Quirigua in the Maya World

I was naturally anxious and expectant on this my first visit to a Central-American ruin, but it seemed as though my curiosity would be ill satisfied, for all I could see on arrival was what appeared to be three moss-grown stumps of dead trees covered over with a tangle of creepers and parasitic plants, around which the undergrowth had been cleared away for the space of a few feet. However, a closer inspection showed that these were no tree-stumps but undoubtedly stone monuments.... We soon pulled off the creepers, and with rough brushes, made by tying together the midribs of the leaflets of the corosa palm, we set to work to clear away the coating of moss.

As the curious outlines of the carved ornament gathered shape it began to dawn upon me how much more important were these monuments, upon which I had stumbled almost by chance, than any account I had heard of them had led me to expect. This day's work induced me to take a permanent interest in Central-American Archaeology, and a journey which was undertaken merely to escape the rigours of an English winter has been followed by seven expeditions from England for the purpose of further exploration and archaeological research.

Alfred P. Maudslay (1889-1902,vol.5:i), recalling his first days at Quirigua

When the first European and American explorers penetrated the dense jungles surrounding Quirigua more than 150 years ago, the ruins of this ancient Maya ceremonial center fired the Romantic imagination in search of "lost" civilizations. To the pioneer archaeologist of the ancient Maya, Alfred P. Maudslay, the extraordinary carved monuments at Quirigua were an important inspiration. Today we remain impressed by the grandeur and artistic excellence of Quirigua's sculptures, many of which are justifiably considered masterpieces of Maya art. Carved with stone tools, the sandstone monoliths are varied in form and proportion, from short and squat to extremely tall and slender. Many of the sculptures feature idealized portraits of kings dressed in lavish ceremonial regalia. The hieroglyphic texts that accompany these figures reveal that they were erected in honor of local rulers near the end of the Classic period of Maya civilization (A.D. 250-900).

Now thoroughly excavated and converted into an archaeological park, the monuments of Quirigua stand where they were originally erected, in low-lying plazas adjacent to a palace compound that served as the residence of its rulers about A.D. 450-850. (The names of the Quirigua kings are listed in Appendix A.) Unlike other centers, such as Tikal, Caracol, or Calakmul, Quirigua was never a large urban complex but rather served as the ceremonial and market center for a dispersed rural population, in which ethnic Maya were a minority. Quirigua was established on the north bank of the Motagua, a river originating in the highlands of western Guatemala near the ancient trading center of Chichicastenango. Winding its way between the Chuacús range, which lies to the north, and the great line of volcanoes which loom over the Pacific coast to the south, the river gradually drops into the Motagua valley, one of the prominent geological features of Central America (Fig. 1.1). Bordered by the Sierra de las Minas and Montañas del Mico to the north and the Sierra del Merendón and Sierra del Espíritu Santo on the south, the broad valley guides the river's meandering course through hot, moist bottomlands toward the northeast and the Gulf of Honduras. Today the Motagua valley is still the primary artery for travel between the western highlands of Guatemala and the Gulf of Honduras.

The geographical location of Quirigua was undoubtedly selected not only because of the access to the highlands but also because it marked a point where the river crossed the route between the city of Copan and the major centers of the Peten. Heading almost directly north from Copan, the mountain trails passed the Copan satellite Río Amarillo and then connected with the headwaters of the Jubuco and Morjá Rivers, which empty into the Motagua just southeast of Quirigua. Travelers to the Peten could then continue northward from Quirigua over a low pass which placed them on the banks of Lake Izabal. Prehispanic settlements have been documented along this route and at its terminus on Lake Izabal, at modern Mariscos (La Ruta Maya Conservation Foundation n.d.; Orozco et al. n.d.). From there, they could follow the tides of the lake via the Río Dulce to the Gulf of Honduras, which provided access to the numerous sites situated on the coastal rivers of southern Belize, such as Pusilha, Uxbenka, Lubaantun, and Nim Li Punit (Fig. P.1). Alternatively, disembarking at the northeastern end of the lake, they could begin overland treks into the southeastern Peten.

The location of Quirigua at a crossroads between the highlands, the southeastern Maya zone centered on Copan, and the Peten heartland suggests the importance of trade in its economy (Ashmore 1884; Sharer 1878, 1990; Sharer et al. 1983:48; Sheets 1983). Although excavations suggest that Quirigua was unusually poor in jade compared to other Maya centers, there is archaeological evidence for the city's trade in obsidian, derived primarily from the Ixtepeque source located near the upper Motagua. In addition, the highly fertile bottomlands of the valley no doubt supported agriculture, and there is some evidence for cacao as a local cash crop in the Classic period (Ashmore 1984). The vast forest resources of the lower Motagua valley also probably contributed significantly to the local economy. Despite all these advantages in location and natural resources, however, Quirigua grew slowly and even collapsed for a time, before achieving a period of growth in the eighth century A.D. At its height, Quirigua consisted of a settlement center of only about four square kilometers with a population of no more than two thousand persons (Ashmore 1980a:23, 1987:221). Even including the many outlying groups that surrounded the floodplain center in the eighth century, Quirigua was very small, especially compared to its neighbor, Copan, where fifteen to twenty thousand persons occupied a small mountain valley during the Late Classic period (Fash 2001; Webster, Sanders, and van Rossum 1992).

Maya Kingship

While Copan appears to have been settled far earlier than Quirigua and grew much larger, kings ruled both cities during the Classic period. (Lists of events at the two centers appear in Appendices B and C.) Like the kings of many other centers, the Late Classic rulers of Quirigua were considered both political and spiritual leaders. One of the royal roles emphasized in hieroglyphic texts and monumental art is that of a medium between the social and supernatural worlds. Rulers could serve as mediums for supernatural entities during ecstatic ritual (Freidel and Schele 1988a; Freidel, Schele, and Parker 1993; Schele and Freidel 1990). Conjured using ritual implements, represented in figural art, or embodied in sacred masks and costume, deities were manifested in diverse forms so that the kings could communicate with and direct them. Such acts of supernatural communication were closely connected with the sacrifice of blood and other precious substances. Astronomy constituted an important aspect of supernatural contact, for through this knowledge rulers were able to anticipate auspicious moments for activities such as warfare or political ceremonies. Astronomy, numerology, and other sacred knowledge became the basis for the chronology of official histories, as recorded in hieroglyphic codices, painted ceramics, and inscribed monuments. Such knowledge had to be publicly affirmed through performance, however. In this sense, Classic kingship emphasized power through personal charisma.

For about a millennium, beginning around A.D. 100, rulers of ancient Maya sites generally conceived of the transfer of power as dynastic or carried through lineage that was reckoned to a deified ancestor. Rulership was patrilineal and often determined by primogeniture, although occasionally it could pass through brothers. Following the death of the previous ruler, a lord underwent a series of complex accession rituals that associated the ruler with certain distinctive supernatural entities. Their culmination was a ritual death and rebirth, signaled by coronation with a white headband (sak hunal) made of bark paper, which might include jade ornaments that revealed their living spiritual essence. Additional personified headdresses were sometimes presented, and the ruler displayed a snake-footed deity (God K) scepter, called k'awil. As a sign of his new identity, the ruler also assumed a new name, usually derived from a (typically celestial) deity. Frequently, this name was identical to that of a prominent ancestor; and in a real sense the king became the present manifestation of that former personality.

An ancient Maya king was entitled to a certain political status, embodied in the emblem glyph title that he usually bore. The emblem glyph is a title naming a person a supreme ajaw of a certain polity, ideally, of equal status with other emblem glyph-bearing rulers. For example, the Quirigua emblem glyph (Fig. 1.2) consists of a dotted element reading k'uhul "divine," prefixed to a sign depicting a gourd, which was the ancient name for the site. The small sign above the gourd reads ajaw "lord." In general, the polity referenced by the emblem glyph signified a city and probably a certain amount of the surrounding land. In many cases, small sites were established at strategic locations within larger polities, such as El Cayo, built on an island near Piedras Negras. The rulers of some subordinate centers were merely called ajaw or had specialized titles such as sajal instead of the full emblem glyph. Many of these sublords acknowledged the dominance of their overlord in the texts they commissioned. In some rare instances, lords of subordinate centers used the same emblem glyph title as their overlords. An example is B'alam Ajaw of Tortuguero, who was a war leader under K'inich Janab' Pakal I of Palenque during the seventh century. Many of these political hierarchies were expressed through complex references to "overkingship" in hieroglyphic texts. Thus, some lords are stated to be yajaw "the ajaw of" another. Others conducted actions that are said to have taken place ukab'jiy (or uchab'jiy) "under the supervision of" an overlord. Political expansion, therefore, was not defined in terms of territorial acquisition per se but by subordination of rulers and their dynastic centers.

Of further importance in maintaining the hierarchy of different polities was the intense rivalry between Tikal and Calakmul, the largest urban settlements in the Classic Maya lowlands. Recent evidence suggests that the economic success and growth of many Classic-period polities were closely tied to a site's political relationship with these two great powers. Although the precise mechanism of these interactions is still being investigated, intersite marriages, elite visits, presentation of gifts, and military intervention have all been suggested as factors. Most of the larger Classic centers had political relationships with either Tikal or Calakmul that sometimes extended over a long period. Accordingly, intersite relationships often developed into enduring rivalries and alliances. Occasionally, however, sites profited through a change of alliance coupled with military victory. The most famous example of this strategy is probably Caracol: beginning as a client of Tikal, Caracol switched sides in A. D. 562 and, aided by Calakmul, witnessed the defeat of its former overlord (Grube 1994; Martin and Grube 2000). Although Tikal and Calakmul did attack each other and each other's allies directly, sometimes an ally of Tikal would attack an ally of Calakmul or vice versa. As will be seen, neither Quirigua nor Copan was isolated from the tension between Tikal and Calakmul. In fact, Quirigua's explosive growth in the eighth century may be explained by reference to these external political relationships, apparently affording its most famous ruler a new route to power through warfare and sacrifice rather than dynastic inheritance.

The focus of this book is the history of this ruler, who led Quirigua into its period of maximum political power during the eighth century, reigning from A. D. 725 to 785. According to the hieroglyphic inscriptions, his name was K'ak' Tiliw Chan Yo'at/Yo'pat, or K'ak' Tiliw for short (Fig. 1.3a). Like many elite names of the Classic period, this name derives from that of a deity, thereby evoking both his superhuman power and divine ancestry. The first part of the name includes the words for "fire" (k'ak') and tiliw, which is probably a derived form of the root til, meaning "burn," followed by chan "sky." The last element of this name, yo'at/yo'pat, alternates with a glyph that depicts a lightning deity who holds a lobed stone object, often in a quatrefoil shape (Fig. 1:3b). This object symbolizes the caves in which the Classic Maya considered many deities, especially the lightning spirits, to reside. It is also utilized by the Yo'at/Yo'pat lightning spirit to crack the carapace of the cosmic turtle, resulting in the rebirth of maize, as discussed below (Fig. 1.4). The approximate translation of this ruler's name as "fire-burning celestial lightning god" is truly awesome, representing a significant claim to divine identity.

