Metapragmatics in Archaeological Analysis: Interpreting Classic Maya Patron Deity Veneration

Abstract

Anthropological archaeology has long had an interest in understanding the semiotic properties of material objects, since it is through such objects that most archaeological analysis takes place. In recent decades, a new emphasis on materiality has focused attention on the links between material objects and social relationships in the past. In this article I argue that, just as in modern societies, the indexical meaning of material objects, and their role in social relations, were shaped by metapragmatic discourses in the past. Thus, in order to understand the role of material objects within ancient societies, it is necessary to analyze these discourses by means of historical records, their archaeological context, and analogical examples. I give an example of this method by analyzing Classic Maya inscriptions and the ways that the discourses recorded in them characterize material objects such as temples, cult objects, and tribute payments.

While many archaeologists agree that the evaluation of meaning is an essential component of the study of past societies, the ways in which meanings were established and negotiated, as well as the ways that archaeologists can access this meaning, are continually debated within the field. One of the key themes of these debates is the similarity and difference between meaning conveyed by language and that conveyed by material culture (e.g., Deetz 1967; Hodder 1982a; 1982b, 1987, 1989, 1992; Miller 1982; Wylie 1982; Shanks and Tilley 1987; Tilley 1991; Wynn 1993; Preucel and Bauer 2001; Olsen 2003; Preucel 2010). Since archaeological analysis is primarily focused on the latter, archaeologists have a profound interest in exploring the semiotic properties of objects and the methods of interpretation of such meanings.

In recent years, studies of “the social constitution of self and society by means of the object world” (Preucel 2010, 5) fall under the broad umbrella of “materiality” (e.g., Miller 2005; Tilley 2007; Maran and Stockhammer 2012). But while the term is now widely dispersed across archaeological literature, its users invoke a variety of different theoretical concepts in its application. Recently, some archaeologists have taken an approach explicitly based on Peircean semiotics (e.g., Preucel and Bauer 2001; Coben 2006; Preucel 2010; Bauer 2013), and this article follows their lead. Others, such as Knappett and Malafouris (2008), take a stance that borrows heavily from Actor-Network Theory, developed by French sociologists and science studies scholars (Callon and Latour 1981; Callon 1986; Law 1986; Latour 2005). Actor-Network Theory proposes that material objects, much like human beings, have agency. Instead of considering them as inert, these theorists see humans and objects as members of a single network of social relationships. This increased focus on material objects is an attractive position from the standpoint of archaeology, and the application of this model is epitomized by advocates of “symmetrical archaeology” (Olsen 2003, 2010; Shanks 2007; Webmoor 2007; Witmore 2007; Olsen et al. 2012). Olsen, for example, defines archaeology as “the discipline of things par excellence” and contends that we should refocus our attention on understanding “what material culture is, the ‘nature’ of it so to speak, [and] … the role it plays in human existence on a more fundamental ontological level” (2003, 89–90).

When taken to the extreme, however, this approach can be problematic. By privileging material culture (particularly durable material culture of the kind archaeologists study) it risks losing sight of other modes of meaningful human behavior, especially language. And, as argued by Bauer (2013, 3), it also risks creating a simple equation between artifacts and people. Webmoor (2007, 571), for example, argues, “San Martin orange ware is consistently found at Teotihuacan. … A symmetrical archaeology would treat Teotihuacanos and orange ware of this period as inextricable. For understanding prehistoric practice, is it helpful to distinguish the users of the ubiquitous ceramics from the ceramics themselves?” Webmoor’s point is that humans and objects can be treated as cyborgs—joint participants in social action. However, in this example, he seems to treat the totality of San Martin orange sherds as synonymous with the totality of persons at Teotihuacan rather than considering either the vast array of other categories of meaningful artifacts—both durable and ephemeral—in use at Teotihuacan or variations in the use and meaning—what Herzfeld (1992) calls “metapatterns”—of San Martin orange ware itself that might have implications for the enactment of social personae. Such considerations were the very reason that processualist and postprocessualist archaeology rejected the pots = people paradigm.

Instead, archaeologists should be aware of the ways that these different artifacts and artifact categories, taken as “signs” in the Peircean sense, interact with one another to create higher-order signs and more complex “interpretants” such as identities, social hierarchies, and ideologies. A key concept in this interpretive process is that of “metasemiotics,” in which signs such as San Martin orange ware pottery are rendered meaningful by way of other signs, such as the food and drink contained in these vessels, the norms engaged in their use, and the patterns of their distribution. And because language is such an important element of human communicative systems, “metapragmatics” is an essential component of metasemiotics.

Silverstein (1976) coined the term “metapragmatics” to refer to that realm of discourse that functions to regiment the many indexical dimensions of language by means of reference to some of the contextual aspects of the communicative situation, broadly conceived. This category of discourse includes, for example, terms that characterize reported speech by verba dicendi such as “exclaim” or “whine” as well as discourses about the proper usage of English or the aesthetic value of particular accents. Since Silverstein’s article was published, metapragmatic activity has become widely recognized as an essential element of linguistically enacted social relationships (see Lucy 1993). It has also been expanded somewhat from its original meaning to include linguistic activity that regiments nonlinguistic indexical signs (Silverstein 1993, 36). It is this expanded use of the term metapragmatics that I borrow here.

Metapragmatics has been recognized as an essential process in communication, responsible for many of the regularities of norms and behaviors observable within human societies. In this article, I argue that archaeologists should consider the role of metapragmatics in past societies in order to make inferential links between the artifacts we typically study and large-scale social phenomena such as status hierarchies and power relations. I demonstrate the utility of this approach by examining the role of metapragmatic discourses among the Classic Maya (AD 250–900). During this time, the Maya world (what is today parts of Mexico, Belize, Guatemala, Honduras, and El Salavador) was divided into a series of semiautonomous polities. Each of these communities venerated a set of local patron deities (Baron 2013). These gods were believed to inhabit physical effigies that were regularly handled, dressed, fed, and bathed. The importance of these physical objects in the negotiation of identities and power relations was established through a set of linguistic discourses that metapragmatically described the relationship between deity effigies, temples, offerings, and different categories of human beings. In order to access these discourses, my research draws upon linguistic data from Classic Period inscriptions, Colonial era documents, and ethnographic accounts (see Preucel 2010, 230–38).

I will start with a brief discussion of previous approaches to meaning in the archaeological record. I will then advocate an approach that draws on Peirce’s description of the sign and Silverstein’s account of metapragmatics. Turing to my archaeological case study, I will give a general account of patron deity veneration followed by a more in-depth analysis of its significance at one particular site. By using this evidence, I will demonstrate the productivity of a metapragmatic approach to the archaeological record: the objects most commonly left to us for archaeological study, such as pottery and architecture, existed within a much wider semiotic context that included many other signs, both linguistic and material (Herzfeld 1992). While context is a concept every archaeologist is familiar with, the challenge comes from the fact that the archaeological context left to us is a poor reflection of this much more complex system that once existed. Therefore, it is important for the archaeologist to examine all available semiotic information in order to overcome this evidential scarcity. This, in turn, allows for a fuller understanding of social relationships in the past.

