Applicable to features both designed by natural selection and generated by cultural design forces, costly signaling theory predicts the evolution of signaling systems that reliably communicate information about some underlying quality. For instance, an individual's superior access to resources will not be obvious unless they advertise it, so advertising is necessary for the individual to gain the prestige benefits of superior access. To be believable, however, the advertisements must be impossible to fake unless the individual actually controls a disproportionate share of resources. Wasteful, resource-consuming advertisements (e.g., conspicuous consumption or conspicuous generosity) are the result. For Early Formative period Mesoamerica, this simple, though perhaps counterintuitive, formulation provides expectations about empirical patterning borne out by existing observations on the material record of cultural elaboration within the Gulf Coast Olmec heartland and by changes over time in patterns of inter-regional interaction.

Interest in the Olmec, arguably Mesoamerica's first civilization, has surged over the past decade, with two full-length books (Diehl Reference Diehl2004; Pool Reference Pool2007), a recent Ancient Mesoamerica (Spring 2010, Vol. 21) special section titled “Rethinking the Olmecs and Early Formative Mesoamerica,” a major synthetic analysis and catalogue of Olmec art in the collection at Dumbarton Oaks (Taube Reference Taube2004), and numerous journal articles. The Olmec even play a role in modern identity politics, setting claims of an African stimulus for the Olmec in contrast with enthusiasm for archaeological evidence of home-grown artistic and technological creativity.Footnote ¹

The Olmec emerged during the Early Formative period, roughly 1800–1100 b.c., a time when people throughout Mesoamerica were first congregating in permanent settlements and making pottery. Some additional innovations, such as carved monumental and portable stone sculpture, and long-distance movement of fine pottery and other goods, seem to have been introduced by people of the Gulf Olmec heartland, also called Olman (Clark Reference Clark1997; Coe Reference Coe, Sharer and Grove1989; Diehl Reference Diehl2004, Reference Diehl2005; Diehl and Coe Reference Diehl, Coe and Coe1996; Lowe Reference Lowe, Sharer and Grove1989; Pool Reference Pool2007). Because the heritage of these Olmec innovations can be traced in later Mesoamerican art and material culture, Mesoamericanists have referred since the 1940s to the Olmec as Mesoamerica's “mother culture” (Bernal Reference Bernal1969; Covarrubias Reference Covarrubias1942, Reference Covarrubias1957).

A variety of archaeological evidence favors the view that some Gulf lowland individuals or factions were better able than their contemporaries elsewhere to harness energy and resources toward non-subsistence ends. For one thing, the scale of construction activities at San Lorenzo exceeded by a substantial margin the scale of construction elsewhere (Clark Reference Clark1997; Coe and Diehl Reference Coe and Diehl1980a; Cyphers Reference Cyphers, Stark and Arnold1997; Diehl Reference Diehl2004; Symonds and Lunagómez Reference Symonds, Lunagómez, Stark and Arnold1997; Symonds et al. Reference Symonds, Cyphers and Lunagómez2002). Second, monumental stone and wood carving were well developed on the Gulf Coast long before either appeared in other regions (Diehl Reference Diehl2004; Ortíz and Rodríguez Reference Ortíz Ceballos, Rodríguez, Clark and Pye2000). Third, as documented by a recent ceramic provenance study, Gulf Coast Olmec people were exporting substantial quantities of fine white and carved-incised gray pottery at a time when other centers exported little if any pottery (Blomster et al. Reference Blomster, Neff and Glascock2005; Neff and Glascock Reference Neff and Glascock2002; Neff et al. Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Diehl, Houston, Joyce, Lipo, Stark and Winter2006a, Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Cyphers, Diehl, Houston, Joyce, Lipo and Winter2006b).

If Olmec innovations appeared concurrently across a single contiguous region, there might be little debate over the importance of the Olmec as inventors of many features of Mesoamerican civilizations. However, “Olmec” things have variable temporal and spatial distributions, appearing at various times between 1500 and 400 b.c. over a wide area of Mesoamerica, from El Salvador and Honduras (e.g., Joyce and Henderson Reference Joyce and Henderson2010) to the Basin of Mexico (e.g., Clark Reference Clark, Scarborough and Clark2007; Diehl Reference Diehl2004; Niederberger Reference Niederberger, Clark and Pye2000; Pool Reference Pool2007). The broad temporal and spatial distribution of the “Olmec phenomenon” has given rise to a competing metaphor that describes the Olmec not as Mesoamerica's “mother culture” but as one of several “sister cultures” (Hammond Reference Hammond1988) that participated in the genesis of Mesoamerican civilization. Proponents of this alternative metaphor include Kent V. Flannery (Reference Flannery and Benson1968), David Grove (Reference Grove, Sharer and Grove1989), Arthur Demarest (Reference Demarest, Sharer and Grove1989), Joyce Marcus (Reference Marcus, Sharer and Grove1989), and Christine Niederberger (Reference Niederberger, Clark and Pye2000).

Despite assertions to the contrary (e.g., Cheetham and Blomster Reference Cheetham and Blomster2010), the mother- and sister-culture metaphors still guide most research and frame most discussions of the Olmec. This is evident in the recent Ancient Mesoamerica special section, where Blomster (Reference Blomster2010) highlights the asymmetry between the Oaxacan chiefdoms and the nascent Olmec state, Reyes González and Winter (Reference Reyes González and Winter2010) describe Gulf Olmec influence on ceramics at a small site on the southern Isthmus of Tehuantepec, and Cheetham (Reference Cheetham2010a) suggests that Gulf Olmec colonized southern Chiapas based on his excavations at Canton Corralito. John Clark (Reference Clark, Scarborough and Clark2007) makes an even stronger statement of the mother-culture position, arguing that the Gulf Olmec capital of San Lorenzo became a state over 1,000 years before the runner-up, Monte Alban. On the sister side, Joyce and Henderson (Reference Joyce and Henderson2010) emphasize the local adoption and exploitation of “Olmec” symbols by Early Formative period Hondurans. Other papers in the special section highlight the variable expression of “Olmec” within Olman itself, which leads Pool et al. (Reference Pool, Ceballos, Martínez and Loughlin2010:102–103) to suggest that San Lorenzo may not have been particularly powerful even within the heartland, and Wendt (Reference Wendt2010:119) to conclude that assessments of the strength of Olmec influence outside of Olman may be flawed by incomplete understanding of patterning within household assemblages.

The mother- and sister-culture metaphors have been highly productive in the sense that they have generated hypotheses that are testable with the empirical data of archaeology (e.g., carved-incised vessels were exported from Olman but not vice-versa [Blomster et al. 2005]). Adequate existing theoretical frameworks are liberating in science, since they set scholars free to get on with the job of making observations and examining whether expectations are met. Thus, ironically, while Cheetham and Blomster (Reference Cheetham and Blomster2010:93) eschew the “polarized debate” between the two sides in their introduction to the recent special section, it is this very debate that has made it possible for the authors of the special section to “present robust data sets that permit well-grounded, testable, and diverse interpretations about the nature of the Gulf Olmecs and their impact abroad.”

Robust data sets are certainly desirable, and it may be, as Cheetham and Blomster suggest, that the best course for Olmec archaeology and for archaeology in general is to get on with the job of meticulously documenting empirical patterning in the record. But there is another possibility too. Perhaps more explicit theory can be developed that specifies mechanisms and conditions capable of generating precocious accumulation and assertion of political power on the Gulf Coast together with broad and variable distribution of “Olmec” artistic elements across Early and Middle Formative Mesoamerica. The mother- and sister-culture metaphors have served us well, but it may be time to rethink the conceptual basis of our understanding of Olmec.

