Of gods, glyphs and kings: divinity and rulership among the classic Maya.

Houston, Stephen ; Stuart, David

'the mere fact of royal divinity was not so important as the relations which the king formed with other gods and men, and the contexts in which he was able to assert his divinity'.

(BURGHART 1987: 237)

New hieroglyphic decipherments now allow us to address several fundamental questions about the conceptual and religious underpinnings of Maya rulership. We can now explore the Maya concepts of relationships between deities and kings. Of particular interest are the ritual expressions of these relationships in the political and social arenas of various kingdoms. We can also attempt to delineate how relations between royalty and divinity changed over time in the Maya area, most notably after the fall of numerous kingdoms at the dawn of the Post-classic era.

The implications of these issues reach far beyond the Maya region. Scholars studying cultures from Ancient Egypt to China have confronted the question: how can rulers embody characteristics of both the human and the divine? Comparative studies show this question to be relevant to many traditional systems of authority, since rulers may tend to connect themselves with an immutable, divine order 'which transcends mere [human] experience and action' (Bloch 1987: 272). The power and mystery of divinity provides the ultimate sanction of worldly authority. There is, however, an apparent difficulty with attributing godhood to human rulers, namely, the fact that rulers are observed by their subjects to undergo the same processes as commoners do. Rulers are born, they live and die, demonstrating mutability and frailty as they do so. Some scholars have suggested that rulers may seek identification with the divine precisely because of their mortality and evident human weakness (O'Connor & Silverman 1995a: xxiii).

And yet, despite what many researchers consider to be the paradox of the concept of divine humans, cultures ruled by such hybrid divinities do not seem to find any inherent contradiction in it. As this article will make clear, a large part of the 'paradox' is created by scholarly preconceptions of what a 'god' is. The Western concept of a god as one who is all-powerful, without faults, whose existence is not marked by either birth or death, is at times indiscriminately applied to other cultures. In a belief system where gods or supernaturals are born and can die, are changeable and even capricious, and have their own vulnerabilities, it is less necessary for a ruler to explain away these qualities in him- or herself.

In 19th-century Fiji, the 'stranger king' and his family were established as beings that were ontologically and historically separate from their subjects. Rulers did not 'spring from the same clay as [their] people' (Sahlins 1981: 112). In other parts of Polynesia, rulers were likened to sharks travelling on land, rapacious, unpredictable, wholly foreign in origin - dangerous (Sahlins 1981: 112). In a very different place and time, legal theorists in Tudor England found it useful to distinguish between the king's 'body natural' and his 'body politic', the domain of 'certain truly mysterious forces (which) reduce, or even remove, the imperfections of . . . fragile human nature' (Kantorowicz 1957: 9). These societies framed authority in terms of mystical and religious forces, vested in a king 'who reigns not by force, still less by illusion, but by supernatural powers . . . [within] . . . him' (Kertzer 1988: 52).

Throughout the world's history, culturally accepted linkages between rulers and the supernatural fall into recognizable patterns, demonstrating the ease with which such associations could be made in a context of appropriate beliefs and values. Cross-culturally, arrogating divinity and its attributes directly to the ruler occurs in three typical ways.

1 The ruler claims to be divine, in direct descent from other divinities, or receives divine honours after death (Price 1987: 104).

2 The ruler is rhetorically described in terms of qualities and 'epithets appropriate to a deity', although remaining recognizably distinct from a true god (Moertono 1968: 43-4; Liebeschuetz 1979: 238). Sacrality may hinge then on the possession of legitimating icons, such as the royal drum of the Ankole kingdom in East Africa, the magical pusaka, 'holy relics of inheritance', of 18th-century Java or the ting tripods of early China (Moertono 1968: 65; Pemberton 1994: 32; Ferrie 1995: 317). Alternatively, sacrality may connect with an aura of 'dangerous, sacred force' emanating from royalty: the tapu restrictions surrounding traditional Hawai'ian elites or prohibitions regarding the imperial person in 17th-century Japan exemplify this force (Kertzer 1988: 46-7).