By 725, when this ruler assumed the title of divine lord of Quirigua, many of the sites in the Maya lowlands were experiencing growth and concomitant political tensions. The neighboring site of Copan in particular was undergoing a population explosion that had begun to stress the valley's carrying capacity. Its ruler, Waxaklajun Ub'ah K'awil (formerly known to scholars as "18-Rabbit"), witnessed the expansion of Copan during the reign of his predecessor and probable father, Smoke Imix, who had reigned for most of the seventh century, from A.D. 628 to 695. Even so, the end came sooner than Waxaklajun Ub'ah K'awil could have anticipated, when he was captured and sacrificed under the auspices of K'ak' Tiliw in 738.

In this regard Copan was not alone, for this was a time of ruthless conflict and power struggles among elite centers, many of which witnessed the humiliation of defeat in war and the capture of their rulers. Calakmul, for example, suffered the loss of its ruler, Jaguar Paw, in 695. In 711 Palenque also lost its king, K'an Xul, to its enemy Tonina. The victors in these struggles often commissioned major art programs. Tikal, for instance, was enjoying a renaissance under Jasaw Chan K'awil; and master artists at Yaxchilan, under the auspices of Shield Jaguar, were working on Temple 23 and its great lintels featuring his wife, Lady Xok. The ruler of Naranjo, K'ak' Tiliw Chan Chaak, had just completed a series of successful raids in the Yaxha region and commissioned a number of exquisite stelae to commemorate himself and his redoubtable mother, Lady Six Sky. One of the most astonishing success stories of the times, however, was that of Dos Pilas, a renegade dynasty that split from Tikal in the mid-seventh century. Led by a series of aggressive rulers who had allied themselves with Calakmul, this polity expanded rapidly, conquering several sites in the region. Ruler 2 of Dos Pilas, who acceded in 698, oversaw the translation of his polity's new wealth and status into massive architectural programs, such as the El Duende group. Quirigua's political strategies bear comparison to those of Dos Pilas in some respects. It seems likely that those in power remained well informed concerning developments in polities both near and far and adjusted policy accordingly, waiting for the perfect moment to strike at those in their path.

What is particularly significant about the history of K'ak' Tiliw is the singular role of monumental texts and images in celebrating the ruler's exploits, by presenting these acts in certain supernatural contexts. During his long reign, Quirigua was embellished with eight known stelae, one large zoomorph, and two smaller zoomorphic sculptures. The monuments are of intrinsic significance to archaeology and art history for their massive scale, elaborate carving, and excellent state of preservation (Fig. 1.5). In view of their colossal size, their high sculptural quality, and the eloquent poetics of their hieroglyphic texts, the sculptures of Quirigua stand out as some of the greatest achievements of Classic Maya civilization. They are also nearly all in situ, which locks them into a precise spatial and temporal context. But even more important is the survival of the Quirigua monuments as a complete series between the dates of A.D. 746 and 810, spanning the reigns of at least three kings. Few Maya sites provide such a comprehensive record of artistic development over time. In spite of these qualities, previous studies have not adequately contextualized the art or politics of Quirigua within the greater Maya or Mesoamerican traditions. In Maya studies, Quirigua is usually considered of secondary importance, owing to its marginal location and relatively unassuming architecture. This study highlights the importance of the sculptures of Quirigua as a major source of information concerning ancient Maya spirituality and political theory that can be related to a specific historical context.

Sculptural Formats and Practices at Quirigua

Artistic traditions clearly express the political and spiritual ties between Quirigua and other Classic Maya centers. These practices drew indirectly upon traditions that had been developed by one of the most ancient civilizations of Mesoamerica, the Olmec of the Gulf Coast of Mexico (de la Fuente 1973; Drucker 1952; Milbrath 1979). One of the major centers associated with this culture, La Venta, flourished between 1000 and 600 B.C. Sculptural technique at La Venta was varied, with execution in both high and low relief. Among the Olmec innovations seen at La Venta were some of Mesoamerica's first upright stone monoliths or stelae as well as rectangular thrones and volumetric sculptures in the forms of humans, colossal human heads, animals, and supernatural beings. The Olmec also sometimes associated altars with stelae, as at the highland site of Chalcatzingo (Grove 1984: 62-64). The stela form may have evolved from the Olmec celt or ceremonial axe, which was identified with maize (Porter 1996; Taube 1996). This symbolism is expressed in a set of celtiform stelae set up at the foot of La Venta Mound C (Fig. 1.6). These monuments depict supernatural beings wearing elaborate headdresses crowned with a trefoil maize icon. Together with the upright form of the stela, such botanical imagery has led some researchers to associate these monuments with concept of a "world tree," a symbolic axis of communication between levels of the universe (Freidel, Schele, and Parker 1993: 134-135 Reilly 1994). But the Olmec stela was not exclusively a supernatural effigy; it could also portray historical personages. The La Venta stelae sometimes show rulers in a narrative mode portraying ritual action (e.g., Stela 5), but these rulers can also be represented iconically, bearing the implements of office and/or placed in cosmological or supernatural settings (e.g., Stelae 1 and 2; Fig. 1.7).

As the Olmec culture at La Venta waned, numerous centers elsewhere in Mesoamerica preserved and elaborated these sculptural forms, including thrones or supports as well as stelae. Upright carved slabs appeared for the first time in the Maya lowlands in the Middle to Late Formative period, at sites such as Nakbe and El Mirador (Hansen 1989; Matheny 1987). Although these monuments were the direct ancestors of Classic stelae, also participating in the development of the stela were centers in Chiapas, the Guatemalan highlands, and the Pacific slope, particularly Izapa, Abaj Takalik, El Badl, and Kaminaljuyu, all of which thrived in the Late Formative period (300 B.C.-A.D. 25o). At these centers, the stela format was exploited even more than it had been among the Olmec. At most of these centers and especially at Izapa, stelae were placed at the base of mounds in a manner reminiscent of La Venta.

Although each of these major Late Formative centers featured stelae bearing varied iconography, one image is common to all four centers: the ruler shown in the ritual of conjuring spirit beings, who appear above him, as on Izapa Stela 4, Kaminaljuyu Stela 11, and El Baúl Stela 1 (Fig. 1.8). When they adopted the stela form in the second and third centuries, lowland Maya rulers preferred this type of scene, the antecedents of which can be traced to Middle Formative Olmec stelae such as La Venta Stela 2. Although iconographic and epigraphic similarities suggest that the early lowland Maya stela was more closely related iconographically and stylistically to the sculptures of El Baúl and Kaminaljuyu than to those of Izapa, the importance of Izapa in promulgating the stela form should not be discounted. As the stela spread throughout the Maya lowlands in the Early Classic period, it retained a number of its Late Formative features. It passed from kingdom to kingdom as a unified conception, replicating the low-relief style and primary function as an expression of the political and religious institution of kingship.

As a defining feature of Classic Maya civilization, the stela has been subjected to intensive study; and several interpretations have been put forth to explain the symbolism of this class of monuments. One of the most important of these is the suggestion that stelae may symbolize the "world tree." According to Mircea Eliade (1964:120,194, 269-274), this concept refers to a cosmic tree located at the center of the world that serves to connect the three cosmic realms of the heavens, earth, and underworld and is a source of life. Part of the original support for the association of stelae with the world tree was an erroneous decipherment of the glyph for "stela" as te' tun or "stone tree." We now know that the Maya termed these monuments lakam tun, possibly translated as "huge stone" or "banner stone." Nevertheless, there is ample support for identifying "world trees" both in the Maya ethnographic record and in ancient Maya art. In fact, most of K'ak' Tiliw's portraits show him wearing the "God-C" apron, which Linda Schele and Mary Ellen Miller (1986: 77) convincingly identify as a representation of the trunk and branches of a sacred tree (Fig. 1.9).

This costume element appears in diverse contexts (such as figurines and carved panels), however, and is not specific to stelae; thus it cannot be taken as proof that the stela itself symbolizes a tree, like the apron. There is in fact no costume element or other icon that specifically marks stelae as symbolic trees. As an alternative to this generic symbolic equation, it seems more productive to look for specific evidence on how the Maya conceived of individual monuments or programs and thereby gain a sense of the complex history of religious meanings conveyed by the monuments. In the context of such an analysis, it is not only the similarities but also the differences between monumental symbolisms that are significant.

While the general status of Classic Maya stelae as arboreal effigies is open to question, there is ample evidence to associate them (in addition to zoomorphs, altars, and other types of monuments) with rituals of cosmic renewal (see Christie 1995; Newsome 2001). Stone monuments were incorporated into elaborate cosmological rituals that established the shape and quality of both time and space. To the ancient as well as the contemporary Maya, time does not unfold in an entirely linear sequence but rather as a perpetual cycle of repeating events, initiated by cosmic reordering or Creation. Monuments and architectural programs reproduced aspects of this cosmic order through their conformation to sacred prototypes and their dedication according to the precise schedule dictated by a complex calendrical system.

The connection between monuments and cosmogenesis was articulated through the use of the Long Count calendar, a system for recording time that emerged during the Late Formative period and later spread through the southern Maya lowlands, appearing first at Tikal in A.D. 292. The Long Count calendar explicitly referenced Creation mythology, as it was used in hieroglyphic texts to count the number of days elapsed since the date of Creation, which was August 13, 3114 B.C., according to the Classic-period sources. In fact, Long Count records on stelae are featured information, usually occurring first in the text and sometimes even written larger than other glyphs.

As exemplified by the west text of Quirigua Stela C (Fig. 1.10), the Long Count begins with an oversized initial series introductory glyph (ISIG) which may read tzik hab' "count of years," into which is infixed a glyph or "patron" associated with the appropriate month in the 365-day hab' or "vague year." Following the ISIG are five units of time, each with a numerical coefficient. The highest unit, which scholars designate the b'aktun (144,000 days or about 400 solar years), is followed by the k'atun (7,200 days or about 20 years), then the tun (360 days), winal (20 days), and finally fin (single day). On Stela C west, the date is written with the numeral nine (a bar representing five units and four dots representing single units) in the b'aktun position. A single dot (framed by two space-filling curls) precedes the k'atun glyph, while glyphs for "zero" accompany each of the smaller temporal units. Combining the units with their coefficients, this date can be calculated in the following manner: [9 x 144,000] + [1 x 7200] + [0 x 360] + [0 x 20] + [0] days after the beginning of the current cycle. Traditionally, scholars represent the date on Stela C west in an abbreviated form, listing the coefficients only, in descending or der and separated by periods: 9.1.0.0.0. In our calendar, this date corresponds to August 27, 455. On this date, Stela C records that an early king of Quirigua set up a stela. In fact, stelae were usually erected to commemorate such whole k'atun endings. Often, however, monuments were also dedicated on quarter-k'atuns, which Mayanists term hotuns. At Quirigua, for example, the known stelae of K'ak' Tiliw were set up on 9.15.15.0.0, 9.16.0.0.0, 9.16.5.0.0, and so on. Mayanists refer to such anniversaries of the Creation as "period endings" (see Thompson 1950:181).