While it is not the only approach to theory building in historical science, evolutionary theory has a track record far superior to any other. Evolutionary theory has generated powerful explanations of historical entities (organisms) in biology for well over a century, something that no other theoretical perspective can claim. This prior success provides one rationale for turning to evolutionary theory as a source of explanations for the historical entities of interest to us as archaeologists.

I argue in what follows that predictions derived from the evolutionary theory of costly signaling (Zahavi and Zahavi Reference Zahavi and Zahavi1997), introduced to archaeology by Neiman (Reference Neiman, Burton and Clark1997) under the name “wasteful advertising,” are consistent in surprising detail with observations on the Early Formative archaeological record. In brief, I argue that different kinds of Olmec cultural elaboration were designed as prestige signals intended for different audiences. Not only are the designs of such artifacts as colossal heads, fancy ceramic exports, and carved jades compatible with the hypothesis that they are prestige signals, but the context in which they first appear, the way they were consumed, and the temporal trajectory of their use in various locations also conform to predictions derived from costly signaling theory. Costly signaling also leads to hypotheses that can be tested in future research. One byproduct of viewing Olmec cultural elaboration from a costly signaling perspective is an understanding of how both the “mother culture” and “sister culture” metaphors contain grains of truth. Pool (Reference Pool2007:17) also concludes that proponents on both sides of the debate make valid points. Nonetheless, because ecological circumstances allowed Gulf Olmec populations to invest more heavily in prestige signaling, they contributed disproportionately to the artistic and symbolic heritage of later Mesoamerican people.

COSTLY SIGNALING

While evolutionary theory has produced some credible accounts of material patterns generated by subsistence activities (Cannon Reference Cannon2003; Winterhalder and Kennett Reference Winterhalder, Kennett, Kennett and Winterhalder2006) or stylistic variation in artifacts (Eerkens and Lipo Reference Eerkens and Lipo2005; Lipo et al. Reference Lipo, Madsen, Hunt and Dunnell1997; Neiman Reference Neiman1995), the material trappings of “complex society” (temples, palaces, rich burials, long-distance circulation of goods, etc.) have been less susceptible to evolutionary analysis. Some theoretical foundation is therefore required before proceeding to the case of Olmec cultural elaboration.Footnote ²

Elaborate structures are puzzling outcomes in both biological and cultural evolution because, in both cases, generating and/or maintaining the elaborate structures would seem to reduce fitness by reducing the resources available for subsistence and mating. One solution to this problem that is applicable in both biological and cultural evolution builds on the evolutionary theory of costly signaling (Zahavi and Zahavi Reference Zahavi and Zahavi1997). Bleige Bird and Smith (2005) have recently argued that a wide range of cultural phenomena that are seemingly intractable to evolutionary analysis, including ritual, symbolism, and prestige-related activities, may be explicable in terms of costly signaling. One may add, as I do here (and see Neiman Reference Neiman, Burton and Clark1997), that archaeological phenomena that seem puzzling from a fitness-maximization perspective, such as wasteful expenditure in stone carving and long-distance transport of bulky, fragile ceramic vessels, also conform to predictions derived from costly signaling theory.

Also called the “handicap principle” (Zahavi and Zahavi Reference Zahavi and Zahavi1997), costly signaling predicts the evolution of signaling systems that reliably communicate information about some underlying quality. As Cronk (Reference Cronk2005) points out, “costly” may be a misleading label because the essence of the principle is that the signal must be hard to produce, except by individuals who actually possess the underlying quality being advertised. For individuals possessing the quality being advertised, the advertisement may entail no great cost in energy or time, whereas individuals without the quality may be incapable of advertising it no matter how much energy or time they expend. Put another way, higher-quality individuals incur a lower marginal cost for advertising (Grafen Reference Grafen1990b; Zahavi and Zahavi Reference Zahavi and Zahavi1997).

Costly (or hard to fake) signals may be either between- or within-species. As an example of between-species signaling, Arabian babblers studied by Amotz Zahavi (Zahavi and Zahavi Reference Zahavi and Zahavi1997) fly to the tops of trees and call loudly when they see a circling raptor. Although a commonsense explanation might be that the babblers are calling to “warn” others in their group, natural selection should not produce such a group-beneficial trait because of the high cost to the caller's fitness, and in any case such an explanation is inconsistent with the specifics of the display: it is unnecessarily loud, repeated too many times, continues even in the absence of other babblers, and is sometimes a group activity rather than something done by an individual sentinel. Zahavi and Zahavi (Reference Zahavi and Zahavi1997) explain the behavior instead as a signal to the raptor that he has been spotted, which means that the necessary element of surprise has been lost. The babbler gains if the raptor moves on to other prey and the raptor gains because it avoids wasted effort. The underlying quality being conveyed by the call is the babbler's knowledge of the raptor's whereabouts. The signal's reliability is ensured because the high cost of exposure (the possibility of being caught by the raptor) ensures that babblers who issue these calls have, in fact, spotted a raptor; if babblers tried to trick the raptors by calling whether or not they had seen a specific raptor, the raptors would still have surprise on their side, and the behavior would be eliminated by selection.

Another classic example of between-species signaling is the stotting behavior of gazelles, wherein, when stalked, the gazelles repeatedly jump straight up into the air, thus making their location obvious, while at the same time expending energy that could be put toward escaping from the predator (Bleige Bird and Smith 2005; Cronk Reference Cronk2005; Zahavi and Zahavi Reference Zahavi and Zahavi1997). The costly signaling explanation of the gazelle's behavior is that it enhances the fitness of both predator and prey: the gazelle's fitness is enhanced to the extent that the display discourages a predator from initiating pursuit, which would certainly entail energy expenditure by the prey and might eventuate in the prey becoming dinner for the predator; the predator's fitness is enhanced if the gazelle's display reliably communicates underlying health and strength, thus leading the predator to avoid pursuits with a low probability of success. Although in the long run the interests of the gazelle and the predator are in conflict, there is a temporary confluence of interests if the gazelle being stalked is, in fact, capable of outrunning the predator; the selectively advantageous choice for both parties is for the predator to abandon the pursuit. One might object here (as one reviewer did) that this example leaves unspecified the mechanisms by which the predator decodes the prey's signal, but this is an anthropomorphic red herring: no decoding is needed because natural selection has coupled the stotting of the gazelle with reluctance on the part of predators to waste energy in low-probability pursuits.

Threats are a good example of within-species signals. Two rivals both benefit if the likely loser walks away from a physical contest, since both conserve energy and avoid injury. Thus, there is a mutual advantage to be gained if a threat accurately conveys information about motivation, confidence, resolve, strength, and other characteristics necessary to win a physical contest. Zahavi and Zahavi (Reference Zahavi and Zahavi1997:17–22) point out that threat postures of many species, including humans, increase vulnerability to a rival's attack, and that this cost is only bearable by highly motivated individuals who are confident that they will win. Insufficiently motivated or weak individuals do not threaten or threaten unconvincingly because the likelihood and direct physical costs of losing are too great. Tendencies to make threats without sufficient motivation and ability would be eliminated as fakers pay high fitness costs, including death, in contests that escalate to violence. This is the mechanism by which selection ensures that threats are reliable signals in the sense that there is a direct correlation between an individual's willingness to remain vulnerable in case of attack (a high cost) and his/her ability to win a physical contest.

HUMAN PRESTIGE DISPLAYS AS COSTLY SIGNALING

A common form of costly signaling in humans involves activities designed to advertise prestige (Bliege Bird and Smith Reference Bliege Bird and Smith2005; Hawkes and Bliege Bird Reference Hawkes and Bliege Bird2002), a quality that has an evolutionary basis in greater reproductive success. Driving a Ferrari, holding a potlatch, and philanthropy command attention and confer prestige because they provide reliable evidence of an individual's ability to harness resources. Individuals may have a host of idiosyncratic motivations for engaging in such activities and may deny that they have anything to do with prestige seeking or reproductive success. But, no matter what else they are intended to communicate, the actual costs of these activities guarantees that they also are true signals of resource access; only someone with real money can afford to drive a Ferrari; day laborers and university professors normally cannot.