3 The ruler achieves divine status only on occasion, through the ritual summons of god-like forces which he appropriates for himself (Hocart 1970: 92-3). By this form of possession, godly words form on a ruler's tongue. His statements pass into the realm of unexamined, unquestioned truth, and his body becomes, as in ancient Egypt, 'suffused with the same divinity manifest in his office and the gods themselves' (O'Connor & Silverman 1995: xxv).

Another mode of relating kings to divinity involves less the practice of sharing in divinity - the three customs outlined above - as fulfilling a central role in communications between gods, humans and, frequently, royal ancestors, who operate as crucial intermediaries (Bendix 1978: 18; Keightley 1978: 212-13). To the ruler goes the important task of interpreting divine will and controlling 'human approaches to the divine and the communication of gods to men' (Beard 1990: 30; see also Moertono 1968: 40-41). Contrast with this situation the case of Classical Athens, where religious functions took place on many different levels, involving people of varying status (Garland 1990: 90). Perhaps the unique characteristic of royal interpretation of divine will is its applicability to all subjects in a polity. Finally, a divine ruler's human qualities, particularly his mortality, may be cast in such a way as to exemplify larger, cosmic cycles or patterns. In this manner the symbolic attributes of the ruler negate common, human ones, or at least elevate them to another dimension of meaning (Feeley-Harnik 1985: 281-2). To quote Bagehot, 'a princely marriage is the brilliant edition of a universal fact, and, as such, it rivets mankind' (cited in Cannadine 1987: 7).

Royal divinity can also be reinforced by myth and ritual. To those who believe, myths provide incontrovertible, narrative rationales for why things exist in the way that they do. A subset of myths includes royal charters, stories that justify or explain regal behaviour. Similarly, as 'highly structured, standardized sequences', rituals often engage distant events, forces, or beings that are described in myth or charters and make them tangible and potent in the present (Kertzer 1988: 9). The parading of god effigies - seen extravagantly in ancient Egyptian processionals (Kemp 1989: 205) or Sumerian Gotterreisen (Sjoberg 1957-1971: 481) - underscores royal pretensions of affinity with the gods. To spectators, the gods concretely and visibly participate in the ruler's ceremonies. In much the same way, Mesopotamian rulers boast of 'housing' gods in sumptuous dwellings and enjoy, particularly in the late 3rd millennium BC, the role of physical proxy in the marriages of gods (Bottero 1992: 225-6).

Broadly speaking, then, there exists considerable variety in royal identifications or interventions with gods. Rulers may lay direct claim to divinity, or do so rhetorically by using godlike titles and demanding the ritual veneration due to gods. They may possess divine force intermittently, employing godly costumes and behaviours to summon supernatural presences. Further, lords may invite gods to witness and validate ceremonies (Liebeschuetz 1979: 43), often through god effigies or physical proxies that may be paraded ceremonially. A more subtle invocation of divinity consists of stories that liken royal lives to the immutable patterns set by gods. Implicit here is not only the notion of remote events and beings, but the continual repetition of such patterns in later times. For Mesoamerica, Nicholson (1971a) calls this 'pattern history', founded on the idea of recurrence: as calendar cycles, or certain permutations of these cycles, repeat, they produce like-in-kind repetitions of mythological or historical events.

The artistic and documentary sources of the Classic Maya employ all these methods for linking rulers with the divine. Kings make frequent use of the explicit title k'ul ahaw,(1) or 'divine lord'. In rituals, lords also frequently 'impersonate' gods by the wearing of deity masks, clothing and ornament. Gods or spirits may also appear in ceremonial situations as actual witnesses or participants, perhaps as effigies of wood, stucco, or stone. Before documenting these manifestations of divine rule, however, we must address the complex nature of 'gods' in Classic Maya and Mesoamerican belief.