Carved on the opposite (east) face of Quirigua Stela C is an inscription that clarifies the connection between the monument dedication and the events of Creation (Fig. 1.11). This text is one of the most detailed accounts of these events that survives from the Classic period, containing many unique elements. It begins with a Long Count record of the "zero" date of Creation, rendered as 13.0.0.0.0. Following this are the corresponding positions in the tzolk'in or 260-day calendar, 4 Ajaw, and the hab', 8 Kumk'u. Together, these notations are referred to as the Calendar Round. Several events are associated with this date, including a list of sacred platforms or thrones set up by supernatural beings. The first of these objects is dedicated by two deities known as the "Paddlers," aged beings who in ceramic scenes are often shown paddling a canoe. This stone is set up at a place called nah ho' chap "First Five Sky" and is identified as a "jaguar platform/throne stone." The second stone dedication is performed by an unknown deity at a location that may read lakam kah "Large Town." The second stone is referred to as a "snake platform/throne stone." Finally, the third stone is bundled by Itzamnah, a prominent patron of rulership. The stone set by Itzamnah is stated to be a "water platform/throne stone," and its place of dedication is "??-Sky, First Three-Stone place." The entire process is overseen by an entity called "Six Sky ajaw," which David Freidel, Linda Schele, and Joy Parker (1993: 73-74) identify as the "Maize God," but for which I offer a different interpretation (see Chapter 5). The narrative of Stela C is a metaphor for monument dedication by the ruler. His rituals reenact the ordering of the cosmos and compare him to the supernatural beings associated with each of the three stone platforms or thrones.

As discussed in subsequent chapters, specific details of this text were elaborated in order to emphasize the meaning of certain monumental art programs at Quirigua. In particular, boulder sculptures in the form of composite animals were conceived as effigies of these thrones or platforms of Creation (Fig. 1.12; Looper 1995b, 2002b). For example, Zoomorph G is named with a logograph (T150) which depicts a bundle of bones (Fig. 1.13a). Elsewhere in Maya art, the bone bundle is employed as a throne for supernatural beings (Fig. 1.13b) or a support for sacred objects (Fig. 1.13c). A polychrome vase shows a spirit seated on the T150 glyph, which is placed atop a round personified stone that is similar to the zoomorphs of Quirigua (Fig. 1.13d). There are unfortunately no archaeological data from Quirigua to prove exactly how these monuments were used in ceremony. What is clear is that the unusual elaboration of zoomorphic sculpture at Quirigua was related to a local interpretation of the lore of cosmogenesis.

It is noteworthy that while the Quirigua account is extremely detailed, parts of its content are consistent with texts from other Maya sites. For example, both the Tablet of the Cross at Palenque and Piedras Negras Altar 1 mention the events of Creation at the First Three-Stone place, which is named in the same manner as on Quirigua Stela C (Fig. 1.14a, b). The Creation text of the badly damaged Dos Pilas Panel 18 also mentions the First Three-Stone place (Fig. 1.14c). Usually, local elite traditions embroidered the narrative of cosmogenesis by incorporating dynastic ancestors as observers of the events. The key motif of the erection of sacred stones, however, was a widely accepted component of Classic-period lore. Its codification in the inscriptional record may have been historically linked to the spread of period-ending ceremonies involving stelae and other monuments.

The setting of primordial stones was both a principal structuring concept for space and time and a metaphor for social order. As promulgated by the Classic kings, the lore of Creation took on a decidedly elitist tone, implying that the paradigms established by the gods were the inheritance of rulers. As such, cosmogenesis became a royal prerogative that was periodically enacted through ceremonial performance. Through various techniques, rulers drew upon the aesthetic and symbolic significance of popular technologies, such as domestic architecture and agriculture, transforming them into statements of dynastic legitimacy and esoteric power. In the Classic period, the stela gained widespread popularity due to its suitability as a vehicle for political expression. A king's ritual action of stela erection replicated the actions of the creator gods. Further, the workings of the Maya calendar placed each period ending on a day with the same name as the king's political office, Ajaw. Thus, when a king commissioned a stela in his own image, his identity became conflated with the cycle of 360 days. In this way, the religious significance of the anniversary of Creation was appropriated. The stela allowed the king to be linked to the most fundamental definitions of space and time, thereby asserting his supernatural nature.

Beyond its inherent symbolic value, the stela had other ritual functions as a supernatural interface. Since their conception, stelae had been physically associated with mounds and pyramids. The universal Mesoamerican conception of mounds as effigy mountains and of mountains as the abode of spirits and ancestors suggests a function of stelae as portals to the supernatural world. As Evon Vogt (1970:14-16) notes, the function of the modern cross shrines of the Tzotzil Maya of Zinacantan as supernatural "doorways" may be close to that of Classic stelae. There are numerous parallels between the uses of such crosses and ancient stelae, including the practice of "dressing" the object. Postconquest Maya crosses are adorned with flowers and vegetation as well as actual clothing, not only to make the object ready for ritual but also in recognition of the nature of the cross as a living being (Bricker 1981:102-109; Vogt 1970: 14-16). Similar wrapping or binding ceremonies were central to the use of stelae in the Classic period, recorded prominently in the inscriptions (Stuart 1996). In one of the rare depictions of a Classic stela, the monument is shown wrapped with a cloth sash (Fig. 1.15). In the New Year's pages of the Postclassic Dresden Codex (pp. 26d-28d), upright wooden posts are also adorned with capes and sashes (Fig. 1.16). It has been argued that these posts are analogous to Classic stelae (Grube and Schele 1988; Schele and Stuart 1985). The dressing or wrapping of these posts suggests that they, and perhaps Classic stelae as well, were considered to have been vessels for living spirits.

The Dresden Codex images and Classic vase scene noted above also suggest that stelae served as loci for sacrifice. While the codical image shows an offering plate and incense burner placed before the wooden post, the vase depicts a flat stone in front of the stela, upon which is shown a sacrificed child. This image relates to the scenes of bound captives that adorn many actual altars, such as Tikal Altar 8 (Fig. 1.17). Here the carved image preserves the sacrificial offering. The Dresden Codex scenes show blood offerings before the post, a ritual implied by the form of actual altars such as that of Copan Stela 4, which has a shallow depression on its upper surface and drainage channels. In fact, many altars are carved in the image of the quatrefoil portal to the underworld, implying the specialized function of the altar as the point at which energies of sacrifice are magically transferred to the spiritual beings that wait behind or alight upon a stela, such as the jaguar shown on the vase in Figure 1.15.

Hieroglyphic texts also contain references to sacrifices performed upon or in front of stelae in the context of their dedication. The text of Quirigua Stela F (Fig. 1.18a) records the commonest of these events, a "scattering," which in this case is performed on the monument itself. Here, as elsewhere, the substance scattered is ch'ah "drops (of incense)" (Love 1987). A common Classic title, ch'ahom(a) (Fig. 1.18b), refers to the king as "one who offers drops (of incense)." The interpretation of ch'ah as "incense" is convincing, as a ch'ahom(a) glyph from Copan depicts a figure depositing a glyph which reads pom "copal incense" into a censer (Fig. 1.18c; W Fash, in Schele 1989c). Nevertheless, it is likely that blood and other precious substances were burned along with the incense, providing a rich feast for the spirits. The scattering ceremony may relate to planting practices of Maya farmers, in which liquid offerings are poured into the ground. In this sense, royal ritual structurally reproduced popular practices, establishing connections with common people but at the same time veiling rulers in an aura of awesome spiritual power.

Several stela scenes which include burning incensarios, such as Nim 1.i Punit Stela 15 (Fig. Lig), demonstrate that the burning of these offerings was essential to the proper ritual use of the stela. This image shows the scattering ritual in progress, in which standing figures cast drops toward an incense burner placed on the ground. The burning of offerings before a Classic stela strongly recalls the rituals carried out before adorned crosses of modern Zinacantan, in which the cross is readied for supernatural communication by the burning of incense and candles, the "souls" of which provide nourishment for the supernatural beings assembled behind the cruciform "doorway" (Vogt 1970: 14-16).

Even though some sculptors signed their works, there is little additional information about the profession from the Classic period (Montgomery 1995; Stuart 1989b, 1989c). An unprovenanced panel in the museum at Emiliano Zapata in Chiapas, Mexico, displays what appears at first glance to be the sole Classic-period image ofa carver at work (Fig. 1.20). He is shown seated with crossed legs, touching a zoomorphic carved stone with what appears to be a bone stylus. The text above the sculptor's hand contains the "lu-bat" compound which introduces sculptors' signatures, thus identifying the nature of the event depicted (Stuart 1989b, 1990).

The depiction of a bone stylus on this panel, however, suggests that this is no scene of actual carving, as bone would be suitable for carving only the softest stone. For the sophisticated relief sculpture of sites such as Copan and Quirigua, the sculptor would probably have begun by pecking out the rough form with rough tools of flint, followed by work with a wooden mallet and small chisels of varying sizes made of flint or quartz. Drills were employed as well, and much of the undercutting seen at Copan was probably begun by drilling at an angle to the surface of the block. Glyphic portions of monuments were first roughed out into blocks, as demonstrated by several examples of unfinished texts at Dos Pilas, and then finished as the rest of the monument (Schele and Miller 1986: 39). Rubbing with an abrasive stone such as sandstone would have provided the smooth finish desired for most sculptures and was probably a technique used to sculpt sandstone at Quirigua. The Madrid Codex shows gods carving deity heads or masks using the axe, awl, and drill; however, the heads being fashioned in these scenes are probably made of materials other than stone (Fig. 1.21). At Quirigua, K'ak' Tiliw's sculptors employed primarily sandstone, which--when freshly quarried and moistened--would likely have yielded fairly easily to stone tools and is amenable to either deep or shallow relief.

As a final step, most Maya monuments were probably painted. While evidence for polychrome painting exists at some sites such as Yaxchilan and Piedras Negras, the Quirigua monuments preserve only traces of red pigment (for example, on Zoomorphs B and P). It is possible that the Quirigua sculptures were uniformly coated with red paint, a color symbolic of powers of birth, sacrifice, and cosmic renewal. There is no evidence for a naturalistic use of color at the site, nor for the use of color to differentiate sculptural details.