Where do the human motivation to seek prestige and the corresponding tendency to admire and cater to prestigious individuals come from? Status in most social species, including our closest relatives, the chimpanzees, comes through dominance, which has an etiology distinct from prestige (Henrich and Gil-White Reference Henrich and Gil-White2001). Joseph Henrich and Francisco Gil-White (2001) develop an “information goods” theory of how the psychology of “prestige” or “freely conferred deference” could have evolved in the human lineage. Following Boyd and Richerson (e.g., Reference Boyd and Richerson1985), they argue that cultural transmission circumvents some of the costs of individual learning, and this explains how prestige and cultural transmission could have evolved in tandem within the human lineage. The capacity for cultural transmission, which established the complementary roles of model and learner, changed the selective environment, so that traits engineered “to extract reproductive benefit from the flow of socially transmitted information” (Henrich and Gil-White Reference Henrich and Gil-White2001:167) were favored. From the learner's point of view, such an environment favors strategies that gain access to the best models. An ability to show deference is thus favored by selection because it enhances access to more-capable models. Sexual selection enters the picture because, from the model's point of view, reproductive advantage can be gained from being emulated—more emulators equates to greater prestige, and since prestige signals quality, prestige leads to more mating opportunities. Note that “learner” and “teacher” are not invariant identities, but rather variable roles that individuals may adopt in different social situations. Because of the reproductive payoff from being recognized as skilled or knowledgeable (i.e., the reproductive payoff of prestige), selection favors efforts to accumulate and advertise prestige. In sum, Henrich and Gil-White's model posits that humans have an evolved capacity for showing deference and a corresponding propensity to exploit freely conferred deference for reproductive benefit.

Why would humans pay the costs of prestige displays when human language can effectively communicate hidden qualities? Or, in historical-evolutionary terms, once language evolved, why would selection maintain largesse, magnanimity, and other costly status displays, when underlying qualities could be communicated much more easily and cheaply in words? In fact, the low cost of language is its downfall: using language, humans can inflate, lie, and distort. Potential mates and allies would be much better advised to pay close attention to signs of quality that are guaranteed by their cost to be reliably related to quality. The interests of signalers, who benefit from the prestige associated with widespread appreciation of their superior qualities, and the interests of audience members, who benefit from guaranteed-reliable information, are both served by the wasteful, non-linguistic prestige-signaling system.

Anthropologists and other human scientists have long recognized the importance of prestige-signaling systems in human relations. As Bleige Bird and Smith (2005) point out (see also Cronk Reference Cronk2005; Zahavi and Zahavi Reference Zahavi and Zahavi1997), this theoretical tradition extends from the “practice theory” of Pierre Bourdieu (Reference Bourdieu1977) back to earlier work of Marcel Mauss (Reference Mauss1924) and Thorstein Veblen (Reference Veblen1979 [1899]). In the work of these theorists, gifts, altruistic acts, conspicuous consumption, and conspicuous generosity are viewed as means to accumulate “symbolic capital”—obligations on the part of others, respect from others, etc.

Accumulation of symbolic capital provides an incentive for senders of wasteful advertisements, but why would giving lavish gifts or other forms of conspicuous wastefulness generate any response in the receivers of the gifts? Zahavi's handicap principle, foreshadowed by Veblen's notion of “invidious comparison” provides the answer: the reliability of information conveyed, which is ensured by the actual high cost of lavish gifts and displays, is the basic incentive for receivers to participate in the signaling system (Bleige Bird and Smith 2005). Invidious comparison enables all members of an audience to gauge their own relative status and to decide whom to consider a potential model, patron, or ally.

Prestige signaling characterizes even mundane, non-elaborate aspects of human existence. For instance, careful analysis of return rates for hunters and gatherers often reveals that male hunting makes little sense in terms of direct fitness benefits, such as caloric return (Hawkes Reference Hawkes1993). Returns from hunting large animals tend to be lower than gathering or hunting small, abundant animals; large prey are typically shared, with no preference for related individuals or mates; and there is little evidence that meat sharing can be explained by reciprocity (Hawkes Reference Hawkes1993; Hawkes and Bliege Bird Reference Hawkes and Bliege Bird2002). Why would males work harder than necessary to get enough to eat, and why would they share their earnings with kin as well as non-kin, mates as well as non-mates, with little chance of repayment? Hawkes and Bliege Bird (Reference Hawkes and Bliege Bird2002; Smith and Bliege Bird Reference Smith and Bird2000) argue that hunting is better understood as a means to advertise quality, and that observers of the advertisements have a variety of incentives to pay attention to the information that relative hunting success provides. Ultimately, the payoff for the hunter is enhanced prestige and, consequently, enhanced access to mates. In archaeology, prestige competition has been hypothesized to explain an increase in large animal hunting from middle through late Holocene times in California (Hildebrandt and McGuire Reference Hildebrandt and McGuire2002), and Aldenderfer (Reference Aldenderfer, Kennett and Winterhalder2006:190) has recently suggested that herding in the Andes may have originated as “an alternative means by which a man can signal his positive qualities to potential mates and allies.”

The focus here, however, is on cultural elaboration, and if costly signaling theory is to promote understanding of cultural elaboration, it must point to explanations for the tremendous variability in elaboration seen in the ethnographic, historical, and archaeological records. One useful result (Grafen Reference Grafen1990b) is that the level of costly signaling will scale with its payoff, which is a function of the number of competitors, i.e., audience size, and with the variance in the underlying quality being signaled (see also Neiman Reference Neiman, Burton and Clark1997; Zahavi and Zahavi Reference Zahavi and Zahavi1997). Applied to human social status displays, the underlying quality is control of resources, which means that lavishness of displays must depend on resource availability and structure: more total resources available for human consumption and relatively predictable resource yields will permit population growth (increasing audience size) together with increasing variance in accumulated resources (increasing difference between haves and have-nots). Bleige Bird and Smith (2005:234) make the second point, arguing that predictability of subsistence yields enhances the reliability of subsistence output as an indicator of producer quality: under stable conditions, industrious farmers with the best land or talented salmon fishermen with the best fishing gear and best knowledge of salmon habits consistently out-perform their colleagues and are able to accumulate larger and larger bases for funding public works or other advertisements of their own quality. Over time, relatively stable subsistence systems thus give rise to more stable status hierarchies and more elaborate advertising of status.

The dependence of prestige-signaling's conspicuousness on the overall potential to accumulate wealth and the overall variance in wealth, both of which depend on the stability and productivity of the resource base, may account for the near-universal inclusion of “food surpluses” in the list of characteristics that define “complex society” (e.g., Evans Reference Evans2004). In Mesoamerica, public architecture, monumental art, and other physical manifestations of “complex societies,” which embody accumulated wealth and thus are statements about control of wealth, post-date the advent of settled villages and emergence of stable subsistence systems based predominantly on agriculture. Moreover, as I argue in greater detail below (see also Neff Reference Neff, Lauriers and Love2011a), their most salient early expressions occur in a region of prodigious and stable biotic productivity, the lowland region stretching from the Gulf Coast of Veracruz to the Pacific coast of southern Chiapas and Guatemala.