The nature of Maya gods

Ancient Maya sources are replete with depictions and mentions of supernatural beings, most of which are commonly called 'gods' by students of Mesoamerican religion. The Mayan word is k'u or ch'u, the pronunciation being dependent on the particular language, be it of the Yucatecan or Greater Tzeltalan branches. But 'god' is not always a satisfactory translation. K'u or ch'u - which more accurately means a 'sacred entity' when used as an adjective k'ul or ch'ul (as in k'ul ahaw, 'holy lord') - has the meaning of 'holy, sacred, divine'. With these cautions, we retain the term 'god' for most major deities while recognizing its limitations and understanding that not all supernatural entities can be grouped under a single, inclusive term.

Partly in reaction to Schellhas' compilation of Maya divinities (1904), Tatiana Proskouriakoff and others (Proskouriakoff 1965: 470-71; 1978: 113, 116-17; Marcus 1978) make four, related assertions: that the idea of 'gods' results from the spurious application of Old World parallels (Marcus 1978: 180; Proskouriakoff 1978: 113; Marcus 1983: 345,349,351; Marcus & Flannery 1994: 57); that the concept pertains only to a few, late periods in Mesoamerican antiquity, especially those at a state or imperial level of political organization (Kubler 1969:32; Grove 1987: 426; Marcus 1992: 270-71); that the notion of a 'god' inherently distorts nuances of indigenous belief (Beals 1945: 85; Marcus 1989: 150-52); and that most 'gods' in Mesoamerica represent euhemerized ancestors (Proskouriakoff 1978: 116-17). Rather than devising a pantheon, a roster of gods organized into a family on a Greco-Roman model, ancient Mesoamericans categorized and worshipped vital, impersonal forces of nature. These forces embodied essences that animated all (or most) things in nature and incorporated the powerful, intercessionary spirits of ancestors (Spores 1984: 85). There is some merit to such views, and the critics are correct in questioning indiscriminate use of the term 'god'.

To illustrate the complexities, we can point to an important category of supernaturals known as wayob (singular way), the 'animal companion spirits' that helped constitute the psychological and spiritual make-up of Maya lords, rulers and places (Houston & Stuart 1989; Grube & Nahm 1994). Among modern native Mesoamericans, these entities are often called naguales, and remain an essential aspect of native Mesoamerican spirituality. They are consistently viewed as an aspect of the human soul, sometimes wandering at night from their sleeping hosts. This connexion is no doubt reflected in the alternate meaning of way as the verb 'to dream'. In their depictions in Classic Maya art, usually on the exteriors of polychrome drinking vessels, wayob are shown as animal composites or as animals with unusual behavioural or bodily attributes. They are also explicitly linked with people. These depictions, then, are of ancient royal souls, or parts of these souls, and constitute an important key in the study of ancient Maya religion. Significantly, the way entities seem separate from the notions of k'u or ch'u, and we prefer not to call them 'gods'. This, at least, is a distinction the Maya were apparently careful to make.

Ch'u is the foundation of the word ch'ulel, which appears in Chol Mayan and the Greater Tzeltalan languages with the meaning like 'vitality', but perhaps more literally 'holiness' (the term is composed of ch'u and the abstractive suffix -lel). Widely translated as 'soul' or 'spirit', it more correctly refers to the vital force or power that inhabits the blood and energizes people and a variety of objects of ritual and everyday life (Vogt 1969: 369-71). This general Maya conception is essentially identical to conceptions of divinity found elsewhere in Mesoamerica. The Classical Nahuatl word teotl, also widely translated as 'god', is more appropriately understood as 'a numinous, impersonal force diffused throughout the universe' (Townsend 1979: 28; see also Hvidtfeldt 1958). Burkhart (1989: 37) aptly classifies this system of belief as 'polytheist monism . . . (a) divine principle manifested . . . in multiple forms, some ambivalent, some expressing opposite principles in their different manifestations'. Ethnographic research among Mixtec-speaking peoples in Mesoamerica confirms the durability of this concept: potent forces, some linked to the earth, others to wind, water and the sky, present different faces or aspects for human apprehension (John Monaghan pers. comm.). Humans receive only partial glimpses of a divine totality, often in manifestations we call 'gods' (Townsend 1979: 28).