Material Meanings in Archaeology

Parmentier (1997, 43–44) notes a reflex within Western culture to view linguistic signs as meaningful, decontextual, and impermanent, while material objects are seen as practical, contextual, and permanent. Within archaeology, debates about the relationship between language and material objects reflect this tendency as well as attempts to move beyond it.¹ These have been key debates within the field, especially since the transition to postprocessual archaeology (e.g., Hodder 1982a, 1982b, 1987, 1989; Miller 1982; Wylie 1982; Shanks and Tilley 1987; Tilley 1991; Wynn 1993; see Preucel 2010 for a more in-depth discussion). Many of these authors see linguistic signs as specific, arbitrary, and referential, while material signs, due to their physical nature, are ambiguous, motivated, and subconscious. In addition, it is argued, material signs’ durable nature allows them to retain meaning over time and, contradictorily, become more ambiguous over time than linguistic signs (e.g., Hodder 1989, 72–73). More recent authors (e.g., Jones 2004, 330; Olsen 2010, 59; Maran and Stockhammer 2012, 2) have also asserted that physical objects have properties that distinguish them from language and therefore render them more effective in the world in certain ways.

However, in these accounts, the differences between language and material culture are largely based on a Saussurean (1966) model of linguistic signs (Parmentier 1997, 44–46; Preucel and Bauer 2001). This model focuses on the referential function of language and treats language (langue) as an abstract system separate from its actual context of use. In the 1970s and 1980s, linguistic and cultural anthropologists, for whom language was a more central object of study, instead adopted a pragmatic approach borrowed from the writings of Charles S. Peirce (e.g., Silverstein 1976; Singer 1978; 1984; Mertz and Parmentier 1985). This semiotic anthropology made important insights that were long overlooked by archaeologists.

In Peirce’s description, a sign is “anything which is so determined by something else, called its Object, and so determines an effect upon a person, which effect I call its Interpretant, that the latter is thereby mediately determined by the former” (EP 2:478). Interpretants include feelings, thoughts, or physical reactions. Peirce (EP 2:289–99) observed that signs can be described based on certain properties, which fell into three trichotomies: the phenomenological category to which a sign belonged (potentiality, instantiation, or general principle), the sign’s relation to its object (icon, index, or symbol), and the way that this sign/object relationship is interpreted (rheme, dicent, or argument). The second trichotomy is the most well known and widely applied. Iconic signs relate to their objects by some sort of “ground” in resemblance or physical similarity. Indexical signs relate to their objects by means of spatial or temporal contiguity. Symbolic signs relate to their objects on the ground of conventional understanding. Silverstein (1976) argues that language is unique among signs in its use of a purely symbolic mode. This “semantic” function of language is the most transparent to language users: strings of phonemes (like morphemes and words) represent particular things in the world through a purely conventional understanding.

Just as it is easy to grasp the importance of the symbolic mode when considering language, it is easy to grasp the importance of indexicality when considering material objects. Hodder (1992, 204) uses the example of wallpaper, which “may provide an appropriate setting, evoke the ‘right’ atmosphere, but it does not have a specific meaning in the same way that words or sentences do.” In other words, through spatial and temporal contiguity, wallpaper indexes something about a room and its inhabitants. But Hodder mistakenly concluded that indexicality was more important for material signs than for linguistic signs.

Silverstein (1976), however, observed that indexicality is in fact an essential component of linguistic signs as well, and these indexical functions he called “pragmatic.” In fact, he argued that the realm of pragmatic function is much more vast—and socially significant—than the narrower semantic function of language. For example, by analogy to wallpaper, we can consider the pragmatic function of an accent, which indexically signifies the place of origin of a speaker. Other pragmatic functions include tone of voice to signify emotion, pronouns to refer to co-occurring objects or signify differential social ranking, volume to signify authority, and so forth. These pragmatic functions interact with, and can serve as a model for, material culture (Tambiah 1984; Munn 1986; Parmentier 1997; Keane 2003; Rosenstein 2003).

Silverstein (1976) also identified the metapragmatic function of language. Metapragmatic events take the pragmatic function of language as objects of reference. In other words, metapragmatic events comment upon the indexical function of language, with varying degrees of explicitness. While the metapragmatic function is implicit in many types of speech events, metapragmatic discourse is characterized by explicit descriptions or characterizations of pragmatic activity (Silverstein 1993). For example, metapragmatic discourse includes descriptions of reported speech via verba dicendi, such as “whine” or “shout,” and the characterization of particular linguistic usage as appropriate or inappropriate. Following Silverstein (1993, 36), the definition of metapragmatic discourse can be extended to include discourses that comment upon the indexical function of material objects. Agha (2007, 254), for example, uses the term to refer to a notice in Philadelphia taxicabs describing the appropriate attire of drivers. In this case, a linguistic artifact (a written notice) characterizes nonlinguistic artifacts (bathing suits, tank tops, etc.) as inappropriate for a specific social setting.

Metapragmatic discourses, both those that comment on language and those that comment on material culture, are in fact one of the most important means by which signs take on indexical meaning. For example, metapragmatic discourses about the proper pronunciation of English shape the interpretation of “correct” or “incorrect” speech as indexical of a speaker’s status. Vitriolic discourses surrounding the trial of George Zimmerman for the shooting of Trayvon Martin have shaped the indexical significance of hoodies and skittles in recent years.

The process by which metapragmatic discourse renders indexical signs interpretable can be understood within the framework of Peirce’s third sign trichotomy. Rhematic signs are those whose interpretant is purely qualitative—they are interpreted as signs of existence and as such are conditions of possibility for higher-order interpretants. Dicent signs are those that produce an understanding in the interpreter of a relationship between the sign and some aspect of the universe that can be judged as true or false. For example, a weather vane is a decent indexical sign, since it produces the understanding that it points in the direction of the wind. Many of the pragmatic functions of language fall within this category. For example, an accent produces within the listener an understanding of the relationship between a person’s speech and socioeconomic status (an understanding that may or may not be true). Metapragmatic discourses can also be dicent signs, for example, by producing an understanding of a relationship between certain pragmatic functions of language, such as accent, and certain qualities of a person, such as “low-class.” The same is true for material signs that index personhood such as clothing. Finally, arguments are signs that propose general norms or laws. Thus, arguments are important for understanding cultural systems, since they propose that certain behavioral norms are widely appropriate across individuals. Metapragmatic discourses can also fall within this category if they propose a general relationship between indexical signs and categories of people.