In general, our knowledge of the details of the sculpting process in the Classic period is limited. Nevertheless, Diego de Landa's account of the carving of deity images in wood among the sixteenth-century Yukatek suggests that the activity was accompanied by penitential rituals (Tozzer 1941: 159-161; see also Tate 1992: 30-31, 2001a, 2001b). He observed that when new images of gods were desired, the (male) artisans were shut inside a specially constructed hut and performed their work accompanied by periodic incense-burning and bloodletting. In sixteenth-century Tzotzil, the association between sculpting and bloodletting may be suggested by the term 'an, which means both "to carve" and "to let blood" (Laughlin 1988: 136). Even in the twentieth century, Ch'orti' Maya sculptors who make sacred crosses practice sexual abstinence, fasting, and work in isolation in the forest, in order to remain "in constant spiritual communication with God" (Girard 1995: 279-28o). Such a relationship between the roles of artist and penitent may also have been extant in the Classic period, appearing in the context of the lordly office of itz'at "artist, sage, wise man." The supernatural prototypes of the itz'at are the deity pair known as the Paddlers, who are called chan itz'at "sky artists," possibly in reference to their role as the primordial artists who painted the sky (Barbara MacLeod, cited in Schele 1992b: 257-259). The relationship between the Paddlers and bloodletting is clear from numerous images and texts (Stuart 1984). In addition, a noble bearing the itz'at title is shown in charge of the bloodletting ritual depicted on Dos Pilas Panel 19 (Houston 1993: Fig. 4-19)

Even though the letting of blood during sculpting mentioned by Landa has not been conclusively documented in the Classic period, the collectivity of the art production indicated in his report parallels Classic practices. Where the tradition of signing sculptures existed, larger objects such as stelae often bear the signatures of multiple artists, indicating that large commissions were likely collective undertakings. Piedras Negras Stela 12 alone has the signatures of eight different sculptors. Nevertheless, the execution was evidently carefully controlled, so that multiple artists' hands can rarely be securely identified on large monuments, including most of those at Quirigua.

Although a few sculptors' signatures include titles which suggest that they were also painters, Classic Maya elites seem to have placed a higher value on the arts of writing and painting than on sculpture. Not only are there many more images of scribes than of sculptors in Maya art, but writing and painting are often shown as being of divine origin. On a bone from Burial 116 at Tikal, an artist's hand holding the Classic calligraphy brush emerges from the maw to the underworld (Fig. 1.22). Even the supernatural patrons of artists, the Pawatuns, are never represented with the tools of sculpture--only the paint pot and brush of the scribe (see Fash 2001: Fig. 74). Such profound elevation and deification of the scribal arts may explain in part why large-scale Classic Maya sculpture designed for public display is so overwhelmingly graphic in style, as the planar nature of relief technique requires thinking in graphic terms. With the few exceptions of certain periods at Copan, Tonina, and perhaps very late Piedras Negras, Classic Maya sculptors conformed closely to the aesthetics of the graphic arts, usually treating monumental sculpture as little more than enriched paintings and often retaining the hairlike, fluid lines characteristic of the calligraphy brush and stylus. In contrast to these norms, sculptors at Copan often moved beyond the realm of the graphic, sometimes creating truly volumetric ("in-the-round") altars and thrones in the forms of animals and composite creatures.

At Quirigua the earliest stelae are clearly subordinate to architecture, being located on or adjacent to platforms in the typical Classic Maya manner. During the reign of K'ak' Tiliw, however, sculptures achieved an elevated status, becoming nearly independent objects. The vast open space of the Quirigua Great Plaza served as the setting for these monuments, which were arranged according to cosmological patterns (Fig. 1.23). Although it was based on a design that originated in Copan, K'ak' Tiliw's Great Plaza is so immense that the sense of surrounding architecture which is always present at Copan is greatly reduced at Quirigua. Among all Maya sites, it was Quirigua that came closest to severing the traditional association between stela and mound/pyramid, which had endured since the time of the Olmec. When standing near them, K'ak' Tiliw's stelae convey the sense of being completely self-supporting, demanding equal viewing from all four sides. Further enhancing the impact of these monuments is their huge size, which completely dwarfs the audience. Given that the original plaza floor was about a meter below its current level, the viewer's head originally would not have reached the level of the ruler's feet on some of the stelae. Such effects of scale and setting maximize the presence of the monuments and suggest the central importance of stone sculpture in the artistic program.

Several other features of K'ak' Tiliw's sculptures set them apart from general aesthetic trends in the Classic period, but these are shared to an extent with nearby Copan. Frequently evident at the two sites is a sense that artists were highly experimental, working within a milieu that favored technical virtuosity. At Copan the sculptors during the reigns of the twelfth, thirteenth, and fifteenth rulers explored the dramatic and dynamic effects achieved through deeply undercut and broken stone surfaces (Fig. 1.24). At Quirigua the best artists manipulated layered parallel planes in moderately low relief to define shapes and create shadows from the intense sunlight of the Great Plaza (Fig. 1.25). There is also an awareness at both sites of the variety in types of cuts and surface treatments possible in stone sculpture, a broad spectrum of which were used at one point or another in the history of Quirigua and Copan. Such sculptural diversity casts a considerable doubt on the concept of a unified "Quirigua style" or "Copan style," which appears frequently in the literature on Maya art. Although the preference for certain basic sculptural formats at each site may certainly be documented, careful formal comparisons over periods of twenty or even five years at either site reveal the relatively dynamic nature of sculptural traditions in the Maya Southeast. By the eighth century, when the monuments of K'ak' Tiliw were created, the history of forms and techniques utilized at the two nearby sites was rich indeed, immersing the sculptors in a complex artistic culture and resulting in spectacular sculptural achievements.

One purpose of this study is to explore the nature of the transformations within this extraordinary sculptural tradition. In particular, how may we reconstruct the factors which fostered the changes in the style and iconography of eighth-century sculptures at Quirigua during the reign of K'ak' Tiliw? Such a question has been asked of Quirigua sculpture previously, although it has never been fully explored. In nearly all discussions of Quirigua sculpture, the approach has been largely formal, with only recent speculation on the relationship of art to religious and political history. In the earliest of these studies, Herbert Spinden (1913: 175-177) attempted to support an erroneous theory that Quirigua was colonized following the abandonment of Copan by noting many similarities in iconography and representational mode between the two centers. In Spinden's (1913: 175) view, stylistic development proceeded automatically, disconnected from politics, the details of which were unknown at the time: "The course of development of the stelae and altars may be said to begin at Quirigua where it leaves off at Copan." Tatiana Proskouriakoff's (1950: 131) brief but more sensitive discussion of Quirigua sculpture likewise avoids political speculations, focusing exclusively on formal developments. The discovery of the historical identities of the rulers of Quirigua by David Kelley in 1962 had little effect on the study of their monuments, which was largely confined to the identification of the subjects of the portraits (Kubler 1969:15-18; Miller 1983). The only major study of iconography at Quirigua is Andrea Stone's (1983) unpublished dissertation on the zoomorphs, which related their imagery to concepts of cosmology and creation. Generally, scholars have avoided discussing the political dimensions of style and iconography. Clearly, a study of the nature proposed here necessitates the development of a theoretical framework for such art historical interpretations.

Art and Ritual

The problem of the relationship between Maya visual culture and politics is itself dependent on the definition of rulership in this society. In this study, two concepts are utilized to interpret the nature of royal power: ritual and persona. The first of these has received less attention in art history than in anthropology, despite its relevance to the field. While ritual is often understood within the context of "religious performance," to which the "secular ceremony" bears little resemblance, this distinction is not useful for the ancient Maya. A working definition of ritual as a "prescribed system of proceeding" (Blier 1996: 189) is useful, not only because it encompasses a variety of performances but because it recognizes the role of political intervention through the regulation of ritual (see also Rappaport 1999: 24). Far from being irrational, meaningless rote ceremony, as it is sometimes popularly conceptualized, ritual is a fundamental mode through which humans create "reality" and bring order to the world. As Suzanne Preston Blier (1996:189) states: "Rituals...offer through their formality and relative fixity a means of measuring, mastering, and making sense of the world at large."

Recent scholarship has sustained a vigorous debate concerning the social functions of ritual (Bell 1992). According to one school of thought, rituals are essentially a symbolic language through which cultural meanings may be grounded in individual experience (Turner 1967). As one of the proponents of this approach, Sherry Ortner (1978: 8), states, "As actors participate in or employ symbolic constructs, their attitudes and actions become oriented in the directions embodied in the form and content of the construction itself, the construct--the model if you will--makes it difficult for them to 'see' and respond to the situation in a different way."

The limitations of this approach are clear. For one thing, it is based on the ethnocentric assumption of a fundamental opposition between the individual and the collective, in which individual difference is a "problem" that is "solved" through ritual. Further, it overemphasizes the cognitive, propositional aspects of ritual. It does not deal with the fundamental nature of rituals as performances, in which nonpropositional, nonsemantic formal elements play a key role in forging social relations. In fact, numerous studies have explored these aspects of ritual. For example, Bruce Kapferer's (1979b) analysis of an exorcism rite in Sri Lanka demonstrated that changes in the relationships of ritual participants were effected through the manipulation of media, space, and audience/participant focus. Rather than merely providing a passive dramatic backdrop for a communicative act, performance may be understood as a medium in which social relations are transformed (see Geertz 1966: 7). An analysis of ritual must consider not only its semantic content, conveyed through verbal texts, but also the way in which the performance reveals experiential truths through bodily praxis.

While this conclusion contributes to an anthropological theory of ritual, it does not constitute a historical model. In order to understand the history of ritual, we must find ways to connect one performance to another, documenting continuities and changes as they are enacted by specific human agents. In Maya archaeology, significant steps in this direction have already been taken, and it is now argued that ancient Maya political history cannot be separated from ritual. The work of Linda Schele and David Freidel in particular has been dedicated to understanding how numerous aspects of ancient Maya political interaction were articulated within a framework of ritual performances. In two studies these authors argued that the origin of Classic-period culture was marked by an abrupt change in ritual (Freidel and Schele 1988a, 1988b). In their view, this took place in the Late Formative period across the Maya lowlands of the Peten and Belize, when monumental architectural structures bearing images of supernatural beings were built as theatrical stages by an emerging nobility. The conduct of rituals in this context provided a basis for these rulers' claims to supernatural ancestry. Eventually, the deity images of the facades were replaced by portraits of rulers, thereby fixing divine identities in a more permanent form.

These interpretations stand in dramatic contrast to previous reconstructions of ancient Maya culture, especially those promulgated by the eminent scholar J. Eric S. Thompson, who saw political (sometimes called "historical") interpretations in direct opposition to ritual (Thompson 195o: 63-65). In Freidel and Schele's view, public performance and charismatic ritual were crucial to the power of Maya rulers, through which they could sway the loyalties of people who viewed and participated in these ceremonies. Performances that displayed differences in regalia, spatial position, and access to sacred materials and objects maintained hierarchical distinction between nobility and commoners. According to Schele and Freidel, the rituals of the ancient Maya elite were carried out principally in order to effect cosmological changes. This definition of ancient Maya power acknowledges the transformational role of ritual and suggests that power is meaningful not in an abstract symbolic sense but to the extent that it is invoked ritually.