Not surprisingly, then, it is difficult to develop an inclusive and satisfactory definition for Maya 'gods'. They may assume special human or animal forms (often both), and embody certain specific natural forces, such as lightning, wind, or the essence of maize (Taube 1985; 1992a).(2) Other, more specialized supernaturals seem to be narrowly conceived in connexion with specific places or socio-political entities. The Maya situation is similar in many respects to that which Nicholson (1971b) has described among the Aztec, where we find hierarchical categories of supernatural figures, each with diverse 'aspects' and sometimes overlapping attributes. According to extant fragments of Classic period mythology, some supernaturals have birth-dates and named parents (Berlin 1963; Kelley 1965). Also, many Maya gods existed in two or more planes, living within sacred narratives far-removed from the present world as well as participating directly in the ritual activities of humans. As Michael Coe (1973: 22) and others suggest, sacred narratives not only worked to explain the patterning of natural events, but could establish charters for human, usually royal, behaviour. As actors and participants in rituals, gods could interact with powerful humans in an almost routine manner. Interestingly, many of these basic features of native religion survived the European conquest, and remain prevalent in Mesoamerica to the present day (Gossen 1986).

In regionalized incarnations, Mesoamerican gods enjoyed tutelary relationships with particular socio-political groups (Lockhart 1992: 16), with whom they had an almost contractual relationship of quid pro quo transactions (Thompson 1970: 170). Nobility was defined in part through its direct association with particular gods. Crucial titles of rulership, as in Chalco, Mexico, involved the concept of 'god-possessor lord', perhaps reflecting an earlier notion of lords as carriers of god effigies (Schroeder 1991: 122-3, 142, 172-3). Similar evidence appears in the Mixtec region of Mexico (John Monaghan pers. comm.). Below we demonstrate the ways in which Maya rulers associated themselves with gods, often in similar ways.

Glyphs for gods

The key glyph in discussing Maya conceptions of divinity is the 'God C' sign [ILLUSTRATION FOR FIGURE 1 OMITTED], deciphered as a logograph with the value K'U(L) or CH'U(L) (Barthel 1952: 94; Ringle 1988). As a proper noun, this hieroglyph conveys the ideas of k'u, 'god, sacred entity', as already described; when prefixed to other signs, it also may be read as the adjectival form k'ul, 'sacred'. The sign is extremely common in the inscriptions of the Classic period, suggesting that the ancient texts are a rich source for understanding Classic beliefs.

A basic function of the K'U(L) sign appears on the so-called 'Vase of the Seven Gods', which shows two rows of seated supernatural figures before an enthroned underworld deity ([ILLUSTRATION FOR FIGURE 2 OMITTED]; see Ringle 1988: 3, 5). According to Maya convention, these individuals - clearly non-human in their faces - rest on two base-lines, sitting side-by-side in two rows. The extruded eyeballs and dark background of the scene lend a sinister, nocturnal quality to the image. The hieroglyphic text in the middle includes a long list of deity names, each followed by the K'U glyph. Each of the right-facing figures, then, are designated as the 'so-and-so "god"'. A verb precedes this list of god names and follows, in turn, the 4 Ahau 8 Cumku date of Maya creation. The event may indicate that these gods are 'multiplied', 'ordered', 'added together' at the beginning of the current creation (Freidel et al. 1993: 67-9).

