesoamerican civilization over the long term, archaeologists draw on a variety of rhetorical forms. The developmentalist narratives of processual archaeology, in which the story told is one of ever-increasing complexity from the origins of agriculture to the rise of the state, are readily at hand. Given three decades of critiques of developmentalism from positions ranging from postprocessualist to Darwinian, it is noteworthy how often we still end up telling such stories. Developmentalism proves particularly attractive (seductive?) in Mesoamerica, where the transition from earliest village to urban state unfolded over just 1500 or 2000 years (depending on how one defines terms) and thus rapidly compared to other world areas. Alternatives to developmentalism include the historicist forms long applied to study of early Mesoamerica: long-term continuity, divergence from initial homogeneity, and convergence from initial heterogeneity. These last are hardly mutually exclusive, and they may be built into very different narratives. Divergence from an initial state of homogeneity is posited in mother-culture arguments that envision an "Olmec" heartland on the Gulf Coast as a crucible of Mesoamerican civilization. Flannery and Marcus, to whom such suggestions are anathema, nevertheless chose the same rhetorical form -- manifested in the divergence of Otomanguean languages -- as a central motif for The Cloud People (1983). The third form, in which heterogeneity is early and interaction leads to convergence, is of obvious interest given the linguistic diversity that characterized Mesoamerica at the time of the Conquest. In The Cloud People, application of the model of early homogeneity specifically to Otomanguean actually implies initial diversity at a larger scale, between the major language families. Accounts positing early heterogeneity in the period 3500-500 B.C. include Grove's (1974, 2007) proposals concerning a Tlatilco culture, Clark and Blake's (1994) Mokaya for the Soconusco, and the synthetic notion of "sister cultures" as an alternative to an Olmec "mother culture" (Hammond 1988, Grove 1989).
A holistic synthesis of social transformations in early Mesoamerica will probably require a variety of rhetorical forms, but the forms prompt different sorts of research endeavor since they face distinct challenges for empirical justification. In particular, convergence from initial heterogeneity -- if it is to be placed on an equal footing with the others -- requires that we resist any urge to assume patterns in one area based on knowledge of another or received wisdom concerning the inevitability of complexity. Sustained empirical scrutiny of individual regions -- with particular attention to the earlier epochs, where evidence tends to be scarce -- is essential for the successful identification of early heterogeneity. The Soconusco, with its robust early record that diverges in many ways from well-known highland patterns, is a prime candidate for such work. The empirical requirements that would allow scholars to weigh the importance of convergence from initial heterogeneity to the genesis of Mesoamerican civilization thus provide one of the larger endeavors to which the present volume contributes.
As we consider the relevance of multiple rhetorical forms, a key issue will be scale. A satisfying narrative of early Mesoamerican civilization will probably involve a complex interplay of the forms at different scales. For instance, we might trace divergence from initial homogeneity within macro-regions -- the Central Highlands, the Southern Highlands, the Southeastern Pacific Coast -- but if those units differ considerably early on, a larger-scale comparison of those as units might reveal instances of convergence from initial heterogeneity. Given the pace of societal change between village and state in Mesoamerica, any effort to banish developmentalism completely -- attractive as the prospect might be in abstract -- seems doomed to failure. A more promising approach would be to test the power of developmentalism with shifts of focus between scales. For example, there is an unfortunate tendency, in Formative-era regional studies across Mesoamerica, to envision chiefdoms arising, de novo, in any and every area studied. Despite the fact that these occurrences are strung out over 1500 years, emphasis is on local causation rather than historical accretions or the snowballing effects of inter-linkages between regions. Historical analysis at large scales is thus one antidote to facile developmentalism. Another strategy would be to scrutinize trajectories that appear developmental at smaller scales of time. For example, change in sociopolitical organization in the Soconusco looks developmental when it is resolved in increments of 300-500 years -- though no single site was dominant for long, complexity increased steadily from initial Early Formative through Late Formative. Resolved at a scale of 100-200 years, however, the narrative becomes more complicated and challenging (Chapters 5, 6, 7).
The upshot of all this is that we approach social transformations in early Mesoamerica with uncertainty about relevant scales of space and time. Choice of narrative scale, in other words, requires empirical input. An important area for research is thus exploratory work on the relations between causal processes and the multiple scales on which they might conceivably have played out. "Macro-regions" like the Southeastern Pacific Coast need to be examined for internal structure. Was the transition to the Formative or the emergence of hierarchical societies a widespread shift across this area or were there centers and peripheries? Work in distinct study areas along the Chiapas and Guatemala coast has begun to reach sufficient density that such questions can be asked, and several of the papers in this volume make progress in that area (see Chapters 3, 8, 9, 10, 11, and 12).
In sum, a volume of collected papers on the social transformations in the Soconusco between 3500 and 500 B.C. has relevance well beyond the narrow culture-historical concerns of archaeological work in that region. The period in question encompasses transformative events of major importance in the rise of Mesoamerican civilization -- adaptive transformations that culminated in sedentary, agricultural societies; the rise of sociopolitical complexity; and the pan-Mesoamerican dissemination of style horizons from 1400 B.C. Soconusco evidence has proven relevant to larger efforts in each of these cases, and patterns in the Soconusco in each case seem surprising or distinctive -- points of great interest to theoretical work increasingly inclined to emphasize the importance of diversity across Mesoamerica. That larger interest in diversity is not idle enthusiasm. Identification of variation -- or the lack thereof -- is important in the selection of rhetorical form and relevant scale(s) for the framing our narratives of early Mesoamerica.1Dates throughout the volume are in calibrated radiocarbon years unless otherwise noted.