While power among the ancient Maya was exerted in the social world, its principal source was perceived as the normally invisible "otherworld," manifested in the form of various spirit forces which together composed the living cosmos. Perhaps the most potent of these was k'uh, roughly translated as "holiness," which was identified with royal blood. Another distinct spiritual force recognized by the Maya translates roughly as "white flower spirit." This essence was thought to reside in the breath but was also profoundly associated with procreation and particularly with umbilical cords. Interestingly, each of these concepts associates spiritual power with substances that emerge from the interior of the body. Accordingly, a fundamental ritual pattern involved the opening of the body so that its immanent forces could be manifested. For example, through the perforation of the body and drawing of royal blood, the power inherent in this substance was revealed and put to use. Likewise, the sacrifice of a captive's intestines magically manifested the powers of the umbilicus. A less violent context for the deployment of spiritual essences was the formal speech and song of the elites, which released the forces of breath and the particular powers of sex and procreation.

Ancient Maya power, then, could be accessed through ritual procedures that centered on the manipulation of the body. Such a focus may imply that Classic Maya political ritual derives from or was otherwise historically related to traditions of shamanistic curing and midwifery. In fact, glyphic texts that accompany such scenes of deity conjuration occasionally refer to the event as the "birth" of the deity. Another expression used in the context of bloodletting is the same as that which relates a child to its mother. Such metaphors may exemplify the elite appropriation of popular ceremonies that existed in Mesoamerica before the advent of kingship.

Rituals designed to release the power of the otherworld required a sophisticated means for channeling these tremendous forces. Such was the function of artifacts that we designate as "art," such as bloodletters, bowls for sacrifice, altars, ceramic burners, and stelae. Many of these objects served as implements or tools, including the stingray spines and obsidian lancets that were used to puncture the flesh. Ritual objects also contained and stored these energies, much like a battery. The dedication sequence of a stela, in which cloth or rope bindings fixed the energy of sacrifices in the monument, illustrates this well. The creation of a work of art may itself have been conceived as the infusion of matter with spiritual power, while ritual use enhanced that power. The intentional breakage or destruction of a work of art was also an essential part of the life history of the object, as its power was thereby released to be put to some other use. This belief, for example, probably lay behind the deposit of fragments of monuments in the foundations of stelae at Copan. Works of art were used to manipulate space and create a sacred landscape for ritual. Accordingly, three or four objects placed in a triangle or square constituted a magical diagram, creating a liminal space appropriate to ritual. In sum, Maya artworks may be conceptualized as technology of ritual transformation, which extended the potential of the human agent to manage sacred energies inherent in certain materials, idealized geometric forms, and chronological symmetries.

In addition to its functions as ritual implement, container, and tool for spatial modeling, visual art served as a communicative medium, a site of ritual inscription through which performances were documented and committed to collective memory. Many monuments, such as the stelae of Quirigua, were relatively accessible to the general populace and featured images of what were likely public dance performances by rulers (Looper 2001). Others, such as Yaxchilan Lintel 24 (Fig. 1.26), bear texts and images that may have been intended for a more restricted audience consisting of elites and their ancestors. This monument was installed in a temple doorway so that it could only be viewed just inside or outside the doorway, depending on the light. The image features a Yaxchilan ruler holding a burning torch over his kneeling wife, who draws a thorn-studded rope through her punctured tongue. Blood scrolls on her face stand as memorials of this ritual, as do the blood-spattered paper strips in the basket before her. The two main framed texts, located at the top and at the middle left margin of the panel, complement the two figures, labeling them as "images in penance." These inscribed images thus awaken thoughts and feelings in the viewer through their sculptural forms, iconography (images), and spatial relations with architecture and landscape and (in ancient times) through interpretation and display in ritual performance. Because the lintel was installed in a manner that made viewing by living humans difficult, its primary intended audience may have been ancestral. Its specifically penitential subject matter may also suggest this, as such an image might have been considered particularly moving to the ancestors.

A similar propitiatory function is suggested by other lintels and wall panels at Yaxchilan and Copan, which feature texts written in mirror image. While this could be an example of scribal virtuosity, such an arrangement may imply that the texts were meant to be read through the walls or from the sky, that is, by a divine audience. One stela at Piedras Negras even has an inscription on its upper surface, invisible to the earthbound human viewer. Similarly, many beautifully carved monuments, such as the sarcophagus of the king K'inich Janab' Pakal I of Palenque, never saw the light of day, being entombed deep beneath massive architectural structures. These examples serve as reminders that artworks were meant to be viewed and cherished not only by a community of living humans but also by the ancestors. As an activity that was sanctioned by traditional convention, the carving of a text or image had the effect of a magical formula, making the inscribed event happen, regardless of whether it was seen by human eyes.

The making of art was not only the creation of reality through ritual but a fulfillment of the ceremonial obligation of the elite. Texts and representational art actualized the rituals that the elites were required to perform, through their control of sacred materials and knowledge. Further, because ancestors are reborn through their descendants, what the living memorialized through art was in a real sense also remembered by the dead. The value of monumental art, and in turn the spiritual power of the elite, lay in its capacity to incarnate memory and to stimulate reflection and emotion in a diverse and yet interrelated audience. Even public monuments may be considered to be primarily offerings, transactional objects intended to restructure social relations among the living and the dead.

The function of Maya art as offerings or gifts was also crucial in negotiating social status. Trade and exchange networks kept Maya courts supplied with luxurious and exotic materials from which art was made. These materials were worked and combined with local materials in special ways to produce other commodities, such as painted pottery (Reents-Budet 1994). Monumental sculpture, for instance, required not only locally quarried stone but also specialized tools made from rare imported stone. The technical and esoteric knowledge implied by the production of art objects enhanced the prestige of the elite. But such objects and knowledge had no social value if hoarded. Like the potlatch celebration of the native peoples of the Northwest Coast of North America, which must be witnessed in order to generate prestige, ritual knowledge embodied in art had to be selectively shared with others. In this way, rituals and art objects participated in a network of social and exchange relationships that bound people to the ruler (Clark and Blake 1994). But while they forged social ties, art objects also masked social and political inequalities both within and between cohesive communities (Earle 1990).

Large sculpted stones erected in a public space could embody social exchange too, as their creation and manipulation implied a massive investment in resources, even to those who had not actually witnessed the process of moving them. (This aspect of the monuments still inspires awe today.) Nevertheless, the fixity and the massive scale of monuments permanently withhold them from free economic circulation. At Quirigua restriction of access was further implied by the impassive, unapproachable features of the royal portraits that stare over the heads of the viewer, as well as perimeter foundations, which functioned rather like a velvet rope at a museum. The sculptures were thus kept from the general populace, even though they were given as a public offering. As such, monumental sculptures exemplify the paradoxical nature of certain gifts, which--as discussed by Annette Weiner (1992)--are retained as much as given. This is especially true for objects of great sacred or cosmological significance, such as monumental images. Their monumentality served as a means of governing their social circulation, of preventing the separation of the objects and the cosmological meanings they embodied from the persons who commissioned them.

In the same way that the monumental object functions as a paradoxical gift, so the information that it communicated was both given and withheld. The formalized genres of text and image simultaneously disclosed and concealed the knowledge of its (elite) designer(s). The great abundance and complexity of the pictorial images, obscured and revealed through overlapping of sculpted forms, suggest the wealth of esoteric knowledge claimed by the rulers and partially manifested for the uninitiated.

Hieroglyphic inscriptions are also of critical importance in this regard, in their selectivity and even through the practice of literacy itself. As discussed by Stephen Houston, the standardization of Maya writing and its high degree of elaboration imply that it was probably not fully readable by much of the population (Houston 1994; Houston and Taube 2000). This would have been especially relevant at ancient Quirigua, where most of the local populace was non-Maya. Although evidence is slim for the Classic period, written texts were performed through song or other oral presentation during the colonial period (Thompson 1972: 13; Tozzer 1941: 153). If such practices of "recitation literacy" were extant in the Classic period, then the public display of written texts may have emphasized the knowledge that could only be accessed through ritual performances. In such a manner, valuable information was selectively distributed to the people, with the implication that additional wealth lay behind the inscriptions. Monumentality, for the ancient Maya, thus provided the elite "owners" of the monuments the potential for retaining their identity and perpetuating it into the future. In the words of Weiner (1992: 8), such objects "bring a vision of permanence into a social world that is always in the process of change."

Art and Personae

The importance of art in stabilizing identities prompts a consideration of the specific ways in which ancient Maya monumental portraiture embodies the social person. Indeed, one of the remarkable aspects of Classic Maya art and one that distinguishes it from other areas of the Americas is its "personalized" quality. Specific persons are represented, sometimes on more than one monument, and histories relate momentous events in the lives of kings. In addition, the relatively naturalistic proportions of Maya art, together with an abundance of incidental physiognomic detail such as fingernails and strands of hair, convey a strong sense of immediacy and physical presence of the subject. A few Maya representational traditions, such as that of Palenque, encouraged highly naturalistic royal portraiture, with particular emphasis on distinctive facial characteristics (Griffin 1976).

Accordingly, in recent decades some scholars have written Maya art history in terms of human actors, even to the extent of using iconography, architecture, and inscriptions to reconstruct personalities and intentions of rulers (e.g., Jones 1977; Newsome 2001; Schele and Freidel 1990). Such an approach is grounded in Western art historical tendencies, exemplified especially by Ernst Gombrich's (1966: 35-57) attempts to glimpse the personalities of the Medicis through the works they commissioned. Despite this precedent, however, such an approach must proceed with caution, at least in the ancient Maya context (Fash 1998; Houston 1989). For example, it seems unlikely that we would be able to "reconstruct" the personality of a ruler, since the artifacts and the ruler are so tenuously associated. In particular, in ancient Maya art history we have very limited evidence concerning patterns of patronage and the specific relationship of the ruler to the artists.

The monuments of K'ak' Tiliw are particularly deceptive in this regard, as the main subject of both their texts and image is the king. The standard rhetoric of the inscriptions claims for the king sole responsibility for their dedication. Given this unitary "personal" focus, it seems only natural for the historian to look for the impetus for these works in the mind of the king himself. In addition, the themes of many of the images and texts are highly subjective, relating to dreams and trance experience. The small scale of the community and the centralization of art production only strengthen our suspicions that it was the king himselfwho planned the images. Nevertheless, while the king could have designed the monuments or have otherwise assumed some responsibility in their programming, it is essential to remember that we have no evidence to confirm or deny this supposition. It is also possible that the design could be attributed to a master scribe, another member of the royal family, a council of lineage heads, a shaman-priest, or some other religious specialist. In short, there is no documented connection between the king and the inception or execution of the work, making the attribution of intention to a particular person difficult.