Often certain prefix and affix signs qualify the sign for 'god' in ways that affect its reading and meaning. For example, the K'U sign often takes the prefix element U-, serving as the pronoun u-, 'his, hers, its', and the nominal suffix -il [ILLUSTRATION FOR FIGURE 3 OMITTED]. In Mayan syntax, these affixes signal possession, so that the name of the possessor - the person to whom the 'god' belongs - comes next in the phrase. In this way the term 'u-k'u-il X' would render 'X's god'. As one might suspect, the appearance of this phrase in the inscriptions helps us to understand the relationships between deities and humans. Several related inscriptions from the 'Cross Group' temples at Palenque, Mexico, contain many such explicit statements of god 'ownership', and thus merit a more detailed analysis.

The texts of the three temples in the Cross Group generally relate information about the 'Palenque Triad' gods, three mystical brothers who were important tutelary deities of the local dynasty (Berlin 1963; Kelley 1968). Each of the three temples concerns one member of the Triad: the Temple of the Cross with the deity known as 'GI', the Temple of the Foliated Cross with 'GII', and the Temple of the Sun with 'GIII'. Each god was born in the far distant past; the main tablets associated with them connect their mythic history with the early Palenque kings (see Lounsbury 1979). Secondary inscriptions located outside the inner shrines of the temple give important dedicatory information on the construction of the temples and the 'housing' of the gods within. Significantly, the inner shrines of these temples are explicitly 'owned' by the deities themselves. This concept is reflected throughout Mesoamerica, where temples are almost universally considered 'gods' houses'. There are a few other instances in which gods possess things. A text on the lid of a stone box parallels the Palenque material (Coe 1973: plate 7). The inscription, beginning with a date and verb, continues with an expression probably reading U-PAS-TUN-li, glyphs spelling 'open stone' (a reference to the lidded box) along with the usual possessive affixes. To judge from the remainder of the inscription, the box belonged to two gods, identified as 'the gods' of a ruler of Tonina, a site relatively close to Palenque.

Typical of the dedicatory texts of the Cross Group is the inscription from the balustrade (or alfarda) of the Temple of the Foliated Cross. According to this text, on the day 1.18.5.4.0 1 Ahau 13 Mac (8 November 2360 BC) the god GII was born at a place called Matawil, and some 3000 years later, on 9.12.19.14.12 5 Eb 5 Kayab (12 January AD 692), the 'god' of K'inich Kan Balam, the contemporary Palenque king, 'entered the house'. One may safely assume here that the ruler's 'god' is GII himself, although he is not named in this second dedicatory passage. A related inscription from the door jamb of the shrine of the same temple presents the information in a slightly different way. This text states the same 'house entering' event involving GII, now named, but states that the deity is the 'cared-for thing' or 'precious thing' (huntan) of the ruler K'inich Kan Balam. Interestingly, the word huntan is more often used to express the relationship between a child and its mother ('K'inich Kan Balam is the precious thing of the Royal Lady Ts'ak', for instance). Although no precise kin relationship is expressed, it would seem that a ruler was thought to 'care for' a god, perhaps through sustaining sacrifices, much in the way a mother cares for her offspring. It is doubtful, however, that such statements can be extended to mean that rulers were considered 'mothers' of deities (cf. Schele & Freidel 1990: 475; Stuart 1984).

These inscriptions and their alternative phrasings demonstrate beyond doubt that the rulers could be considered 'owners' of important deities. The possessed u-k'u-il glyphs do not allude to concepts of 'temple' nor to any abstract, impersonal invocations of 'holiness' (cf. Schele & Freidel 1990: 473, [ILLUSTRATION FOR FIGURE 6:15 OMITTED]), but rather pertain to distinct and personalized sacred entities. One may go so far as to suggest that such references may allude to specific images of gods, for the identical phrase u-ch'u-il means 'his idol' in Colonial Cholti Mayan (Fought 1986). The corresponding hieroglyph, shown in FIGURE 3, is common in many different types of Maya texts, including several touching on themes of warfare and conquest. Inscriptions at Tikal, for example, refer to 'the gods' of a ruler from El Peru and another of a Naranjo lord, both of whom were apparently the victims of military defeats. Although the readings of these passages present certain problems of interpretation, we concur with the suggestion by Simon Martin and Nikolai Grube that the defeat of neighbouring kingdoms may have involved the appropriation, capture or desecration of foreign god effigies 'owned' by royal victims (Martin n.d.).