Metapragmatics is an especially important analytical category if we are to fully understand the meaning of material objects and their relationship to language (Rosenstein 2003; Preucel 2010, 256–57). Parmentier observes (1997, 43): “It is practically impossible to find a … situation where objects can be analyzed apart from linguistic meaning. Objects are implicated in language at every turn, for instance, by being themselves bearers of linguistic signs (e.g., and inscribed stone), by representing linguistic performance (e.g., a religious icon depicting Christ’s extended arm), by being labeled by language (e.g., a titled painting), by being contextually surrounded by linguistic utterances (e.g., a ritual offering), or by being produced according to textualized instructions (a technical instrument).”

This chain-like relationship between signs and their interpreting signs (or “semiosis”) was seized upon by Peirce as far more important than the relatively static, dyadic relationship between expression and meaning (as in the Saussurean model). He realized that interpretants are themselves new signs that produce new interpretants in a process of “unlimited semiosis” (Eco 1979, 68–69). This process links signs together to produce an ever-denser semiotic context in which new signs are interpreted. To quote Parmentier: “[Peirce’s] notion of semiosis as the chaining or spiraling of interpretations is critical for understanding social action as linear, i.e., as emergent in realtime and as multileveled, i.e., as a projection from metasemiotic representation. One of the methodological implications of this for anthropology is that there will be a systematic difference between observation of cultural processes in realtime and elicited after the fact of information from informants” (1997, 8).

The methodological challenge Parmentier identifies is exacerbated in the case of archaeological interpretation, since the real-time process of sign creation and interpretation is lost (Parmentier 2009, 151). This is a problem for archaeology, because “There is no way to fully analyze the organization of signifying forms or the structure of meaningful codes without studying the functioning of signs in the contexts of their use—the principle that applies just as strictly to material objects as to linguistic signs” (Parmentier 1997, 51). Of course, archaeologists are aware of the concept of context as an important aspect of interpretation (see Hodder 1986). However, the archaeological context is only a pale reflection of the original semiotic context in which material objects were produced and interpreted. This means we are operating under conditions of extreme evidential scarcity (Herzfeld 1992).

However, the process of “unlimited semiosis” also links signs and interpretants across time frames at multiple scales beyond that of “real-time” semiosis. In the archaeological record, these time scales can become compressed such that the archaeologist can observe precipitates of semiosis over hundreds of years within a single lot of a single test unit. This ability to observe long-term cultural processes is often seen as a unique advantage of archaeological research. It allows us to partially overcome the radical deficit of information by filling in the gaps with information from better-known periods. This process is typically known as “analogy” in archaeological interpretation. In effect, analogical reasoning allows the archaeologist to take signs and interpretants from a known context—usually a more recent ethnographic example—and plug them into missing sections of the earlier context. There is not space here for a full discussion of the uses and debates of analogy in archaeology (but see Clark 1951, 1953; Hawkes 1954; Binford 1961; Willey 1977; Orme 1981; Wylie 1985). Archaeologists who see broad similarities between many different cultural contexts find cross-cultural analogies appropriate. Those, like myself, who recognize the contextual and contingent nature of signs and interpretants try to restrict analogies to closely related or descendent communities.

Analogical reasoning presents a theoretical contradiction for the semiotically-oriented archaeologist. The unfolding of semiotic chains of signs and interpretants is unique at any given moment. Thus, the use of analogy can be problematic without supporting arguments demonstrating the similarity of the two contexts. But signs and interpretants do have some degree of regularity over time as a result of the long-term scale of semiosis mentioned above. This property allows them to continue to have social effects long after the moment they were created. Therefore, if the chronologically compressed archaeological record exhibits long-term continuity, we can hypothesize that this continuity is due to a relative stability of the metapragmatic discourses that regimented semiotic chains. Correspondingly, changes in the archaeological record might then reflect changes in metapragmatic activity over time (see Leone and Parmentier 2014, 18). Of course, any given society will exhibit continuities in some aspects of semiosis and changes in others. The archaeologist must therefore use this interpretive process with extreme care and recognize the limitations inherent in the analogical method.

In summary, the indexical meaning of material objects cannot be fully understood apart from linguistic signs that metapragmatically modeled their meaning for ancient language users. For archaeologists, this conclusion presents a challenge since many of these metapragmatic discourses are now gone. As Parmentier notes, “Absent other historical, contextual, or comparative information, there is no principled way to [delimit functional and symbolic parameters of objects] from the evidence provided at an archaeological site” (1997, 49). And while this observation may seem to close the door on archaeological analysis, it in fact offers three strategies for overcoming the evidential scarcity archaeologists face:

1. 1. Historical information in the form of written records is the only way of recovering ancient metapragmatic activity and should be used to its fullest extent.

2. 2. Contextual information is already familiar to archaeologists. But we should be reminded to analyze the entire material record rather than reducing it to a single artifact class such as a pottery type. A division of labor within archaeological projects helps us to accomplish this task, with experts in different types of artifact analysis sharing information.

3. 3. Comparative information in the form of direct historical analogies offers a chance to fill in missing metapragmatic information, though it should be used with caution.

These methodological strategies are already used, to one degree or another, by most archaeologists. However, when analyzed within a semiotic framework, they are much more analytically powerful. Peirce’s sign typologies offer a vocabulary with which sign functioning can be described with precision (Parmentier 2009, 142). And Silverstein’s concept of metapragmatics proposes a mechanism by which signs acquire indexical meaning cross-culturally. These concepts thus allow the archaeologist to trace linkages between artifacts and cultural norms (arguments) that shaped social interaction in ancient societies.

Case Study

The following case study presents an example of these strategies in use. I interpret the relationship between religious cult objects and categories of personhood by analyzing the metapragmatic discourses recorded in ancient historical records, considering their material co-text (archaeological context), and applying analogies from later periods where metapragmatic activity is scarce or unavailable.

The Classic Maya world (250–900 CE) was divided into a series of semi-autonomous polities, each headed by a ruler (ajaw). Some of these rulers were more influential than others, presiding over tributary networks that incorporated smaller polities and were maintained through royal marriages, visits, and gift-giving (Martin and Grube 1994). Each polity had its own set of patron gods that were believed to protect and sustain the community and to represent it in its relationships with other polities. The veneration of these patron deities had profound implications for higher-order social models related to politics and power relations between different groups and individuals of the Classic Period. So too did the metapragmatic discourses that surrounded these religious practices, including descriptions of how patron deities helped their home communities, the ways that patron deities were cared for, and the relationship between patron deities and human rulers of particular sites.