Another important consideration in understanding human prestige displays is that their specific form varies according to historical circumstances. Veblen (Reference Veblen1979 [1899]:166), perhaps the first to offer a systematic treatment of human prestige signaling, attributes this historical contingency to the forces of cultural selection:

Veblen, as clearly as any social scientist influenced by Darwin, understood that humans are products of natural selection, and this means that humans are predisposed to “conspicuous wastefulness” when it serves their own reproductive interests. However, human dispositions alone are insufficient to predict the specific manifestations of status displays (“usages and customs and methods of expenditure”), which depend for their specific form on ongoing selection of culturally inherited variants in specific historical circumstances.

Natural selection may also help determine the expression of human prestige displays. This argument has been developed by James Boone (Reference Boone1998, Reference Boone and Diehl2000; Boone and Kessler Reference Boone and Kessler1999), who suggests that if generosity toward subordinates emerges as one component of prestige signaling, it will improve the group's competitiveness in ongoing conflicts with other groups and will also increase the lineage's chances of surviving through unpredictable population bottlenecks.

Costly signaling theory is potentially useful to archaeology because it provides a set of expectations about how signals will be designed given the varying interests of senders and receivers. Similar expectations emerge, whether the design is wholly by natural selection (as in many signaling systems in nature) or partly by other “design forces,” such as human problem solving, subconscious learning, or cultural selection (Bleige Bird and Smith 2005:226; Cronk Reference Cronk2005:610). If we can identify components of the material record that may have served as signals, we can use costly signaling theory to make predictions about the conditions under which they would have been created and the historical trajectory of their expressions.

THE HANDICAP PRINCIPLE IN MESOAMERICAN PREHISTORY

Fraser Neiman (Reference Neiman, Burton and Clark1997) introduced Zahavi's understanding of the handicap principle into Mesoamerican archaeology by arguing that dated Maya monuments constitute reliable (though wasteful) advertisements of the ability of elite individuals to come out on top in political contests. (“Wasteful,” in this context, refers to an expenditure of resources that could be used for the basic biological needs of maintenance and reproduction; as noted above, expenditures on elaborate structures, like peacock's tails or pyramids, are puzzling precisely because they are wasteful in this sense.) Neiman's purpose was to provide an explicit, theory-based definition of the “Maya collapse” as shrinkage in the range of competitive quality coupled with a shift in audience size and composition. Focusing on dated monuments made it possible to test alternative hypotheses about the temporal trajectory of terminal monument dates and about the conditions that triggered the “collapse.” Beyond the dated monuments, however, many other features of the Classic Maya archaeological record qualify as energetically wasteful advertisements of the power and influence of elite individuals. In Arthur Demarest's (1992:143) words, “the role of ritual, religion, public display, and monumental propaganda was central to Classic Maya society…[the] bulk of Maya labor was spent in construction of the elaborate stages of their ceremonial centers—plazas, temples, facades, monuments, costumes, exotics, and paraphernalia—all for these ritual events.”

Although Neiman's was the first explicit use of Zahavi's evolutionary theory of costly signaling in archaeology, ideas similar to the handicap principle have some precedent in discussions of the origins of Mesoamerican civilization. John Clark and his colleagues (Blake and Clark Reference Blake, Clark and Blake1999; Clark and Blake Reference Clark, Blake, Brumfiel and Fox1994; Clark and Gosser Reference Clark, Gosser, Barnett and Hoopes1995), for instance, argue that status competition involving altruistic acts drove the emergence of rank societies on the Pacific coast of southeastern Mesoamerica during the Late Archaic and Initial Early Formative periods. The earliest (Barra phase, ca.1800 b.c.) ceramics are well-made serving vessels that would have been appropriate for serving maize beer or other liquids at feasts held to win political allies. Around 1600 b.c., some individual or faction had the wherewithal to fund the building of a 9 x 18 m structure at Paso de la Amada (Blake Reference Blake and Fowler1991). A ballcourt was built at roughly the same time (Hill et al. Reference Hill, Blake and Clark1998), thus providing a setting for political competitions analogous to those of the Classic Maya. Expenditures of effort to import ceramic technology, hold feasts, and build public facilities are “wasteful” in the sense that they are not required for reproduction, yet they entail obvious costs in terms of energy or resource expenditure. Like other kinds of cultural elaboration, they present a puzzle for evolutionary theory. The puzzle is resolved, however, by viewing such activities as reliable signals of an individual's ability to provide for those whom he would want to enlist as followers. By paying attention to the extravagance of the expenditures, potential followers gain reliable information about “aggrandizers” in the community.

A partial version of the handicap principle also appears in Kent Flannery's (1968) seminal statement of the “sister culture” perspective on the Olmec. The interaction of Gulf Coast and highland people at this time, according to Flannery, grew out of prestige-seeking behavior on both sides. Oaxaca Valley people made magnetite mirrors and exported them to Gulf lowland people “to enhance and reinforce previously-existing systems of status,” while Gulf lowland individuals made massive offerings of non-local serpentine, jade, and magnetite as a statement of their own ability to create and destroy wealth (Flannery Reference Flannery and Benson1968:107). Citing Drucker et al. (Reference Drucker, Heizer and Squier1959), Flannery (Reference Flannery and Benson1968:107–108) says of a massive buried serpentine pavement at La Venta, “the act of creating and burying such offerings, made valuable by the difficulty of their acquisition, was the important point.” In other words, such a massive, wasteful expenditure provided reliable information to the local populace about who controlled the most resources. Flannery also contends that Early Formative exchange, which started out as prestige-seeking, would have had adaptive value (positive effects on biological fitness) to the extent that it buffered occasional agricultural shortfalls.

While the insights of previous commentators, such as Flannery (Reference Flannery and Benson1968) and Clark and his collaborators (e.g., Clark and Blake Reference Clark, Blake, Brumfiel and Fox1994) converge in some respects with costly signaling theory, the promise of explicit theory is that it should provide more predictions, more precise predictions, and more novel predictions. Explicit use of costly signaling theory should move us beyond insightful summaries of empirical patterns. The starting premise is that the Early Formative archaeological record is made up in part of the remains of signaling systems created by various design forces (natural selection, decision making by humans, cultural selection, etc.) in a context of competitive interaction among individuals, groups, and factions. With some additional information about context, such as the fact that subsistence systems of the Early Formative period were productive enough to permit accumulation of surplus wealth in some regions, can we use costly signaling theory to make specific predictions about the form and distribution of archaeological remains in the Gulf lowlands and elsewhere in Mesoamerica, predictions that may move us beyond debates over the validity of the “mother culture” and “sister culture” metaphors?

THE DESIGN OF OLMEC PRESTIGE SIGNALS

MESOAMERICA'S EARLY HORIZON

The above discussion touches on a few components of the Olmec archaeological record that qualify as “wasteful advertisements,” that is, reliable signals of the power of individuals and/or factions to harness resources toward non-subsistence ends. Chronology and other elements of the context in which these items appeared were largely ignored in order to highlight design features of the artifacts and monuments, especially their cost. Yet, it is the specific historical context and the trajectory of changes that best warrant the inference that Olmec cultural elaboration was shaped by an evolutionary process that selected for costly, reliable systems for signaling wealth and political power.

Major steps in the evolution of Olmec cultural elaboration took place during the Early and Middle Formative periods, roughly 1800–500 b.c. This section summarizes the archaeological evidence for how practices of cultural elaboration evolved and how they led to the dissemination of ceramic traits that make an “Early Horizon” (Flannery and Marcus Reference Flannery and Marcus1994) or “Early Olmec Horizon” (Diehl Reference Diehl2004) recognizable in the archaeological record. The following section addresses the post-Early Horizon fate of elaborate ritual and artistic practices introduced by the Olmec.