On a more abstract level, the search for a basis of art in a single personality ignores the problems associated with the concept of artistic intentionality. While this study takes as given the proposition that humans are endowed with agency, it also recognizes that the attribution of artistic creativity to a reconstructed historical state of mind is problematic. In particular, the voluntary causes attributed to historical individuals may have been implicit in the cultural institutions in which the actors unreflectively took part. Other intentions may have become acquired through a history of behavior which had once been but was no longer consciously contemplated. In this view, artistic intentions are situated in the relationship between the context of artistic production and the object itself (Baxandall 1985: 42). Intentionality is thus one of the deceptions of art. While we assume that a specific group of people must have made the physical work of art, the integration of these people into a broader webs of social interaction inextricably links agency into systems of behavior that are not reducible to the sum of their parts. As stated by Alfred Gell (1999:163), works of art seduce the viewer/interpreter into a "network of intentionalities whereby, although each individual pursues (what each takes to be) his or her own self-interest, they all contrive in the final analysis to serve necessities which cannot be comprehended at the level of the human being, but only at the level of collectivities and their dynamics." The indeterminacy of art lies in the magical ability of the artist to transcend the understanding of the spectator. Indeed, Gell (1999: 172) describes the artist as an "occult technician," whose work "mediates between creative agency and the power of the collectivity."

Such considerations suggest a different approach to studying the subjectivity and personalization of art at Quirigua, not based on personality as a fact or primal source, but rather with the goal of reconstructing the role of the "self" as defined historically by the society (see Mauss 1985: 3). A number of studies by Stephen Houston, David Stuart, and Carolyn Tate have proposed models for interpreting ancient Maya representations in terms of the "self" (Houston and Stuart 1996, 1998; Stuart 1996; Tate 1992: 11-25). Houston and Stuart in particular observed that Maya monuments were not conceived as false simulacra but rather as living entities that shared in the essence of the rulers. The term used to refer to such images, b'ahil, was derived from b'a(h), which meant not only "self" and "person" but also "head." As such, it identifies the head and face as a particularly significant locus of personality-a concept that goes far to explain the compositional focus on the king's face at Quirigua and other sites. Images of a ruler not only embodied the royal self in multiple permanent forms but were considered to function as active agents on the ruler's behalf. In this way, the Maya overtly acknowledged the general function of art and artifacts as secondary agents, capable of propagating causal sequences of events as extensions of the human agents who made them (see Gell 1998:15-17).

Two examples of the attribution of agency to stone sculptures are illustrated in inscriptions from Quirigua. One is on Stela E east (Fig. 1.27a), where the events of the period-ending are introduced with a glyph that assigns agency, followed by ub'ahil "his image." The next glyph block is of astronomical significance, preceding the name of K'ak' Tiliw 29 This passage suggests that the ceremony was conducted under the authority of the royal image, depicted in an astronomical guise. A second reference to an active, living monument may appear in a partially eroded passage on the west face of Stela D (Fig. 1.27b). Here the text records an event associated with the dedication date of the monument as ajawaj, or "it is made ajaw," a phrase related to expressions for royal accession. Next may be the glyph for "his image" (ub'ahil), followed by a series of illegible signs. The final glyph of the clause is tunil "stone object." This passage, then, may suggest that for the period-ending ceremony a stone monument (presumably Stela D itself) was itself made ajaw.

The recognition that stelae function as surrogates for royal authority has further implications for the understanding of their imagery. In particular, it suggests that the monumental portraits may have served to propagate and perpetuate the gaze of the king. As discussed by Stephen Houston and Karl Taube (2000), ancient Maya conceptions of the sense of sight were not the same as the view developed by modern science, in which the eye is a passive receptor of light. In contrast, the ancient Maya eye was an "emanating eye" that actively changed the world by exerting the will of the viewer. Thus, a common title (or nominal component) of kings, k'inich, meaning literally "sun-faced" or "sun-eyed," expresses the searing heat and brilliant light that were believed to emanate from a ruler's face or eyes. Like the sun, the gaze of the ruler was probably credited with the capacity to engender life.

Part of the dedication ofa monument was the witnessing of its sanctification by the ruler or some other noble, which may have animated the representation (Houston and Stuart 1998; Houston and Taube 2000). In particular, this event may have opened the eyes of the carved figure, investing them with the power of sight on the ruler's behalf. Indeed, the aloof gazes of ancient Maya stela portraits are strongly suggestive of the establishment of a wide visual field, with the implication of control of events within that field. For the contemporary Yukatek, an orientation above or movement upward is considered to be relatively powerful or beneficent (Hanks 2000: 26-28). The Classic Maya glyphic expression yichnal also expresses this notion of an encompassing visual field that was crucial in validating ritual (Houston and Taube 2000). Cognate with modern Yukatek yiknal "in front of," this expression is also inherently hierarchical, linking subordinate persons to a ruler or deity. In fact, certain texts from Quirigua use the yichnal in substitution for ukab'jiy "under his supervision" (Looper 1999: Fig. 15). Stelae such as those of Quirigua may have been conceived in part as instruments for extending and perpetuating the dominant gaze of the ruler, but also as a means for invigorating those in the visual field with the royal "heat. "

A group of portrait stelae such as those associated with K'ak' Tiliw, then, may be interpreted as a means of distributing royal agency throughout the landscape, embodied in a series of distinct visages. A study of ancient Maya personhood thus requires an examination of the meanings not of individuality but rather of the "dividuality" of the self--to borrow a term from Marilyn Strathern (1988)--which was achieved through representational art. In this view, there is no a priori category of the self separate from the collaborative practice of its figuration. Further, we must acknowledge that the self is not a static entity but one which changes and evolves over time.

In order to define the changes in self-presentation in Maya monuments and to interpret the social and historical forces that contributed to their making, a term employed in psychology and literary criticism is useful. The term "persona" is used in literature to distinguish between the author and the narrator (Fowler 1987: 176-177) and in psychoanalysis to refer to an "arbitrary segment of collective psyche" (Jung 1953: 105). Both of these conventions make use of the metaphor of the mask or "second self"; and, in fact, the term "persona" originally denoted the masks worn by actors in Greek theater. Although the concept of persona is not often employed by art historians, this etymology suggests the applicability of the term to representational visual culture. In the present study, "persona" is used to define diverse "selves" as they are manifested in art, thereby implying the critical distinction between the identities of the subject of a portrait and the guises presented in a portrait image. Persona, unlike personality, is a culturally constructed mask or a conventional identity that may be changed in relation to dynamic social circumstances.

In ancient Maya culture, personae served as an important mode of mediation between the individual and society and are thus crucial in understanding the dynamics of agency. When manipulated by elites, personae expressed the nature of social hierarchy and inequality. This function of persona is grounded in the widespread Mesoamerican practice whereby rulers legitimated their authority through the display of powerful emblems, often tied to the body. Materials such as jade, shell, quetzal feathers, and bone were attached to the trunk, head, and limbs to infuse these various parts of the body with their power. Notably, many of these materials were derived from loci that were associated with supernatural forces. Bones were literally at the core of the living body, while other materials were acquired from distant, and therefore symbolically powerful, sources. Jade and quetzal feathers came from the mountains, suggesting a celestial identification, whereas shells came from the sea and were thereby associated with the aquatic underworld. Thus, ceremonial attire was cosmological and transformational, magically infusing the elite person with spiritual energy. These details are highly elaborated on Maya monuments, a testament to their iconic power. The principal emblem that served as a seat of spiritual power is a mask and/or headdress. During ceremonial performances, rulers could become one with the spirits that the mask or headdress embodied, thus effecting their physical and psychological transformation. Maya rulers also commonly signaled control over divine forces through the display of deity images in the hands. In this way, masks, headdresses, costume, and other regalia served as a means of forming and manipulating personae, defining the precise relationships between the ruler and diverse supernatural entities.

An examination of the supernatural personae of rulers entails a consideration of the nature of Maya supernatural identities themselves. Did Maya rulers identify with a generalized impersonal supernatural essence or with distinct divine personalities? The answer to this question, as can be imagined, is far from simple, mainly because the subject of the nature of Maya divinities is still debated (Houston and Stuart 1996; Marcus 1978; Proskouriakoff 1965: 470-471, 1978 Thompson 1970: 198; Vail 2000). The Maya spirit world was and remains complex, populated by entities of distinct types. One was known as way, the spiritual co-essence of a person, which usually took the form of a composite animal (Grube and Nahm 1994; Houston and Stuart 1989). In addition, the term k'uh, which was used to refer to an impersonal divine essence, could also reference a specific incarnation in a deity image (Houston and Stuart 1996). Such complexities suggest that to identify Maya supernaturals indiscriminately as "gods" is inappropriate. Among the most significant of the differences between Maya deities and the modern Western conception of gods is the Maya deities' nonexclusive association with fairly broad domains such as agriculture, war, and death. For example, in a statistical study of the Madrid Codex, Gabrielle Vail (2000) demonstrated that nominal glyphs and attributes are used to group diverse deities into three loosely defined, overlapping contexts. The fluidity of roles and attributes of ancient Maya deities helps to explain their tendency toward multiplicity and hybridity.

The spirits of lightning are a good example of an ancient Maya deity complex (Looper 1991a). Across the Maya area, derivatives of the proto-Maya term *kahoq are used to refer to the thunderstorm as either a physical or spiritual phenomenon (Kaufinan and Norman 1984:117; Spero 1987: 231). Thus, in Ch'ol the lightning spirit is called "Chaak" or "Chaak" (Attinasi 1973: 249; Aulie and Aulie 1978: 46), while in Yukatek "Chaak [cháak]" refers to rain or to the deities of thunder, lightning, and rain (Barrera Vásquez 1980: 77; Bricker, Po'ot Yah, and Dzul de Po'ot 1998: 61). In the Classic period, the term "Chaak" is attested as a designation for various supernatural beings, who share a core cluster of features (Fig. 1.28). These include bivalve shell earflares, reptilian eyes, serpent markings on the body, a shell diadem, a knotted pectoral or belt ornament, and a snakelike snout. Many carry hafted axes and trefoil stones and are shown in a jumping movement.

Not only do these beings seem to preside over diverse domains (such as fishing, sacrifice, and caves), however, but most are named with qualifiers, such as Chaak Xib' Chaak, Ux B'olon Chaak, 'O/'Ohl Chaak, and Yax Ha'al Chaak. The distinctions seem to correlate to the ritual domains with which the deities are associated. For example, Yax Ha'al Chaak frequently appears in codex-style pottery scenes together with a particular "death god" and a jaguar deity who has been thrown upon a mountain (see Robicsek and Hales 1981: 39-43). Other lightning deities are similar in appearance to the Chaaks but are referred to in the inscriptions by the term "Yo'at/Yo'pat" and appear in scenes of the resurrection of maize (Fig. 1.4). Despite these differences, the deities can be grouped into a single complex, based on their shared iconography and associations with thunderstorms and the portals between realms of the cosmos. Their particular manifestations depended on specific ritual requirements which, in turn, were grounded in local histories and traditions.