Patron deities, not unlike modern patron saints, were particular aspects of more widely recognized Maya deities, such as the Sun God or Rain God. Unlike these more generalized deities, patron gods were believed to reside in physical effigies and to actively participate in the human world. They were given credit for the safe passing of calendrical junctures, successful warfare, and the accession of rulers. The recognition of patron gods at certain sites has led some scholars to believe that they were simply deified royal ancestors (Proskouriakoff 1978, 116–17; Marcus 1983, 1992; McAnany 1995, 27; Wright 2011, 232–33). However, the epigraphic data demonstrate that patron deities and ancestors were venerated using different religious practices and that their relationship to living humans was distinct. (A full discussion of these differences is beyond the scope of this article, but see Baron [2013]). Nevertheless, ancestor veneration was significant in Classic Maya communities. As McAnany (1995) argues, Classic Maya lineages were represented by key ancestral figures whose veneration was used to differentiate these genealogy-based factional groups. The existence of these ancestral spirits conferred privileges upon their lineal descendents. But while Maya rulers also venerated ancestors, kingship was more than simply kinship writ large. Instead, kingship was essentially an economically extractive institution that appropriated, politicized, and ultimately superseded kin-based social organization.

Maya polities often had many patron deities that were listed on monuments describing ritual events. Over the course of the Classic Period, these lists often became longer. For example, at Copan, new patron deities appeared periodically in the inscriptions of the site throughout its history, while old deities continued to be venerated (Baron 2013). I suggest that one of the reasons for the introduction and accumulation of patron deities may have been to neutralize kinship-based authority in favor of royal authority by replacing ancestor cults with patron deity cults. By means of a new set of semiotic activities oriented toward caring for patron deities and a new set of metapragmatic discourses emphasizing the community-wide benefits provided by these gods, rulers linked their religious authority to community well-being. These new acts and discourses superseded the older social model, in which the privileges of potential rivals were upheld through ancestor veneration. A good example of this process can be seen at La Corona, Guatemala, and will be described below.

Patron Deities: Materiality and Metapragmatic Discourse

The material nature of patron deities was an important aspect of religious practices among the Maya. As noted above, patron deities resided in physical effigies, lived in stone temples, and received physical offerings from people. However, a full appreciation for the significance of patron deity veneration requires attention to written texts for two important reasons: first, these texts provide us with basic information that has now been lost to the archaeological record. For example, the identification of patron deity shrines is difficult, since they resemble other ritual structures. But at Palenque, Chichen Itza, Tikal, Yaxchilan, and La Corona accompanying hieroglyphic texts label certain structures as deity temples. This allows them to be excavated with the purpose of understanding ritual practices carried out in these spaces, as I did at La Corona. In addition, the preservation of patron deity effigies themselves is frustratingly rare.² This is probably because of site abandonment processes: patron deity effigies were not likely to be abandoned but rather carried away by departing residents or captured or destroyed by hostile armies. Fortunately, patron deity effigies are discussed in historical texts and depicted on monuments, thus giving valuable information about their material aspects. In addition to giving us this basic information, metapragmatic discourses specified the relationship between patron gods and certain categories of persons. Thus, they shaped the social relationships that interest anthropological archaeologists.

The hieroglyphic inscriptions described in this article all come from carved monuments. Some of these monuments, such as stelae and altars, were used in calendar rituals that took place periodically to mark ritually significant passages of time. Others, such as panels, tablets, and lintels, were built into architectural spaces. Their texts are self-referential in nature, giving the date on which the monument or building was dedicated as well as historical information leading up to the dedication. Archaeological and paleographic information indicates that Maya inscriptions were contemporaneous with the archaeological contexts where they are found, only very rarely being modified or reset in new locations. Their self-referential nature suggests that their contents may have been read aloud during dedication ceremonies, although this is impossible to confirm. Equally difficult to assess is the intended audience of these texts—whether human or supernatural—given the sketchy evidence for ancient Maya literacy rates and the accessibility of certain buildings or spaces. However, the ideological discourses they contain, it is safe to assume, mirror those spoken aloud by the rulers who commissioned them.

To these Classic Period hieroglyphic texts, we can add discourses from later periods. These include Maya-authored documents that were brought to light after the Conquest and generally reflect Postclassic practices and ethnographically observed discourses in modern Maya communities. There is good evidence to assert a degree of religious continuity between these different periods. Similar veneration practices for patron deities (now patron saints) can be observed, and Spanish Colonial accounts describe continuities during the period of conversion (see Baron 2013). As I will discuss, this continuity appears to be the result of metapragmatic discourses that have persisted over time, linking the well-being of Maya communities, patron deity veneration practices, and political authority.

Patron deity effigies in the Classic and Postclassic Periods, as well as modern patron saints, are particularly potent indexical signs, since they embody supernatural forces usually understood to be invisible (see Leone and Parmentier 2014). Classic Period monuments depict rulers holding effigies aloft and deity heads emerging from ceremonial royal attire. Larger effigies set on float-like palanquins, such as those depicted at La Corona, Piedras Negras, and Tikal, accompanied armies to war to confer supernatural protection (Martin 1996, 2000). Similar practices during the Postclassic Period are described in the Popol Vuh (Christenson 2003, 248) and Annals of the Kaqchikels (Otzoy 1999, 178). If defeated, the effigies were either destroyed, as recorded at Aguateca and possibly Quirigua (Grube et al. 1991; Looper 1999, 268), or captured by the enemy and brought back to the victorious city, as depicted at Tikal (Martin 1996, 2000). Effigies were also brought to peaceful gatherings between polities: an inscription at Palenque refers to a banquet at which foreign gods were guests of honor. The Title of the Lords of Totonicapan describes a similar gathering during the Postclassic Period (Camack and Mondloch 1983, 196). In modern times, patron saints in Maya communities are carried to visit other towns during feast days (Siegel 1941, 72; Redfield and Villa Rojas 1962, 153; Cancian 1965, 39; Vogt 1973, 101; Watanabe 1992, 114). In all of these contexts, effigies serve as dicent signs, indexing the relationship between the deity and the effigy handler. For example, when rulers held effigies aloft, this act signified that there was a social relationship between god and ruler. Similarly, when effigies were carried into battle or taken on friendly diplomatic missions, the effigy’s presence signified the social relationship between god and home community. But these important social relationships were signified in other ways as well. As we will see, the relationship between god and community, as well as the relationship between god and ruler, were signified through additional ritual acts as well as metapragmatic discourses.

The Classic Maya had good reason to be concerned with the relationship between patron deities and the community. From an outsider’s perspective, it is easy to forget the powers attributed to these gods: through their physical co-presence, patron deities were believed to bring about success in war, oversee critical calendrical junctures, and preside over political transitions. These events benefited Maya communities by providing safety from violence, cosmological stability, and political continuity. In the Popol Vuh, Postclassic patron gods were believed to provide agricultural abundance, human fertility, security, and protection from shame, misfortune, injury, and illness (Christenson 2003, 289–90). Today highland Maya community members say that their patron saints give them general protection and well-being (e.g. Reina 1966, 122). In Santiago Chimaltenango, for example, it is said that the saint protected the community from the worst atrocities of the Guatemalan Civil War (Watanabe 1990, 134).