Part of the argument of the previous section is that prestige signaling should scale with audience size and variance in the quality being advertised (control of resources in this case). Conditions in which these variables are increasing should be associated with more advertising and more elaborate advertising (Grafen Reference Grafen1990b; Neiman Reference Neiman, Burton and Clark1997; Zahavi and Zahavi Reference Zahavi and Zahavi1997). Evidence reviewed below indicates that investments in cultural elaboration during the San Lorenzo phase (ca. 1450–1100 b.c.) in the Gulf lowlands fit these expectations—the predicted increases in signaling intensity become visible in the archaeological record at a time and in a location where resources were abundant and resource structure would have favored increasing variance in access to wealth across the population (Neff Reference Neff, Lauriers and Love2011a).

Predecessors of the Gulf Olmec occupied the lowland tropical forests of eastern Mesoamerica during the Archaic period, but population densities were so low that the Archaic residents are almost impossible to detect except in the paleoenvironmental record (Neff et al. Reference Neff, Pearsall, Jones, Arroyo, Collins and Freidel2006c) or in the littoral zone, where they sometimes left large shell mounds (Voorhies Reference Voorhies2004). Shortly after 2000 b.c., during a dry interval (Neff et al. Reference Neff, Pearsall, Jones, Arroyo, Collins and Freidel2006d), these populations crossed the threshold of archaeological visibility in a dramatic way, with Early Formative house mounds appearing in a nearly unbroken chain along the Pacific coast (Arroyo Reference Arroyo, Laporte, Suásnavar and Arroyo2000; Blake et al. Reference Blake, Clark, Voorhies, Michaels, Love, Pye, Demarest and Arroyo1995; Neff et al. Reference Neff, Pearsall, Jones, Arroyo, Collins and Freidel2006c). Ecofactual, settlement pattern, and bone-isotope evidence all favor the view that this adaptive shift involved intensification of a diversified subsistence system that combined efficient exploitation of estuarine or other wetland resources and horticultural activities on the adjacent dry land. The evidence rules out a full-time commitment to maize agriculture.

As of approximately 1600 b.c., Early Formative populations with similar pottery occupied the lowland area stretching from the Gulf Coast to western El Salvador. Large residences, ballcourts, and other puzzlingly wasteful archaeological manifestations appeared in parts of this region between 1700 and 1500 b.c. As discussed above, an explanation bearing some similarities to the prestige-signaling framework being developed here has been used to account for these incipient elaborations as well as the earlier appearance of well-made ceramic containers (Clark and Blake Reference Clark, Blake, Brumfiel and Fox1994; Clark and Gosser Reference Clark, Gosser, Barnett and Hoopes1995).

Given the mixed wetland/horticultural adaptation of lowland Early Formative populations, one can expect subsistence yields to have varied from place to place as a result of geographic variation in soil productivity, rainfall, wetland hydrology, and other environmental factors (Stark Reference Stark, Clark and Pye2000:34). Mazatan, for instance, had an edge over the eastern Manchon swamp, about 100 km to the southeast, where heavy, clayey soils inhibited horticulture on dry land adjacent to the estuary (Neff et al. Reference Neff, Pearsall, Jones, Arroyo, Collins and Freidel2006c).

For several reasons, the Gulf Coast lowlands would have been favored in terms of overall biotic productivity. First, rainfall on the Gulf Coast is abundant and predictable, averaging between 2,000 and 3,000 mm annually, with less than 20% variation in rainfall totals from year to year (Vivó Escoto Reference Vivó Escoto, West and Wauchope1965:Figure 13) and a relatively short dry season (Coe and Diehl Reference Coe and Diehl1980a). Second, alluvial soils of the northern Isthmus of Tehuantepec are replenished by annual flooding of the Rio Coatzacoalcos (Coe and Diehl Reference Coe and Diehl1980a). Coe and Diehl (Reference Coe and Diehl1980a, Reference Coe and Diehl1980b) note that, unlike other regions of Mesoamerica, maize agriculture is a year-round pursuit in the vicinity of San Lorenzo, with up to four maize crops on different soil types in the region. Fourth, riverine and lacustrine resources are abundant in the rivers, oxbows, and ponds of the region. Fifth, like other lowland areas, once the forest had been partially cleared for horticulture, deer and other game would have flourished.

Not only was the Gulf lowland region favored climatically compared to neighboring regions, the time period of concern here, the latter half of the second millennium b.c., also may have been a period of relatively favorable conditions in the region. In contrast to the 2000–1500 b.c. period, the centuries following 1500 b.c., coinciding with San Lorenzo's rise to prominence, seem to have brought wetter, more stable conditions to southern Mesoamerica. The most relevant climatic record is from the Pacific coast of Guatemala (Neff et al. Reference Neff, Pearsall, Jones, Arroyo, Collins and Freidel2006d), but a high-resolution ocean sediment record from the Cariaco Basin off the north coast of Venezuela (Haug et al. Reference Haug, Hughen, Sigman, Peterson and Röhl2001) also shows a relatively favorable interval beginning shortly after 1500 b.c. Thus, while the Gulf lowlands were always favorable for human occupation, the time period of concern here would have been optimum for the growth of large populations (large audience size). The predicted association of stable, favorable climatic conditions with increased advertising can be tested more rigorously when chronological control improves at key Gulf lowland centers, especially San Lorenzo, and paleoenvironmental data of greater resolution become available from sites within the Gulf lowlands.

The structure of resources in the Gulf lowlands is also important here. In a region of high productivity, San Lorenzo stood out for its agricultural potential, especially the abundance of highly productive river-levee soils (Coe and Diehl Reference Coe and Diehl1980b:147–148; Lowe Reference Lowe, Sharer and Grove1989). Individuals and factions who controlled the best land, such as the levees around San Lorenzo thus enjoyed consistently superior access to the fundamental source of wealth in the lowlands. Additionally, key cultivars, especially maize, were evolving toward greater productivity, so that the levee-land gardens around San Lorenzo would have yielded larger and larger crops (equaling more wealth) over the years. An ameliorating and stable climate after 1500 b.c. would also have accentuated the comparative advantage of the most-productive river levee land, thus increasing the variance in wealth and social status. Finally, San Lorenzo also occupies an elevated position on the lowland landscape, and it was surrounded by water during the Early Formative (Coe and Diehl 1980; Diehl Reference Diehl2004), making it among the most defensible of Gulf lowland locations.

The abundance and predictability of resources in the Gulf lowlands favored the precocious emergence of a pattern in which wealth was accumulated and used in prestige displays. Prestige displays, always pursued by humans, and archaeologically visible in the Gulf lowlands starting around 1850 b.c. at El Manati (Ortíz and Rodríguez Reference Ortíz Ceballos, Rodríguez, Clark and Pye2000), would have increased in sumptuousness as differences in wealth accumulation and social status grew during the favorable interval after 1500 b.c. One salient, archaeologically visible outcome of this process in the Gulf lowlands is the San Lorenzo-phase archaeological record, including the sculptural corpus, which was being assembled by 1450 b.c. (Coe and Diehl 1980). Other evidence that the scale of wasteful advertising in the Gulf lowlands was intensifying shortly after 1500 b.c. include the “great ritual event” of the Macayal A Phase at El Manati previously mentioned (Ortíz and Rodríguez Reference Ortíz Ceballos, Rodríguez, Clark and Pye2000). More precise quantification of the scale of monumental activity and better control over chronology at San Lorenzo and other Gulf lowland sites will furnish additional tests of predictions derived from the costly signaling framework.