Ideas of god ownership are not the sole source of divine qualities ascribed to rulers. In an iconographic usage, the k'u glyph appears as streams of liquid falling from the hands of rulers in sacrificial costume. The streams represent royal blood shed in self-sacrifice (Stuart 1984; 1988); their depiction as the k'u motif most likely signals the concept of the ch'ulel, which according to numerous ethnographic sources, constitutes a part of the soul and inhabits the blood of all humans (Vogt 1969). Power and prestige are defined by many modern Maya groups by the degree of one's 'heat' and the corresponding strength of one's ch'ulel (Guiteras-Holmes 1961: 72). Ancient rulers, much like high-ranking Maya of today, may have had 'stronger' or 'hotter' souls which could be channelled, in effect, to sanctify and bless ritual objects and subordinate persons. The ch'ulel of the rulers - it may have gone by another term in Classic times - was probably a central focus of much royal blood ritual, and, perhaps, a major factor in defining the divine qualities of royal office.

Rulers as gods

Classic Maya rulers made direct claims to divinity by means of certain royal titles that make use of the K'U(L) sign. The 'Emblem glyph', an exalted title used almost exclusively by kings, is the most important of these. The Emblem title includes the term for 'lord', ahaw, the name of a place over which the lord exercised or claimed dominion, and - as its distinguishing attribute - k'ul, 'sacred' or 'holy' (Mathews 1991: 24; Stuart 1993: 326). The 'holy' intimates that the ruler holds a quality shared with few others, a quality that presupposes a claim to divinity or, more precisely, god-like status.(3) In an intriguing pattern, the k'ul ahaw epithet is common only rather late in the Classic period (Houston 1989: 55). An argument can be made that the Emblem title - rather like hueitlahtoani ('great ruler') of the Mexica Aztec - reflects a need for new, more exalted grades in society and a distinction between the ruler and a burgeoning group of nobles, many of royal descent. There is increased emphasis on royal ladies who use the honorific title k'ul ixik, 'holy woman', at about the same time. To restrict the number of ahawob, rulers may have used the expedient of bilateral descent to define royalty through paternal and maternal blood-lines, a pattern well-documented among Mixtec rulers, who lived within 'a closed social universe that could be legally penetrated only by birth' (Spores 1967: 141). Nonetheless, present Mayan hieroglyphic evidence shows somewhat more flexibility than existed among the Mixtec.

Another claim to divinity is evident from the personal names of Maya rulers (see Geertz 1977: 158; 1980: 124, for a similar pattern in Bali). Many names incorporate references to deities, one of the most common being the initial element K'inich ('Sun-faced', a descriptive name for the sun), as in K'inich Kan Balam ('Sun-faced Snake Jaguar'). Other royal names describe aspects of deities, such as the Yaxchilan ruler Itsamnah Balam, or 'Itsamnah Jaguar' (widely known as 'Shield Jaguar'), Itsamnah being the name of the very important deity sometimes known as 'God D'. This name would seem to describe a jaguar that assumes a partial identity with the deity Itsamnah. Two rulers of Classic times share the name Itsamnah K'awil. K'awil is the name of another deity of great importance (widely dubbed 'God K'); again the name seems to intimate that these kings are somehow 'hybrids' of these supernaturals. Other royal names, more descriptive, are hardly less opaque: 'Chaak is born from the sun' [ILLUSTRATION FOR FIGURE 4A OMITTED], 'K'awil is born from the sky' [ILLUSTRATION FOR FIGURE 4B OMITTED], 'K'inich is born from the sky', 'K'awil is conjured' [ILLUSTRATION FOR FIGURE 4C OMITTED] and 'K'awil is born' (Chaak is the Rain God, K'inich the Sun God). The significance of these names is unclear. Decipherments provide readings of names but, to date, little understanding of why the Maya favoured certain names over others. Sometimes royal names skip one or more generations, suggesting that that they could have only one living bearer. Often, scholars focus on the initial elements of names and conclude, erroneously, that the shared signs point to use of the same name at different sites [ILLUSTRATION FOR FIGURE 4 OMITTED]. In fact, it is the final sign, usually a deity, that forms the crucial component; preceding glyphs simply provide subtle, adjectival modifications of the god name. We have seen in the examples just cited how two outwardly similar names beginning with the same glyphs pertain to different gods.