The indexical link between patron deities and community well-being was characterized in a set of metapragmatic discourses that describe the protective power of the gods. Their powers are discussed on hieroglyphic monuments using active verbs such as ukabjiiy, “to make happen.” This differentiates patron deities from deceased ancestors, who are not credited with direct actions in hieroglyphic texts. At Copan, the patron deities are described as koknoom Ux Witik (“the guardians of Copan”; Lacadena and Wichmann 2004, 106). This is significant not only in that it expresses supernatural protection but because the object of that protection is a toponym—a place itself, and by extension, all the people who live there. In other words, these patron gods were said to protect the whole population, not just its rulers or elites. Like acts of parading and holding effigies aloft, these discourses can be understood as dicent signs that used the symbolic mode of language to refer to the important benefits provided by patron gods for their communities.

But patron deities did not simply perform these functions at any time for anyone. First, they had to be brought ritually into a special relationship with a particular community. Then they had to be made materially manifest, and, finally, the effigies had to be continually maintained and cared for through a variety of special rituals. And while metapragmatic discourses emphasized the community-wide benefits of these gods, they also made it clear that the duty of creating and maintaining effigies belonged to rulers alone. At Palenque, inscriptions from the Cross Group are particularly descriptive of mythological narratives surrounding the site’s patron deities. The initial coming of the patron gods to Palenque is described as an “arrival,” “birth,” and “earth-touching.” These three terms, apparently interchangeable, describe the process by which general deities, in this case the God of the Rising Sun, the God of the Daytime Sun, and the God of Lightning, became local patrons at Palenque (Lounsbury 1980, 112–13; Stuart 2006, 173–74). This descent is brought about by the mythological founder of the Palenque dynasty. At Tikal, a set of wooden lintels describe battles in which the ruler of Tikal captures the patron gods of Calakmul, Naranjo, and El Peru (Martin 1996, 2000). The ruler undertakes a series of rituals to domesticate them, turning them into local patrons. These include constructing temples, dancing, and god impersonation.

The most important ritual for making patron deities manifest as effigies was the tzak rite, usually translated as “conjuring.” The verb tzak, however, more accurately means “to grab hold of” in the sense of hunting or capturing (Kaufman and Justeson 2003, 903; Hull 2005, 108). The glyph itself depicts a hand grasping a fish, conveying the sense of catching something slippery. Thus, the descriptions of this ritual characterize patron gods as materially evasive, needing to be caught in order to enter the effigies created for them. Only rulers are described as performing this ritual. Metapragmatic discourses therefore link the god’s capacity for divine intervention with the ruler’s ritual acts.

Once the deities had been made manifest, rulers were also responsible for the continued maintenance of their effigies. Texts from Copan, Naranjo, Piedras Negras, Tikal, and Tonina all describe rulers bathing patron deities (Stuart et al. 1999, 50). An inscription from Palenque gives a long history of the polity in which ruler after ruler gave gifts of clothing and jewels to the gods (Macri 1988, 116–17). The Title of the Lords of Totonicapan describes a similar ritual during the Postclassic Period in which the god Tohil was dressed by the ruler’s son and heir (Camack and Mondloch 1983, 196).

Classic Period hieroglyphic texts from Naranjo, Palenque, Tortuguero, and Yaxchilan also describe rulers feeding patron gods. But here we see an interesting discrepancy between metapragmatic discourses—which credit rulers alone with the care for patron deity effigies—and archaeological and historical evidence suggesting that whole communities participated in their feeding. My excavations at La Corona, along with excavations by other project members (Acuña 2006, 2009; Fernández 2011; Patterson et al. 2012; Ponce and Cajas 2012; Perla Barrera 2013), revealed extensive refuse deposits on and near patron deity temples. This refuse included thousands of ceramics, animal bones, plant remains, and other materials indicating that large-scale food consumption occurred in this area (Fridberg and Cagnato 2012). Ceramics included polychromes usually associated with elites, as well as more common wares associated with lower classes (Parris, personal communication 2013). At the time of the conquest, Duran (1994, 128) described community-wide feasts for patron deities throughout Mesoamerica. In colonial and modern times, patron saint fiestas continue to involve the consumption of large quantities of food by community members (Wisdom 1940, 376; Reina 1966, 115).

This widespread devotion toward patron gods is probably related to the belief in their protective powers over the whole community. It may have been impossible for rulers to prevent community members from giving offerings, or they may have actively encouraged participation in god cults. Nevertheless, metapragmatic discourses recorded on hieroglyphic monuments take care to never mention these community-wide devotions. Furthermore, inscriptions from Yaxchilan and Tortuguero specify the special liquids offered by rulers to gods: chocolate, pulque, and the blood of sacrificed enemies, all of which would have been difficult for nonelites to offer.

Other hieroglyphic texts go even farther to emphasize the unique, affective relationship between rulers and patron gods. For example, the loving care provided by rulers for their deity effigies was described on inscriptions from Palenque and Tikal using the kinship trope ubaah ujuntahn, “his precious one,” usually reserved for descriptions of mother-child relationships (Houston and Stuart 1996, 294). Similarly, the phrase ubaah uch’ab yak’abil, usually expressing the spiritual link between fathers and sons, is used to describe the relationship between rulers and gods on monuments from Caracol, La Corona, and Palenque. Another inscription from Palenque describes a ritual in which a ruler gave gifts to his gods. The phrase utimiw yohl uk’uhil, “he satisfies the hearts of his gods,” suggests an emotional relationship between them (Houston et al. 2006, 189).

All of these metapragmatic discourses—statements of the ruler’s role in taming and making deities physically manifest, statements of the ruler’s continuing responsibility for effigy maintenance, and descriptions of the close, personal relationship between ruler and god—can be understood as dicent signs that characterize the ruler’s unique role in the community’s supernatural protection. What did these metapragmatic statements accomplish for the rulers who commissioned them? Very likely they formed the basis of an argument—a sign understood as a general law—that justified their hierarchical position over other community members. This argument and its logic are never stated explicitly in any hieroglyphic monuments. But they can be reconstructed by means of later metapragmatic discourses and from some additional clues in hieroglyphic texts.

The Popol Vuh describes the responsibilities of the Postclassic K’iche rulers toward their patron gods Tohil, Auilix, and Hacavitz. Just as with hieroglyphic discourses, this text describes how the proper veneration of these gods was the responsibility of rulers, who had to undergo difficult fasts. During these fasts, which could last for close to a year in some cases, they made offerings to the gods, abstained from eating maize, and observed sexual abstinence: “Thus it was that the lords fasted during the nine score days, the thirteen score days, and the seventeen score days as well. They fasted often, crying out in their hearts on behalf of their vassals and servants, as well as on behalf of all their women and children. Thus each of the lords carried out his obligations. This was their way of showing veneration for their lordship. … In unity they would go forth to bear the burden of the K’iches. For this was done for all” (Christenson 2003, 290–91).