Likewise, the ethnographic record suggests that Maya divinities are not conceived as possessing timeless personalities or singular identities but rather undergo periodic and often seasonal transformation. As facets of a cosmic totality, Maya deities are born and die as they satisfy their roles in the universal biography. They may change names, appearances, attributes, specific domains of influence, age, and even gender. For example, one of the chief Ch'orti' deities is a solar being during the dry season but transforms into a maize spirit upon the arrival of the rainy season (Girard 1995: 350). In addition, Rafael Girard (1995: 278) observes the tendency for contemporary Maya deities to multiply geometrically into compound manifestations. The same phenomenon is well known from the Yucatan, where deities commonly have a quadripartite aspect, being associated with the four cardinal directions (Thompson 1970: 198-199). Such concepts provide a basis for the consideration of monumental images as aspects of royal personae, which change depending upon calendrical, historical, and ritual requirements. Multiple images required the intervention of diverse deities on behalf of the ruler.

While costume and other regalia represent an important dimension of royal personae, naming practices were also significant in communicating the divine attributes of a historical identity. In some instances, these identities converged, when headdresses were used to display deity heads and other elements that correspond to rulers' names (see Martin and Grube 2000: 77). Another approach to merging these identities was the performance of ceremonies appropriate to the domain of one's supernatural namesake. As we shall see, this particular strategy was highly elaborated during the reign of K'ak' Tiliw. Names, however, are not equivalent to personae but are a distinct mode of marking social identity. Distinctions are usually made between rulers and the deities after whom they were named. Personae seem to have more in common with royal titles, which often stress the performance of ceremonial duties or cosmological associations and which are not required to express the identity of the ruler as a historical entity.

The potentially complex relationship of names to personae and other social identities is well illustrated by the example of seasonal ceremonial activity of the Kwakwaka'wakw people of the Northwest Coast (Jonaitis 1991). In the nineteenth century the Kwakwaka'wakw winter season was dominated by the tseka or Winter Dance. This was a season in which the spirit world spilled over into the human world, manifested in the performance of masquerades and the induction of men into initiation societies. During this season, people set aside their secular names and assumed sacred "winter names" and identities based on the nature of their participation in the ceremonies. Initiated persons were classified as Seals, who danced, and Sparrows, former dancers who now managed the performances. Participants were also organized according to secret societies, some of which involved masquerade performance as spiritually potent beings. Rights to these diverse identities were generally acquired through marriage.

How this system functioned in a person's biography is illustrated in a narrative called "The Acquisition of Names," recorded by Franz Boas (1925: 113-357). This story describes how a father prepared his son to succeed him by bestowing successive names upon him, accompanied by the distribution of gifts and observance of appropriate rituals. Manhood is marked by the presentation of a special name which gives the son the right to participate in feasts. Following this, the father, in conjunction with the son's father-in-law, sponsors a Winter Dance, in which the son appears as various characters, including Eater-of-the-Ground, a grizzly, and a fool dancer. After two winters, he retires as a Sparrow. This example shows that the purpose of Kwakwaka'wakw masquerade performance is not to illustrate a name but rather to support the change in social status signaled by the acquisition of the name. Both the name and masquerade participation are dictated by a complex genealogical system that is manipulated to enhance status.

Likewise, when considering ancient Maya identities, it is useful to consider them in the context of personal histories and political strategies. In contrast to the ethnographic case of the Kwakwaka'wakw, the only unambiguous evidence of ancient Maya elite ritual is provided by the ceremonial cities themselves, embellished with representational images and hieroglyphic texts. And while it would be inappropriate to treat Maya artworks as reflections of ritual akin to photographic documents, royal portraiture can be used to identify specific personae and to trace their development over time. Monumental personae may also be expected to vary among different sites. Such comparisons suggest distinctive traditions of personae, which associate rulers with diverse sources of power. Usually, these traditions are grounded in local histories of representation; however, sometimes they can be shown to be borrowed from site to site and even to be manipulated competitively. The patterns of difference and correspondence among personae may be construed as evidence of political discourses, articulated through ritual. Ultimately, variation in personae reveals the manner in which power and authority were articulated and negotiated during the Classic period.

Methodologies

Having established the theoretical foundation of this study, it is useful to outline briefly the methodologies employed in the analysis. In particular, it is essential to discuss the value of each set of data in relation to the questions I have posed. One of the most illuminating of these data sources is the corpus of inscriptions that embellish the monuments of Quirigua. (Complete transcriptions and translations of the Quirigua texts appear in Appendix D.) These texts include declarations of dominance and subordination, warfare and alliance, and other political events that are frequently "disguised" in ritual terms. For example, a military victory is phrased as the throwing down of a war implement. The monumental texts thus provide a rich historical background for interpreting iconography and style. The basic approach to the decipherment of these texts is based on linguistic principles of syntax and phonetic substitution, as has been discussed elsewhere (Schele 1982; Schele and Grube 1994: 1-75; Stuart 1987b). Above the level of syntax, a discourse analysis of a text (or several related texts) allows for identification of major events and actors and of episodes in linked events (Josserand 1991). Such analyses expand the possibilities for reconstruction of political relationships, implicit in the actions of human actors and their supernatural patrons.

Epigraphers and archaeologists have occasionally expressed reservations about the relation of Maya inscriptions to history (e.g., Houston 1993: 9; Mathews 1985: 52-53). It has even been suggested that this is no "true history" but one so inextricably entwined with mythology as to render it useless as an interpretive category. Like any history, however, that inscribed on K'ak' Tiliw's monuments represents a carefully selected and integrated narrative mainly concerning human actors, the "truth" of which is dependent on the point of view of the compiler(s). Further, because any historical reconstruction is a dialogue with the past, but limited in that we must formulate the questions, an interpretation of ancient Maya texts is a political reconstruction. Thus, the goal of this project is to reconstruct the propositions made by the texts of Quirigua and contextualize them through local and regional comparisons.

In addition, it would be a mistake to assume that Maya hieroglyphic texts present a "confusion" between myth and history. On the contrary, ancient Maya texts reveal distinctive genres of history (stories about humans) and mythology (stories about the deified ancestors), identified through contrasting time frames, with mythic events taking place prior to and shortly after the renewal of the cosmos in 3114 B.C. (Marcus 1992: 8). In the inscriptions commissioned during the reign of K'ak' Tiliw, not only is this distinction rigorously observed through the temporal sequencing of narrative, but texts describing historical and mythical events are often relegated to different spatial zones on the monument. As at other Maya sites, mythic narratives provided the sacred charter for the actions of the king of Quirigua. Thus, the distinction between history and myth was not one of truth versus invention as it is in the modern Western worldview. Indeed, mythic narratives were probably seen as inherently factual, being handed down through the generations and written in the movements of the stars and other natural cycles.

The Classic-period conceptualization of art and its production as strongly inspired by supernatural powers lent to artworks an a priori spiritual significance, a presence which demanded respect and awe. Such attraction was significantly reinforced by the emotional affectivity of form, whereby anthropomorphic or therianthropic images evoke sympathetic reactions in the viewer's body. In addition, Maya imagery is replete with symbols of spiritual and physical power, conveyed through an iconography of gesturing human figures in costume. As George Kubler (1969: 48) observed, the long duration of the Classic Maya style and iconography implies the existence of a generally agreed-upon system of symbolic values assigned to images. These values were surely stabilized to a considerable degree by the full integration of the writing system with iconography. Such considerations have supported the application ofa structuralist approach to Maya iconography. According to the structuralist paradigm, elements of dress, ritual objects, place of action, posture, and gesture all had conventional conceptual associations, which could be manipulated and configured with hieroglyphic texts into pointed rhetorical statements.

One area of structuralist research that has achieved particular prominence in recent years is the attempt to reconstruct a narrative of Maya cosmogenesis by combining Classic-period texts and imagery with passages from the Popol Vuh. This colonial-era K'iche' epic relates the story of Creation and the origins of the K'iche' people. The following summary of this narrative is based on the most recent reconstruction by Schele and Mathews (1998: 36-37). The basic plot of this narrative concerns the destruction of the previous, third Creation, which ended on 13.0.0.0.0 (August 13, 3114 B.C.), and the establishment of the present cosmos by a pair of Hero Twins named Junajpu and Xb'alanke' (in Classic times, Jun Ajaw and Yax B'alam) and their father and uncle, twin maize spirits. The story begins in primordial times, when the maize deities (called Jun Junajpu and Wuqub' Junajpu in the Popol Vuh) were playing the ballgame, a Mesoamerican sport in which two teams compete using a large hard rubber ball, scoring points by means of floor markers or rings installed in the court in which the game was played.

The vigorous actions of this game disturbed the lords of Xib'alb'a, the Maya underworld. The Xib'alb'ans, portrayed as spirits of disease and death, summoned the maize deities into their abode, subjected them to a series of tests, and then dismembered and decapitated them, burying their parts in a Xib'alb'an ballcourt. The skull of Jun Junajpu was hung in a tree adjacent to the ballcourt, where it remained until the daughter of a Xib'alb'an lord came up to the tree and spoke with the skull. When the woman held out her hand, the skull spat into her palm, whereupon she became pregnant. Fearing her father's wrath, the woman fled the underworld and eventually gave birth to a pair of boys, known as the Hero Twins. They are called heroes because one of their tasks was to destroy various monstrous beings which dominated the previous Creation, such as a false sun named Wuqub' Kaqix ("Seven Macaw").

These twins, like their father, were also avid ballplayers and were likewise called to stand trial before the lords of Xib'alb'a. Being more clever than their father, however, they survived all of the torments their hosts inflicted on them and eventually tricked the Xib'alb'ans themselves into being sacrificed. This being done, they attempted to resurrect their forbears in the ballcourt. The maize spirits, reborn as infants, grew quickly, like maize plants, into young adults and began to make preparations for the dawning of the new Creation. After being dressed by their sons and certain goddesses, they awakened a series of aged deities, including the Paddler gods and a patron of merchants known as God L, who helped cleanse the world through a great flood. The Paddlers ferried the maize deities to a place of Creation marked by a turtle, where they were resurrected through a cleft in the shell made the Yo'at/Yo'pat manifestation of lightning. On 13.0.0.0.0 the maize spirits directed the gods to set up three cosmic hearth stones. On the same day, celestial cords, probably identified with the umbilicus of the reborn maize deity, descended to earth bearing sustenance. On 13.0.1.9.2 (February 5, 3112 B.C.), 542 days later, the maize deities completed their work on this fourth Creation by establishing the three-dimensional space of the cosmos, conceived as a house. They marked the four corners, measured the sides, and in the center planted a great ceiba tree (Ceiba pentandra). As this took place in the dry season, the ceiba tree was in flower and thus was conceived as a tree of life. The final event in this cycle was the spinning of the tree as a world axis, setting the stars in motion.