As in ancient Egypt, the outright assertion that Maya rulers were considered 'gods' remains problematic (Baines 1995: 6, 10-11). Texts which apply god names to rulers suggest they are 'holy' (an epithet often limited to the ruler, his spouse and royal ancestors), but not once are living kings said directly to be gods. Much clearer fusions of gods and royals occur with deceased rulers, who may begin to be venerated as ancestral heroes or founder-leaders but over time take on the guise of deities. As Carrasco (1950: 143, translation in Townsend (1979: 34)) states with regard to Central Mexican religious history, 'one of the main processes by which Mesoamerican religions produce such great quantities of deities is the deification of an ancestral tribal leader, who assumes the attributes of the gods of the tribe he represents'. This type of transformation may have taken place among the Maya of ancient Copan, who in their later years erected the largest temples of the site to the great historical founder K'inich Yax K'uk' Mo' (Fash 1988; Stuart 1992).

Posthumous royal portraits tend to fall into three classes (McAnany 1995):

1 static images of seated or standing lords, pictured as they might appear in life [ILLUSTRATION FOR FIGURE 5A OMITTED];

2 views of rulers in transformation or metamorphosis, usually merged with the attributes of the Maize God or plants [ILLUSTRATION FOR FIGURE 5B OMITTED]; and

3 depictions of disembodied, deceased lords, wreathed in smoke [ILLUSTRATION FOR FIGURE 5C OMITTED].(4)

Of these portraits, only rulers and their spouses seem ever to adopt the features of divinity. Royal fathers may occur as Sun Gods or individuals encased in sun disks, while mothers are identified with the Moon Goddess. Sometimes recently deceased rulers appear in the guise of the Maize God. Thus royal parents pair with the two most prominent features in the sky, each diurnally opposed to the other, while the Maize God, an emblem of youth, sustenance and vegetal regeneration, represents a transformational cycle or a mortuary charter that likens rulers to the first human, who was fashioned of succulent maize dough (Houston 1995). Yet, despite such explicit representations, we must remember that living rulers seldom made unequivocal claims to divinity. Seemingly 'holy' and 'god-like' during their life-times, they were probably set apart from actual Maya deities.

A related topic is the naming of dynastic founders, some of which seem to be described as kinds of stars, as at Dos Pilas (Schele 1992; Houston 1993: [ILLUSTRATION FOR FIGURE 4-5 OMITTED]). By Classic times, the Maya may have endowed these progenitors with divinity, but there is nothing in their titles - many use Emblem glyphs - that would suggest a markedly different status from later lords. One stela at Tamarindito, Guatemala, may show a Late Classic lord dressed as the founder of his dynasty (Houston 1993: [ILLUSTRATION FOR FIGURE 4-5 OMITTED]). Regrettably, and rather strangely, no texts occur on this monument, perhaps because the image was intended to evoke a pre-literate period. (Other early, fragmentary monuments were incorporated into the structure behind the stela, suggesting later refurbishment of a building associated conceptually with the early years of the Tamarindito dynasty.) The only ancestors that seem to be depicted explicitly as gods are those in the ascending generation or deceased kings, often bearing the attributes of the god K'awil. This contrasts with some central Mexican beliefs, in which patron deities of particular socio-political groups merge with the '"deified tribal ancestor" or "first founder"' of a community. Temples dedicated to such gods were 'the symbol of the town's independence and integrity, and, in one sense, its luck and fate', so that military conquest 'was signalized by the burning of the patron deity's shrine, frequently followed by the carrying off of the latter's image' (Nicholson 1971b: 409).