In this passage, the writer emphasizes the unique ability of lords to carry out ritual obligations towards patron gods, the arduousness of these obligations, and the willingness and selflessness with which they were performed on behalf of the people. The text continues on to tell of the reward received by the K’iche rulers for their ritual service: “Great was the price that the nations gave in return. They sent jade and precious metal. … They sent precious gems and glittering stones. They sent as well cotinga feathers, oriole feathers, and the feathers of red birds” (Christenson 2003, 291). In return, the people paid the lords tribute items, sustaining them with wealth. This text is an example of a dicent sign—in which the relationship between gods, rulers, and people is laid out. But it suggests a wider argument: the right of rulers to tribute is a reciprocal exchange for their ritual service.

Although most modern Maya communities have no specific knowledge of the contents of Classic Period texts or Colonial Era documents such as the Popol Vuh, a similar argument appears to have survived to the modern era and is instantiated through similar metapragmatic discourses. Hereditary rulership has now been replaced in many communities with rotating cargo positions held by cofradía members. For example, in his ethnography of Chinautla, Guatemala, Reina (1966) describes the power of the cofradía members, who have influence over “the proper action of an Indian mayor, the proper means of framing the behavior of young people, the controlling of the power of public school teachers, and so on” (Reina 1966, 94). Just as in the passage in the Popol Vuh, this authority is framed as a reciprocal exchange for their selfless willingness to take on the burden of ritual responsibility toward the patron saint. When advising new cofradía members, the tatapish (ritual specialist) admonishes: “All you children are placed here to take care of the saint in the cofradía. Everybody must comply with the obligation toward the saints, and the cofradía should be carried on with desire. Everybody must do it because it is an old tradition. It is a custom that cannot be taken away; our ancestors left this tradition to us, and I am passing it on to you. The people of Chinautla must die with this custom. Besides, the Niño of Chinautla [the patron saint] sees you well when the custom is observed” (Reina 1966, 119–20). This metapragmatic discourse—a dicent symbolic sign—also suggests a wider argument that the willingness to take on materially enacted ritual responsibilities justifiably results in the privileges of cofradía members.

Using these examples as analogies, I propose that a similar argument existed during the Classic Period. This is supported by hieroglyphic texts that seem to reference reciprocal relations between deities, rulers, and common people. This is accomplished by three discursive techniques that frame the relationship between patron god and ruler as similar to that between ruler and subject. The first is the use of the term yichonal (overseen by), usually describing the relationship between rulers and subjects, as a trope to describe the relationship between patron deities and rulers (see Houston and Stuart 1996, 301; Zender 2007; Stone and Zender 2011, 59). Second, inscriptions often refer to patron gods as ajaws (rulers), either as part of the deity’s name or with special titles or accession statements. Finally, rulers are also referred to as k’uhul (god-like; Ringle 1988). When taken together with other discourses I have described, I believe that these three strategies were part of an argument justifying the ruler’s right to tribute and authority:

1. 1. Patron deities served the whole community, for example as “guardians of Copan,” or in bringing about success in war, political stability, or the passage of time. In return, they were entitled to sacrifices, gifts, and temples.

2. 2. The needs of patron deities were the responsibility of rulers. Rulers alone are described bathing, dressing, and feeding deity effigies. This is because they had a special relationship with these gods, described using kinship tropes and references to emotional satisfaction.

3. 3. Therefore, just as gods were entitled to rewards such as jewels, clothing, and chocolate for their spiritual protection, so too rulers were entitled to compensation for their unique ritual service to the gods. As such, rulers were “god-like” (k’uhul), gods were “lord-like” (ajaw), and each had oversight (yichonal) over his own immediate subordinates.

The reciprocity argument created through metapragmatic discourses subverted the system of lineage-based rights. Within the lineage system, rights to resources and authority belonged not to the ruler specifically but to particular lineages through inheritance (McAnany 1995). Ancestor veneration rituals can thus be understood as dicent signs that indexed the ancestral links between the living and the dead, thereby conferring hereditary rights on the practitioners. This meant that any given Maya polity could contain competing elite lineages, each performing its own ancestor veneration rituals in order to make claims to land and authority. Discourses related to patron deities, on the other hand, characterized the ruler as distinct from all other elite lineages with respect to his special ritual obligations toward the patron gods who, unlike ancestors, served the entire community. A good example of the tension between a lineage-based model and a patron-deity based model of authority can be seen at La Corona, Guatemala.

La Corona, Guatemala

La Corona was the location of a medium-sized polity, itself periodically beholden to a more powerful polity based at Calakmul, Mexico. La Corona has both a rich archaeological record of patron deity veneration and a detailed historical record that can be compared to this material evidence (Canuto and Barrientos 2011). When these data are examined together, it becomes evident that patron deity veneration played an important role in the negotiation of political relationships within the community.

Most of La Corona’s monuments were commissioned by one lineage that tried to present a narrative of unbroken authority stretching back into the distant past. This lineage apparently traced its origin to a mythological location known as “Six Nothing Place,” and in two separate inscriptions this phrase is incorporated into the titles of rulers from this family. For this reason, I will refer to them as the “Six Nothing Lineage.” However, the historical and archaeological record suggests that there was an additional lineage or multiple lineages that competed, throughout the site’s history, for political control and legitimacy.

Nearly all of the known monuments of La Corona were commissioned by members of the Six Nothing Lineage and thus contain a rather one-sided representation of the site’s history. The Six Nothing Lineage also had a close political alliance with the rulers of the Kaan dynasty, who ruled from Calakmul (Canuto and Barrientos 2013). This alliance was reinforced by marriages between the two ruling families (Martin 2008; fig. 1). Rulers from other lineages at La Corona apparently did not maintain this friendship with the Kaan dynasty. The rivalry between these families may explain many of the events at the site, especially those relating to patron deity veneration.

Figure 1. Family relationships of La Corona’s rulers. Illustration by the author.

I excavated four patron deity temples between 2008 and 2012 under the auspices of the La Corona Regional Archaeology Project (Proyecto Regional Arqueológico La Corona). The goal was to understand their construction history and the ritual practices associated with patron deity veneration (Baron 2013). These temples form a line along the eastern side of the site’s 13R Group Plaza, also known as the “Coronitas” Group (fig. 2). This architectural group also contains a number of other structures, including an elite residence, another pyramidal platform, and several auxiliary structures, also under investigation by the La Corona Project.

Figure 2. Map of the Coronitas plaza (after map by Marken in Canuto et al. [2006], fig. 4)

Structure 13R-5, the southernmost of the four patron deity temples, was a small platform with a masonry superstructure, approximately five meters high (fig. 3). An accompanying hieroglyphic panel, La Corona Panel 1, was discovered by Marcello Canuto in 2005. It records that this structure was a temple (wayib) dedicated in CE 677 by the Six Nothing Lineage ruler, K’inich [?] Yook, for a patron deity called [?] Winik Ub (Guenter 2005). The panel also makes reference to the dedication of three other patron deity temples in the year AD 658 by K’inich [?] Yook’s father, ruler Chakaw Nahb Chan (Guenter 2005). These three temples probably correspond to the adjacent Structures 13R-2, 13R-3, and 13R-4, each approximately nine meters high.