The real crux of the mother vs. sister culture debate, however, concerns the means by which a fairly uniform horizon style spread to many parts of Mesoamerica at roughly the same time that monumentality took off in the Gulf lowlands. Observers on both sides of the debate (e.g., Coe Reference Coe, Sharer and Grove1989; Flannery and Marcus Reference Flannery and Marcus1994:390; Grove Reference Grove, Sharer and Grove1989; Lowe Reference Lowe, Sharer and Grove1989; Tolstoy Reference Tolstoy, Sharer and Grove1989) recognize shared ceramic and figurine styles at many sites in Mesoamerica between about 1450 and 1200 b.c. The debate centers on the question of whether this “Early Horizon” or “Early Olmec Horizon” was spread by Gulf Olmec or invented piece-by-piece in multiple locations. The discussion in the previous section outlined a process by which a horizon style could arise out of an interregional signaling system that employed easily falsified items, such as ceramics.

The central element in the process hypothesized to have generated the Early Horizon is prestige signaling via long-distance export of ceramics from the Gulf lowlands. According to this argument, with humans as beasts of burden, long-distance transport of ceramics makes little economic sense, since the functional equivalent of any cooking, carrying, or storage vessel can be manufactured almost anywhere and supplied to local consumers without the high cost of interregional transport (Neff Reference Neff2011b). On this basis, Early Formative ceramic circulation is hypothesized to arise through inter-regional prestige-signaling.

Prior to about 1400 b.c., ceramic exchange in Mesoamerica seems to have been localized. Relevant data come from the Pacific coast, where, despite widespread ceramic similarities across Pacific coastal Chiapas and Guatemala during the initial Early Formative phases (Barra, Locona, and Ocos), instrumental neutron activation analysis (INAA) of several hundred ceramics yields no evidence of interregional ceramic circulation (unpublished data on file at the Missouri University Research Reactor [MURR] and IIRMES, CSULB). Ceramics may have traveled with people during periodic residential moves (e.g., Arnold Reference Arnold, Skibo and Feinman1999; Neff and Arroyo Reference Neff, Arroyo, Harrison, Gillespie and Peuramaki-Brown2002; Neff et al. Reference Neff, Pearsall, Jones, Arroyo, Collins and Freidel2006c), but a signal of ceramic movement over more than a few tens of kilometers cannot be detected. The earliest incontrovertible evidence of long-distance exchange so far comes from the Acapetahua Estuary region of southern Chiapas, where plain gray pottery identified by INAA as imports from San Lorenzo appears in Cherla Phase contexts associated with radiocarbon dates of around 1400 cal b.c. (Kennett et al. Reference Kennett, Neff, Voorhies, Roberts and Glascock2004). Unpublished INAA results from Cuauhtemoc, on the coastal plain south of the Acapetahua Estuary, also document San Lorenzo imports in Cherla phase contexts (Robert Rosenswig, personal communication 2005).

The trickle of ceramics out of San Lorenzo around 1400 b.c. soon turned into a surprisingly large flow. The INAA study (Blomster et al. Reference Blomster, Neff and Glascock2005; Neff et al. Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Diehl, Houston, Joyce, Lipo, Stark and Winter2006a, Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Cyphers, Diehl, Houston, Joyce, Lipo and Winter2006b) identified multiple examples of carved-incised gray and fine white pots typologically consistent with San Lorenzo Phase Calzadas Carved and Xochiltepec White that found their way from San Lorenzo to coastal Chiapas, and to highland centers like San Jose Mogote, Etlatongo, and Tlapacoya (Blomster et al. Reference Blomster, Neff and Glascock2005; Neff et al. Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Diehl, Houston, Joyce, Lipo, Stark and Winter2006a, Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Cyphers, Diehl, Houston, Joyce, Lipo and Winter2006b). This boom in ceramic exports from the Gulf lowlands coincides with the upsurge in investment in sculptural art evident at San Lorenzo, both occurring around 1450 b.c.

Like the boom in sculptural art, the boom in ceramic exports from the Gulf lowlands can be linked to ecological conditions and the resulting structure of lowland subsistence resources. The Gulf Coast was a favorable setting, the centuries following 1500 b.c. were relatively favorable, and consequently the occupants of the San Lorenzo ridge, with its highly productive river-levee soils and defensibility, found themselves relatively well off. An ameliorating climate would have permitted population to increase, thus increasing audience size, both locally and in neighboring regions of the Gulf lowlands. By around 1400 b.c., some Gulf lowland individuals had acquired the means to fund long-distance export of ceramic pots. Costly signaling theory explains why they would have exported pots once they had the means. Advertising in this way yielded sufficient payoff in the form of obligations and enhanced prestige, given the audience size (potentially obligated exchange partners) and the magnitude of the advantage they enjoyed in control over resources.

One measure of the magnitude of the Gulf lowland advantage is the lopsidedness of ceramic circulation. While substantial numbers of pots found in southern Chiapas, the Mixteca Alta, Oaxaca, and the Basin of Mexico are attributed by INAA to production centers in the vicinity of San Lorenzo, not a single sherd found at San Lorenzo could be identified as an import, and no sherds sampled anywhere else could be identified as imports from anywhere other than San Lorenzo. Examples of exports from sites other than San Lorenzo almost certainly exist, but so far they have not been detected (Neff et al. Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Diehl, Houston, Joyce, Lipo, Stark and Winter2006a). While the ceramic-provenance data alone are silent on the issue of how exactly interregional circulation of Olmec pots was funded, the fact that the only known ceramic exports are ones from the Gulf lowlands is a clear indication that interregional prestige signaling was a game dominated by Gulf lowland chiefs, especially those from San Lorenzo.

The hypothesis that Early Formative ceramic circulation originated as interregional prestige signaling also implies that locally made imitations of high-value imports should appear in places that received imports, as local potters endeavored to duplicate the valued items (Neff Reference Neff2011b). In accord with this expectation, the Olmec INAA study (Blomster et al. Reference Blomster, Neff and Glascock2005; Neff et al. Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Diehl, Houston, Joyce, Lipo, Stark and Winter2006a, Reference Neff, Blomster, Glascock, Bishop, Blackman, Coe, Cowgill, Cyphers, Diehl, Houston, Joyce, Lipo and Winter2006b) identified numerous locally made examples of carved-incised pottery along with Gulf imports in almost every sampled location. This result makes it clear that both the “mother culture” and “sister culture” metaphors contain grains of truth: It was the Gulf Olmec who initiated the export of ceramics as interregional prestige signal, so it was they who stimulated the initial wide dissemination of “Olmec” ceramic traits; at the same time, as potters in highland communities imitated elements of Olmec iconography and introduced their own variations, they helped invent Mesoamerica's “Early Horizon” piece-by-piece.

The same process that spread traits associated with the Early Horizon beyond the Gulf lowlands would have undermined the use of ceramics as prestige signals and led to a rapid decline in long-distance ceramic circulation. This leads to a further prediction, namely that the Early Horizon, while perhaps highly visible archaeologically, represented a relatively brief interval of time. Chronological control is perhaps insufficient in most cases for a definitive test of this prediction. In southern Chiapas, however, Gulf coast imports, while present in the Cherla Phase (see above), predominantly appear during the Cuadros Phase, which may last no more than about 150 years (David Cheetham, personal communication 2006).

Better tests of the prediction that Mesoamerica's Early Horizon was short lived are needed. Enhanced chronological control in key regions, such as the Valley of Oaxaca and the Basin of Mexico, will be crucial. Direct dating of ceramics, either through luminescence or the new technique of rehydroxylation (Wilson et al. Reference Wilson, Carter, Hall, Hoff, Ince, Savage, McKay and Betts2009) may someday provide a direct estimate of the longevity of the ceramic circulation pattern hypothesized to have engendered the widespread sharing of ceramic traits of the Early Horizon.