This story as reconstructed above is not merely a model for the establishment of cosmic order but can be viewed as an allegory of ancestor veneration, a fundamental concept in Classic Maya religion. In addition, the narrative may be interpreted as a model for Classic Maya rulers as manifestations of the Hero Twins, keepers of cosmic order and caretakers of the ancestors. Further, Schele and Freidel suggested that this narrative was read in the movements of the Milky Way during the course of the year (Freidel, Schele, and Parker 1993; Schele 1992b). According to their model, certain canonical orientations of the Milky Way, such as the two extending from north to south and one from east to west, are symbolized by specific icons in Maya art, such as the crocodilian known as the Cosmic Monster and the centipede jaws that mark the entrance to the underworld. The linkage of the Creation narrative to regular celestial movements strongly implies not only its universality but also its coherence as a discrete sequence of events.

While there is much to be said for the idea of interpreting Maya iconography in terms of nocturnal celestial images, caution should be exercised in the application of the "master narrative" of cosmogenesis, as described above, to isolated examples of Maya art. In particular, it can be observed that the Creation story presented above is assembled through a process of bricolage, in which elements from diverse historical traditions and contexts are combined into a single historically disconnected narrative. The entire process is based on the assumption of the existence of an underlying collective and transhistorical Maya narrative that is expressed in fragmentary form in art and literature. In fact, such an approach is not

unique to Maya art history but has been attempted in diverse fields, such as the art of the Dogon (e.g., Laude 1973)- In addition to the problem of historical confusion intrinsic to the pastiche, however, the very existence of elaborate collective myths is suspect. As noted previously in the discussion of the symbolism of stelae, it is highly unlikely that a cultural zone as large and diverse as that of the Classic Maya would be characterized by such uniformity in narrative traditions. In fact (as we shall see in subsequent chapters), not only were the monuments of Quirigua distinctive, but the stories of cosmic ordering inscribed on them preserve many unique motifs that are specifically tied to local historical circumstances. While structuralism is useful when developing theories about motifs, it cannot by itself explain why a motif appears in a given instance in history. A historically engaged interpretation of images must consider both structure and context.

The most basic level of contextual analysis is analysis of the program in which an image is located. In this book, the term "program is used in its traditional art historical sense, as developed especially in the field of medieval and Renaissance art (Gombrich 1972; von Simson 1988: 228). It refers to a complex of images and texts and the conceptual scheme that underlies this complex. To be considered a program, a group of monuments must be located in a contiguous space, oriented in the same direction or along the same axes, and erected in the same general period. At Quirigua, as at many Maya sites, the webs of meaning established among written texts and images are not necessarily limited to single monuments but extend to multiple monuments arranged in groups. In this regard, Classic Maya art is highly sophisticated, comparable to medieval church portals or Buddhist architecture.

A focus on the programmatic aspects of art privileges the designer's point of view. This is an important point, because the Quirigua monuments were created within a multiethnic milieu, with a substantial non-Maya component. Presumably, different social groups at Quirigua would have participated in various ways in the execution and use of the sculptures, resulting in diverse interpretations of their meaning. Even within the Maya minority, distinctions in social status must have related to different points of view with regard to the monuments. A complete understanding of the social significance of the Quirigua program is beyond the scope of the present book, as it would necessitate a status-sensitive comparative analysis of both Maya and non-Maya monumental practices within the region (see Ashmore n.d.). In contrast, the explanatory perspective taken in the following chapters begins by reconstructing the often esoteric messages of the monuments themselves and then expanding their interpretations into ever widening social spheres by integrating archaeological data. The reader should remain aware of the limitations and biases of such an approach.

Because of the complexity of monumental sculpture programs, it is essential to characterize precisely the relationship of images to spatially linked written texts. As discussed by Janet Catherine Berlo (1983: 13), written texts that are physically linked to images may be either conjoined (juxtaposed) or embedded (integrated into an image). The standard monumental mode at Quirigua segregates pictorial images from written texts, placing them on distinct faces of the monument. In all cases, pictorial images are presented as primary information, placed on the broadest faces of the monuments or directed toward major performance areas or processional routes, while written texts are relegated to a secondary position, usually placed on the narrower sides of a monument or on the reverse. This hierarchy of image over written text is standard for Maya art and has implications for the interpretation of the meaning of monuments.

A useful model for understanding these relationships, presented by Flora Clancy (1986), draws on Roland Barthes's theories of relay and anchorage to suggest that written texts can either complement or supplement images in Maya art. Thus, texts may either constrain or anchor meaning, by describing the events depicted; or they may extend the significance of the image by providing additional information relating to it, through a process of relay. At Quirigua images and texts are related to each other through both of these processes. Because the texts often relate multiple events which take place at different times, the text is related to the image through relay. One or two of the clauses in a monument's text, however, will normally be referenced (anchored) in the image. Thus, there is often a clear resonance between the poetics of image and text that contributes to the aesthetic impact of the artwork. In addition, the reading order of the text may suggest a reading order of images, in the cases where there is more than one image. In the end, the decipherment of the patterns of relations of image, text, and space allows for the reconstruction of the particular message that the sculptures present to the audience.

Previous studies of Maya sculpture have noted a strong retrospective focus of iconography, wherein elements from earlier sculpture are frequently quoted in later works (Proskouriakoff 1950; Schele 1979). Such repetitions through time contribute significantly to the local distinctiveness of art from the largest Maya centers. In addition, as Carolyn Tate (1992: xi) observed, the Maya "conceived the imagery of each monument in relation to nets of meanings woven by the symbols on previously existing monuments placed throughout the city." According to Tate, such webs of significance had a political interpretation, as they served to foster a sense of local identity among the inhabitants of the locales where the art was displayed. An implication of this observation is that deliberate copying of ritual iconography from site to site could be taken to indicate political relationships and positioning. Quirigua serves as an excellent test case for this theory, through the richness of its iconography and completeness of its sculptural record. It will become apparent that the sculptures of Quirigua are replete with iconographic quotations not only from the local past but from the ceremonial traditions of Copan. The specific iconographic and textual targeting of these works reveals the dynamism of local and regional ideologies of political ascendancy.

While an examination of iconography within a temporal and spatial matrix provides a means for evaluating the politico-religious history of art, stylistic continuities and disjunctions are also worthy of detailed analysis. In the words of Willibald Sauerländer (1983: 254), "Style is the mirror which makes all the buildings, the statues, the images of the past accessible to aesthetic historicism, for its dreams and for its files." As will be demonstrated, clear patterns of stylistic development can be distinguished in Quirigua sculpture during the forty-year period of continuous sculptural commissions. The most obvious change is an increasing emphasis on the sculptural block and its rectangular cross-section. Further, relief becomes increasingly shallow, so that by the last decades of the eighth century carved designs are conceived and executed as little more than slightly "enriched" drawings, wrapped around the surfaces of a three-dimensional block. How can these changes be explained? Are they the result of an intentional move by the artist(s) to express some concept, or are they merely a sort of artistic "drift," the secondary result of other cultural processes? Moreover, can stylistic changes be related to political power--and, if so, how?

As noted by Whitney Davis (199o: 26), current conceptions of style are firmly rooted in traditions of Classical rhetoric, in which style involves rules for intended verbal effects, such as persuasion or elaboration. Accordingly, these discussions center on informational content of style and intentionality. Nevertheless, as defined traditionally in art history, style is not some inherent quality or occult entity residing in a work of art but an abstraction, based on comparison between artifacts. Further, patterns of similarity in artifact styles cannot necessarily be attributed to common historical causes; in other words, we cannot always successfully read "from style to history" (Davis 1990: 26).

This qualification is especially relevant to the case of ancient Maya sculpture, when we know relatively little about its context of production and use and even less about indigenous concepts of style. For example, the apparent conservatism in stelae from a certain site could be attributed to a number of factors, such as workshop histories, deliberate copying of models, or working practices. Judging from style alone and without independent data, it is not possible to determine which of these interpretations is correct. Conversely, neither can differences in sculptures, such as the changes in those at Quirigua, be taken as sole evidence of particular historical relationships. Instead, as argued by Davis (1990: 25), style is necessarily the index or symptom of the presence ofa historical entity rather than the result ofit. This being the case, considerable care must be taken in correlating sculptural style and politics. In this study, political relationships are constructed primarily by using inscriptional and iconographic data, with stylistic comparisons serving to enrich and extend these interpretations. In such a manner I hope to bridge the gap between style and social history, which has been a perennial problem in Precolumbian art history.

The strong historical focus of this book dictates its organization. It begins in Chapter 1 by tracing the origins of Quirigua as a Classic Maya center back to its very humble beginnings as a small trading colony. Despite its subordinate status, the artistic legacy of Early Classic Quirigua is of great importance for the development of more ambitious programs during the reign of K'ak' Tiliw. Chapter 2 documents the early years of this ruler's reign (725-738), when Quirigua was directly subordinate to Copan and its sculptors worked closely in line with the practices of the larger center. This chapter theorizes that the concept of personae in sculpture, which was already present in the Early Classic period, was actively suppressed through the prohibition of portrait images during this period.

Chapter 3 discusses the political and religious significance of the sacrifice of Waxaklajun Ub'ah K'awil, which was celebrated in a series of portrait stelae erected soon after the event. These monuments, which become increasingly ambitious in scale and execution, reveal evidence of archaism, an apparent reference to the local dynasty and the ruler's legitimacy therein. In addition, the decapitation provided the basis for the development of a distinct association of K'ak' Tiliw with the lightning deity Chaak, which remained crucial to his legacy. Chapter 4 introduces the most complex program of K'ak' Tiliw's career, a group of stelae erected on Platform 1A-1 between A.D. 761 and 780. These sculptures imbued the site with a living presence and permanent ritual authority of the king through the inscription of multiple personae. Nevertheless, the entire program celebrates the cosmological significance of K'ak' Tiliw's dominant identity as a manifestation of Chaak. The climax of this program, discussed in Chapter 5, presents a local twist on the lore of Creation, demonstrating the cosmological implications of the dedication of thrones/platforms by the ruler.

Chapter 6 documents the transformation of the complex personae of K'ak' Tiliw after his death, by his successors. The change is dramatic, as his successors at first focused on K'ak' Tiliw as a great warrior then later shifted to a more ambiguous presentation under new political circumstances. Despite these transformations, the characterization of the deceased ruler as a source of divine power attests to the cumulative impact of K'ak' Tiliw's own monuments on historical consciousness at Quirigua.