Figure 3. North profile of Structure 13R-5. Illustration by the author.

My excavations on all four of these structures revealed a number of sequential construction phases that spanned the history of the site (table 1). Structures 13R-2, 13R-3, and 13R-4 had a similar occupation history, with three main construction phases each, while Structure 13R-5 had just two main phases (Canuto 2006), one early and one late, that book-ended the sequence of the other three temples. I gave each of these construction phases a unique identifying name, which I will refer to in the text and in accompanying figures for reference.³

Table 1. Summary of Construction Phases of La Corona’s Patron Deity Temples

The very earliest construction on the four temples was the first phase of Structure 13R-5 (fig. 3). I have named this the K’inich Phase. It was a low platform below which a tomb was carved into bedrock (Canuto 2006). Ceramics date this construction to the third or fourth century CE. This is one of the oldest structures known at La Corona, probably dating to the first generation of the site’s occupation. The tomb from this phase was looted, but some human bone was recovered—painted reverentially with red pigment—as well as small pieces of carved jade. These artifacts suggest an elevated status of the tomb’s occupant.

The later phase of Structure 13R-5, called the Ub Phase, contained Panel 1, which carries the earliest known historical date at La Corona. This panel was commissioned by Six Nothing Lineage ruler K’inich [?] Yook, and it describes how a person named Tahn K’inich Lajua’ came to La Corona in CE 314. The use of the term K’inich in his name, meaning “radiant,” suggests an elevated status, as it refers to the Sun God and is often used in the names of rulers. (Somewhat confusingly, this term is found in the names of two different La Corona rulers described here.) Given the early date, Tahn K’inich Lajua’ may have been an early settler at La Corona, perhaps some sort of community founder. He is most likely the occupant of the early third- or fourth-century tomb in K’inich Phase of this structure.

However, Panel 1 describes the arrival of Tahn K’inich Lajua’ not with the verb huli (“to arrive here”) but rather tali (“to come in this direction from somewhere else”). Typically, the verb huli carries connotations of dynastic founding or arrivals of great political import and would have been an appropriate choice to describe the arrival of Tahn K’inich Lajua’. The fact that a different verb was chosen for this inscription indicates that K’inich [?] Yook, when he commissioned the panel, denied Tahn K’inich Lajua’ the status of dynastic founder, while still acknowledging his early arrival and high social status.

Instead, K’inich [?] Yook and the rest of the Six Nothing Lineage trace their dynastic founding to completely different events. The all-important verb huli is used to describe arrivals of princesses from the Kaan Dynasty, the Six Nothing Lineage’s political patron. Three such princesses arrived throughout the history of the site, all of them marrying rulers from the Six Nothing Lineage, including K’inich [?] Yook himself (Martin 2008; fig. 1). The use of huli indicates that Lineage A saw these women as the origin of their dynastic authority, not the early settler Tahn K’inich Lajua’.

There is reason to believe that Tahn K’inich Lajua’ did not belong to the Six Nothing Lineage at all, but rather to a different family. Not only is he denied the status of dynastic founder on Panel 1, but he is also denied the status of earliest ruler of the site. Instead, the panel describes the coming of a person to La Corona in 3805 BCE (long before the actual occupation of the site). This mythical early date is thus contrived, probably for the purpose of contrasting it with the relatively recent dates of Tahn K’inich Lajua’. This ancient person is said to have come from the Six Nothing Place, the mythological location associated with the Six Nothing Lineage. The text of Panel 1 implies, therefore, that the Six Nothing Lineage had a far greater antiquity at La Corona than Tahn K’inich Lajua’. Furthermore, the god for whom the Ub Phase was built, [?] Winik Ub, is described as a “Six Nothing Place God,” apparently intimately connected to the mythical ancestor and to the entire Six Nothing Lineage. Thus, the funerary shrine of Tahn K’inich Lajua’ was replaced by a shrine for a deity probably associated with a rival lineage.

After the K’inich Phase, the next earliest construction program identified belonged to Structures 13R-2, 13R-3, and 13R-4 and is called the Mam Phase (Baron 2011, 2012; fig. 4). It consisted of a series of three meter high platforms dating to the mid-sixth century. They have not been thoroughly explored, however. On top of these platforms, a series of slightly later platforms were built; these are called the Muk Phase. The earliest of these belongs to Structure 13R-4, built around 550 CE, followed by 13R-3, and finally 13R-2, built around 600 CE (Baron 2009, 2011, 2012; see fig. 4). These were substantial, eight meter platforms covering tombs. (Burials have been discovered in Structures 13R-2 and 13R-4. Another likely one exists in Structure 13R-3, but excavations were not attempted due to safety concerns.) Their close proximity and formal similarity to the tomb of Tahn K’inich Lajua’ suggests that these structures served as a necropolis for his lineage.

Figure 4. South profile of Structure 13R-2. Illustration by the author.

Only the tomb from Structure 13R-2 remained unlooted. Archaeological evidence indicates that its occupant was a ruler of the site (Baron 2012). Grave goods include fifteen high-quality ceramic vessels, hundreds of local freshwater shells, a turtle, and a crocodile (Baron et al. 2011); a decomposed headdress, of which remained pieces of marine shell; a woven mat—often linked to Maya rulership—that left an impression in the dried mud of the tomb roof; and a deposit of 20,000–30,000 lithic flakes, similar to those found in other royal Maya burials (Moholy-Nagy 1997; Demarest et al. 2003). The large funerary temple associated with the tomb (the Muk Phase platform) is another index of royal status.

The tomb below 13R-4, though looted, came from a similar architectural context and also likely contained the remains of a ruler of the site. Significantly, the monuments of the Six Nothing Lineage are silent about the latter half of the sixth century, when these rulers would have been buried. Along with their close proximity and similarity to the burial of Tahn K’inich Lajua’, this further supports the conclusion that they were not members of the Six Nothing Lineage.

Soon after the final Muk Phase burial in Structure 13R-2, hieroglyphic inscriptions indicate that a Six Nothing Lineage member named Sak Maas came to power at La Corona in 625 CE. However, he did not enjoy unchallenged political authority. In 655, K’uk’ Ajaw, an apparent usurper, came to power, and in the following year Sak Maas died violently “with the edge of a stone” (Grube et al. 2002, 85). His coup d’état was short lived, however, and K’uk’ Ajaw met his own violent end in 658 on the same day that Six Nothing Lineage member Chakaw Nahb Chan, son of Sak Maas, came to power.