One specific prediction about chronology needs to be tested. Flannery and his colleagues (Flannery et al. Reference Flannery, Balkansky, Feinman, Grove, Marcus, Redmond, Reynolds, Sharer, Spencer and Yaeger2005; Flannery and Marcus Reference Flannery and Marcus1994, Reference Flannery and Marcus2000; Sharer et al. Reference Sharer, Balkansky, Burton, Feinman, Flannery, Grove, Marcus, Moyle, Price, Redmond, Reynolds, Rice, Spencer, Stoltman and Yaeger2006) argue that highland regions, particularly the Basin of Mexico, show greater diversity of “Olmec motifs” than San Lorenzo, and this is one basis for the claim that San Lorenzo had no special role in creating the “Early Horizon.” However, the Olmec profile heads that Flannery and colleagues use as examples of motifs that appear in the Basin of Mexico—but not at San Lorenzo—can be hypothesized to pertain to the beginning of the post-Early Olmec Horizon time period (Manantial Phase in the Basin of Mexico [Cheetham Reference Cheetham2010b; Clark 2007]). Consistent with this hypothesis, a profile-head motif also appears on a Jocotal-Phase (post-Early Olmec Horizon) sherd from Pacific coastal Guatemala (Clark and Pye Reference Clark, Pye, Clark and Pye2000). The Olmec profile faces on these ceramic examples are most similar to designs incised and carved on jadeite and other portable carved stone, which become common during the Middle Formative period (see Taube [Reference Taube, Clark and Pye2000] for numerous examples). An account of these profile “were-jaguar” representations requires a broader treatment of the evolutionary changes between Early and Middle Formative periods.

MIDDLE FORMATIVE OLMEC

As Gulf Olmec ceramic exports diminished, dissemination of ceramic decorative conventions declined as well. As a result, the widespread ceramic similarities that prompt investigators to recognize an “Early” or “Early Olmec” Horizon (Diehl Reference Diehl2004; Flannery and Marcus Reference Flannery and Marcus1994) declined, and Middle Formative ceramic assemblages show increasing regionalism. At the same time, jadeite and other carved greenstone celts, figurines, and jewelry show up in increasing frequency, both in the Gulf lowlands and elsewhere (e.g., Coe Reference Coe, Sharer and Grove1989; Diehl Reference Diehl2004; Lowe Reference Lowe, Sharer and Grove1989). In addition, Olmec-style carvings, mostly low relief on natural boulders, appear at Chalcatzingo and across a wide area of eastern Mesoamerica, from the Isthmus of Tehuantepec to El Salvador (Clark and Pye Reference Clark, Pye, Clark and Pye2000; Diehl Reference Diehl2004; Grove Reference Grove, Sharer and Grove1989).

This section focuses on jadeite and other portable carvings and the symbolic elements often carved on them. My argument is that these portable objects replaced ceramics in interregional distribution systems because they have inherent design features that make them more reliable as interregional prestige signals. Like all evolutionary arguments, this one is about trait-frequency changes, which are driven by differences in the suitability (fitness) of the variants given the social and environmental context in which they occur.

Jadeite and other greenstone have a clear advantage over ceramics as stores of wealth to be used wastefully (as prestige signals) or otherwise. Whereas ceramics can be made almost anywhere, jadeite is limited in its occurrence in Mesoamerica to the Motagua fault zone (Harlowe Reference Harlowe and Lange1993; Seitz et al. Reference Seitz, Harlow, Sisson and Taube2001), and serpentine is restricted to metamorphic zones, though it occurs more widely than jadeite. Except within source regions then, greenstones are inherently more valuable and less subject to counterfeiting than ceramics. Jadeite is also extremely hard, and shaping and carving it is time consuming. Like carved basalt monuments, the raw material for which also had to be imported, the serpentine pavements and caches of greenstone celts and figurines at La Venta are honest statements about control of wealth. Jadeite and serpentine differ from basalt, however, in being useable for portable sculptures and personal ornamentation. As a result, they could be used to broadcast a statement about wealth beyond the local La Venta community. The impossibility of counterfeiting combined with portability make portable sculptures suitable for reliably advertising prestige extra-locally, and their advantage over ceramics in this regard may be hypothesized to explain why the focus of investment in export from the Gulf lowlands shifted from ceramics to greenstone.

Olmec expenditure to secure difficult-to-fake jadeite raw material helps account for Middle Formative Olmec presence in southern Mesoamerica. The evidence for this presence includes the carved-relief “marching Olmecs” on boulders along the coast of Chiapas, at Takalik Abaj, Guatemala, and Chalchuapa, El Salvador (e.g., Clark and Pye Reference Clark, Pye, Clark and Pye2000). These sculptures trace a supply route leading to the Motagua Valley jadeite sources. Ultimately, this hypothesis may be testable through reconnaissance and discovery of evidence of Olmec mining activity (e.g., Seitz et al. Reference Seitz, Harlow, Sisson and Taube2001) and by materials-characterization research that compares Olmec jades to raw materials from extraction sites in the Motagua Valley.

Another part of the explanation for proliferation of carved jadeite and serpentine during the Middle Formative may be that greenstone became explicitly linked in the Olmec belief system with the underlying source of all wealth, maize (e.g., Coe Reference Coe, Sharer and Grove1989). Building on the work of Peter David Joralemon (Reference Joralemon1971), Karl Taube (Reference Taube and Guthrie1996, Reference Taube, Clark and Pye2000) identifies many of the Olmec profile faces of the Middle Formative as representations of the maize god (see also Coe [1989] and Schele [1996]). This iconography evinces a fundamental concern on the part of Gulf lowland people with the sources of wealth and power, mainly maize plants and rain.

The cleft-head “were-jaguar,” which Taube (Reference Taube, Clark and Pye2000) identifies as a maize god, is present in the sculptural art of San Lorenzo (e.g., Monuments 30 and 52 [Coe and Diehl 1980a]). The cleft also appears commonly on carved-incised pottery at San Lorenzo and at other Early Formative sites. Maize-god symbolism becomes much more common and much more explicit, however, during the Middle Formative period (Taube Reference Taube, Clark and Pye2000). In Middle Formative contexts, the maize god appears on sculpture, relief carvings, cave murals, jades and other portable art in the Gulf lowlands, Oaxaca, Guerrero, Morelos, the Basin of Mexico, and Pacific coastal southeastern Mesoamerica (Diehl Reference Diehl2004). Incised representations also appear on pots from Oaxaca and the Basin of Mexico as already mentioned (Flannery and Marcus Reference Flannery and Marcus1994, Reference Flannery and Marcus2000; Niederberger Reference Niederberger, Clark and Pye2000). As Taube (Reference Taube, Clark and Pye2000) points out, the proliferation of Olmec maize symbolism during the Middle Formative coincides with unmistakable evidence of a fulltime commitment to maize agriculture in the Gulf lowlands (Rust and Leyden Reference Rust, Leyden, Johannessen and Hastorf1994) and elsewhere in lowland southern Mesoamerica (Blake et al. Reference Blake, Chisholm, Clark, Mudar, Gebaur and Price1992a, Reference Blake, Chisholm, Clark, Voorhies and Love1992b, Reference Blake, Clark, Voorhies, Michaels, Love, Pye, Demarest and Arroyo1995; Rosenswig Reference Rosenswig2005). The Middle Formative also saw dramatic population growth and expansion into previously unoccupied zones in many Mesoamerican regions, a point to which I return below.

In Middle Formative representations, the cleft of the maize god is sometimes depicted with an ear of corn, or a greenstone (jade or serpentine) celt, emerging from it, and Taube (Reference Taube, Clark and Pye2000) argues that celts are components of maize symbolism. Celts are comparatively rare at San Lorenzo, but they were part of the great ritual event deposit at El Manati (Ortíz and Rodríguez Reference Ortíz Ceballos, Rodríguez, Clark and Pye2000), and they are reported from a possible San Lorenzo Olmec colony in the Mazatan zone (Cheetham Reference Cheetham2005). By the Middle Formative period, their numbers appear to increase dramatically. Massive numbers of celts and axes were excavated in Complex A at La Venta and they are commonly featured in maize god depictions (Taube Reference Taube, Clark and Pye2000).