K’uk’ Ajaw’s brief challenge to the Six Nothing Lineage’s authority might suggest that he was a member of Tahn K’inich Lajua’’s lineage. The actions of Chakaw Nahb Chan immediately after deposing him support this conclusion as well. His first act as ruler, as described on Panel 1, was the dedication of three shrines for new patron gods “Newness Lord,” “Yellow Chaak,” and “Great Temple Chaak”—the latter two being aspects of the Rain god (Guenter 2005). This dedication took place only thirty-five days after his accession and corresponds to the K’uh Phase of Structures 13R-2, 13R-3, and 13R-4 (Baron 2012; fig. 4). While the funerary shrines of the Muk Phase had been built sequentially, the K’uh Phase was a single continuous architectural program. On all three structures, it consisted of a thin layer of construction between one and two meters thick. It does not appear to have continued onto the back of the structures, but rather is only present on the front, adding new staircases and new floors of the superstructures. The superstructures themselves were constructed from perishable materials. These features are all consistent with a hurried construction effort that took just thirty-five days to complete. Ceramics from the K’uh Phase are consistent with a date in the early to mid-seventh century.

McAnany (1995, 150) has suggested that the destruction of ancestor shrines of other lineages was an effective way for emergent rulers to negate the claims of potential rivals. This is because ancestor shrines indexically linked surviving descendents to the privileges of previous generations, and their destruction therefore severed these links. In addition, as I have discussed, metapragmatic discourses linked the veneration of patron deities to the protection of the whole community and to the exclusive rights of rulers. Thus, the construction of the K’uh Phase of Structures 13R-2, 13R-3, and 13R-4 was a dicent sign that indexed both Chakaw Nahb Chan’s elevated status and proclaimed the illegitimacy of other claims to the throne.

In 667, Chakaw Nahb Chan died, and his son K’inich [?] Yook acceded to power. In 677, as described above, he completed the Ub Phase of Structure 13R-5 and its accompanying Panel 1 (Baron 2012; Canuto et al. 2006; fig. 3). The text of Panel 1 explains that the rededicated shrine was now the temple of the “Six Nothing Place God.” Like the K’uh Phase of Structures 13R-2, 13R-3, and 13R-4, therefore, the Ub Phase also replaced the 13R-5 ancestor shrine of Tahn K’inich Lajua’ with a new patron deity temple closely associated with the Six Nothing Lineage. To justify this act, the inscription compares the relatively recent Tahn K’inich Lajua’ to the ancient mythological ancestor of the Six Nothing Lineage, who is said to have come to La Corona in 3805 BCE, long before the site was actually occupied. Thus, the panel claims a greater antiquity at La Corona for K’inich [?] Yook’s own lineage rather than the one that probably actually founded the community in the fourth century. And it uses this claim to prior antiquity to justify the replacement of Tahn K’inich Lajua’’s ancestor shrine with a temple for a patron deity associated with the Six Nothing Lineage.

The archaeological and hieroglyphic record from Structures 13R-2, 13R-3, 13R-4, and 13R-5 reveals a series of ritual acts and discourses spanning the site’s history. These acts and discourses served as dicent signs with which rival lineages at La Corona supported competing arguments about authority. According to a lineage-based argument, rulership and its privileges are passed down hereditarily to the descendents who venerate deceased rulers through temple construction and burial offerings. According to the patron deity-based argument, royal authority and tribute is instead a reward for ritual service to supernatural protectors of the entire community. The Six Nothing Lineage apparently hedged its bets—sometimes emphasizing a patron-deity based model, at other times emphasizing the deep antiquity of Six Nothing ancestors at La Corona.

Rulers of La Corona and other Maya communities instantiated the patron deity-based argument for authority through metapragmatic discourses—dicent signs that used language to signify the supernatural protection afforded by patron gods to the whole community and the special qualifications of rulers to venerate the gods on its behalf. These discourses in turn shaped the meaning of material signs such as deity effigies and temples, which indexically linked the ruler to supernatural forces. They also shaped the meaning of other material signs such as ceramic offering bowls and tribute items—artifacts not obviously related to patron deity veneration in the absence of linguistic clues. Thus, an analysis of the metapragmatic work done by these discourses allows the archaeologists to understand the semiotic links between the artifacts that we readily collect to arguments—social models about the reciprocal obligations between persons—that existed in ancient societies like the Classic Maya.

Conclusion

Because archaeological analysis is largely restricted to durable objects, archaeologists have an obvious interest in understanding the ways that these objects convey meaning and help humans establish social relationships. However, it is easy to forget that the meaning of these objects was shaped through accompanying metapragmatic discourses. This is because all language users, archaeologists included, are often unaware of the long-term metapragmatic effects of language, although these effects are highly palpable in the moment of interaction (Silverstein 1976; Agha 2007). Archaeological analysis can be greatly enriched if due attention is given to the unique and emergent ways that material objects and linguistic signs interacted with one another ancient contexts.

Such an approach may seem antithetical to some archaeologists, for whom archaeology is the “the discipline of things par excellence” (Olsen 2003, 89). But in fact this approach merely proposes an expanded definition of material things. As Agha (2007, 2–3) points out, utterances and discourses are only perceivable to us because of their materiality—whether through sound waves, ink on paper, or in this case carved in stone. And although spoken language quickly disappears, this is not a question of its materiality, but rather its durability. Humans take steps to counteract the ephemerality of speech through repetition and written records. This long-term preservation of discourse is obvious in the case of Maya patron deity veneration: similar metapragmatic discourses are apparent in Classic, Postclassic, and modern contexts.

Linguistic and cultural anthropologists, who have access to the real-time unfolding of semiotic behavior, can easily observe the interrelationships between durable material objects and linguistic signs. Unfortunately, archaeologists operate under conditions where both linguistic utterances and a good deal of nonlinguistic material objects have been permanently lost. For this reason, archaeology cannot be seen uniquely as a discipline of things but rather a discipline of radical evidential scarcity (Herzfeld 1992). In this article, I have proposed, following Parmentier’s remarks (1997, 49), that we can offset this deficit in part by means of methods already familiar to many archaeologists: careful attention to context, the use of historical texts, and the use of ethnographic analogies. These techniques, coupled with theoretical model of metapragmatics informed by semiotic anthropology, allow archaeologists to trace the linkages between common artifacts and the social relationships and cultural norms of ancient societies. I have used patron deity veneration among the Classic Maya to demonstrate this process.

I realize that not all archaeologists have access to the same amount of historical information available in the Maya area. And for this reason, it is likely that some semiotic processes of the past are beyond all recovery. This is to be expected however—in all scientific disciplines the likelihood of full analysis and explanation diminishes as available data decreases. The ethnographic analogies I advocate here alleviate this problem to a certain extent, but they can never themselves replace the data lost to the passage of time. Archaeology is a challenging discipline, and the best efforts of archaeologists cannot always ensure a successful interpretation of past semiotic events. However, our efforts can be aided by understanding the ways that linguistic and nonlinguistic signs interact with one another in human social relationships.