Taube (Reference Taube, Clark and Pye2000) argues that celts, as symbolic maize ears, were storage devices for wealth generated by maize production. Under this interpretation, their interment in caches can be seen as a statement about capacity to destroy maize wealth, which also makes it an honest statement about access and ability to create maize wealth. Likewise, export of maize wealth embodied in greenstone, especially carved and incised Olmec jadeite, is an honest advertisement about the wealth of Gulf lowland individuals and factions. It is honest by constraint, because without the underlying quality—wealth—it would be impossible for an individual or faction to export the wealth embodied on carved jades. Whereas Early Formative Olmec chiefs at San Lorenzo advertised their wealth and power to foreign audiences by exporting ceramics that acquired value through the effort expended to transport them across long distances, Middle Formative chiefs at La Venta preferred to export explicit embodiments of their superior access to maize wealth.

In sum, during the Middle Formative, maize agriculture, Olmec prestige signaling, which had a long tradition in the Gulf lowlands, and Olmec maize symbolism became linked in a kind of coevolutionary relationship. Taube (Reference Taube, Clark and Pye2000) has established the symbolic linkages in this complex, whereas costly signaling theory establishes the causal linkages. The component traits of this complex were a legacy from Early Olmec Horizon times, but, as the productivity of maize agriculture increased during the Middle Formative, maize production became more and more central to the generation of wealth. In this context, advertisements about wealth and power focused increasingly on control over maize wealth, with embodied maize wealth (greenstone) being the medium for many of these statements. The memes of a belief system that centered on maize and identified greenstone as embodied maize were favored by cultural selection that arose through the efforts by Gulf lowland people to create reliable signals of wealth and power. As a result, “the power and wealth accrued by their agricultural base is readily seen even today in their monumental art, architecture, and fine objects exquisitely worked in precious and exotic materials” (Taube Reference Taube and Guthrie1996:101).

Ultimately, this coevolving complex was also favored by natural selection. Use of productive maize varieties and planting practices had direct biological fitness effects that promoted their spread. The archaeological evidence for this effect includes increased site density during the Middle Formative in many regions, particularly lowland regions of eastern Mesoamerica (Love Reference Love, Grove and Joyce1999, Reference Love2002), and penetration of agricultural populations into new areas (Robinson et al. Reference Robinson, Neff, Pye, Garnica and Powis2005). Cultural traits linked to productive maize varieties and planting practices may not necessarily have had direct, positive biological fitness effects, but linkage to increasingly productive maize agricultural practices meant that they were favored by selection too.

Favored by both cultural and natural selection, the Middle Formative package—productive maize agriculture and an elaborate ritual and symbolic system centered on maize—were passed on and became dominant features of later Mesoamerican societies. The result, as Covarrubias (Reference Covarrubias1957) and others noted some time ago, is that when one examines artistic and symbolic representations from later periods, there is a clear Olmec heritage. Coe (Reference Coe, Sharer and Grove1989) refers to this as the “continuity hypothesis.” Continuity is especially clear between Olmec maize and wealth symbolism and that of the Maya (Schele Reference Schele and Guthrie1996; Taube Reference Taube and Guthrie1996, Reference Taube, Clark and Pye2000). The monuments and architectural stages for ritual performance constructed by the Maya (Nieman 1997; see also Demarest Reference Demarest, Demarest and Conrad1992) thus represent a legacy of prestige signaling that originated in Olmec monumental art, architecture, offerings, and other forms of wasteful ostentation. Classic Maya society was not an anomalous, mysterious cult invented in the jungles of northern Guatemala 1,800 years ago. Rather, it was the outcome of a long historical evolutionary process stretching back to the mid-second millennium b.c. and to the Gulf lowland Olmec heartland.

In effect, the complex of traits assembled during the Early and Middle Formative periods in the Gulf lowlands swamped competing cultural variants. The effect is analogous to swamping of genes and language by the genes and language of expanding early agricultural populations (Diamond and Bellwood Reference Diamond and Bellwood2003). By assembling a complex of subsistence, symbolic, ritual, and material culture traits that enjoyed relative success, the Gulf Olmec became Mesoamerica's metaphorical “mother culture.”

RESUMEN

Un debate corriente entre los mesoamericanistas se centra en el papel que jugaron los olmecas de la costa del Golfo en los orígenes de la civilización mesoamericana. Por un lado, investigadores como Covarrubias (Reference Covarrubias1942, Reference Covarrubias1957), Bernal (Reference Bernal1969), Coe y Diehl (1981; Diehl y Coe 1995), y Clark (Reference Clark1997) han argumentado que las olmecas introdujeron innovaciones claves y estimularon el desarrollo de otras áreas durante el formativo temprano y, consecuentemente, contribuyeron más que otras poblaciones al subsiguiente desarrollo del arte, simbolismo, instituciones políticas y otros rasgos de las sociedades de la región. Otra punta de vista, sostenido especialmente por Flannery (Reference Flannery and Benson1968), Marcus (Reference Marcus, Sharer and Grove1989), Grove (Reference Grove, Sharer and Grove1989), y Niederberger (Reference Niederberger, Clark and Pye2000), es que la contribución de las olmecas no fue más importante que las contribuciones de otras poblaciones contemporáneas.

Algunas de las cuestiones circunvecinas a este debate se pueden resolver por medio de más investigación arqueológica, como han planteado Cheetham y Blomster recientemente en una sección especial de Ancient Mesoamérica (2010, vol. 21, no. 1). Otra estrategia para avanzar la discusión es enfocar en el desarrollo de una teoría que pueda explicar a la vez la manifestación relativamente temprana de poder político dentro de las poblaciones del Golfo junto con la distribución extensa de rasgos materiales asociados con los olmecas.

Desarrollado en de la biología (Zahavi y Zahavi 1997) tanto como en las ciencias sociales (Veblen Reference Veblen1979 [1899]), la teoría de señalamiento costoso puede explicar la inversión de recursos hacia la construcción y mantenimiento de estructuras elaboradas, cuyos costos parecieran reducir la viabilidad de los individuos que los paga. Ejemplos biológicos incluyen la cola elaborada del pavo real, y el ejemplo sociológico original de Veblen (Reference Veblen1979 [1899]) fue el consumo conspicuo de los ricos nuevos. La predicción central de la teoría es que sistemas de señalamiento para comunicar el valor de una calidad subyacente evolucionan hacia un estado en que solamente individuos que verdaderamente se caracterizan por cierto valor de la calidad pueden dar la evidencia de este valor. Por ejemplo, un individuo que tiene acceso superior a recursos podría querer que otros reconozcan sus capacidades, porque tal reconocimiento puede beneficiarlo en términos de prestigio o la deferencia mostrado por otros. Pero si todo el mundo puede hacer la misma señal, nadie pondrá atención y el sistema de señalamiento desapareciera. Por eso, la evolución favorece tendencias hacia una situación en que los individuos con los más altas valores de la característica subyacente invierten lo máximo en la advertencia de esta característica. En otras palabras, la predicción es que estructuras costosas evolucionan en el mundo biológico tanto como en el mundo cultural porque el costo asegura la veracidad de la señal en relación a la característica subyacente que es el objeto de la comunicación.

Para el período formativo temprano de Mesoamérica, esta formulación provee expectativas sobre patrones materiales que se pueden observar en el registro arqueológico de elaboración cultural en la costa del Golfo y en cambios a través del tiempo en los patrones de interacción interregional.