The gods made animate

Under special circumstances, the distinction between rulers and deities appears to have been purposefully vague. Kings and high nobles possessed the special ability to assume the identities of certain gods through ritual impersonation. Geertz (1977:157) describes impersonation as an aspect of the ruler's charisma, 'liminally suspended between gods and men'. At this point the Aztec concept of teixiptla becomes important, for it allows us to understand the subtleties and implications of god impersonation in Mesoamerica. The teotl, the divine energy, manifests itself in the teixiptla, 'the physical representation or incarnation of the teotl . . . [which is] called forth by the creation of a teixiptla' (Boone 1989: 4; see also Hvidtfeldt 1958: 76-100). In Postclassic central Mexico, costumes, masks and effigies of gods do not 'represent' deities. They are gods in the sense of being partial extensions of divinity. In some instances, there 'is such a resemblance between image and god that . . . visible forms charged with sacred power are considered to be gods themselves' (Lopez Austin 1993: 137, 138). When dancers don masks or other elements of godly costume, the 'essence of the god . . . become[s] present in material form', much as it does for Puebloan Kachina dancers (Markman & Markman 1989: 69). In his study of 'man-gods' in the Mexican Highlands, Serge Gruzinski (1989: 22, 23) comments that, through association with divine force, often present in 'sacred relics', 'something penetrated the man, possessed him, transformed him into a faithful replica of god', made him part of 'the very authority he adored'.

Recent decipherments of Maya hieroglyphs indicate that similar concepts prevailed among the Classic Maya. A distinctive, formulaic phrase is now identified that introduces the names of kings and their close relatives, often accompanying portraits of rulers as impersonators. The relevant glyphs read u-bah-il, possibly 'his body' or 'image', followed by a sign representing a banner or flag decorated with bar-and-dot numbers. This banner sign can be replaced by the syllables a-nu, suggesting its full value of ANUL or ANUM, perhaps related to Yucatecan anum, 'famous' [ILLUSTRATION FOR FIGURES 6A, 6B OMITTED].(5) Occasionally, these signs are conflated or compressed into a single glyph block [ILLUSTRATION FOR FIGURE 6C OMITTED]. One example includes the addition of the suffix K'U. After these combinations come two phrases that complete the expression: first, the name of a deity (along with some rare prepositional phrases); and second, the personal name of a ruler or noble. We interpret this expression as '(it is) the image of . . . the famous "god"', followed by the name of the ruler, lord, or lady who impersonates the god. As it happens, tracing this pattern throughout the inscriptions leads to the identification of several previously unknown Maya deities.

Several deities mentioned in this 'impersonation phrase' are clearly identifiable. The Maize god occurs in several cases [ILLUSTRATION FOR FIGURE 7A OMITTED], as does the Sun God [ILLUSTRATION FOR FIGURE 7B OMITTED] and a sinister-sounding deity named the 'the black hole lord' [ILLUSTRATION FOR FIGURE 7C OMITTED]. Several of these phrases occur in direct association with portraits of rulers in the actual guise of the named deity. FIGURE 8 shows three examples. The first illustrates a correspondence between rulers holding staffs and dressed in elaborate capes with agnathous jaguars; jaguar markings occur on the face of one figure [ILLUSTRATION FOR FIGURE 8 OMITTED]. The god impersonation glyphs show this deity is a being, the Jaguar God of the Underworld, whose name in two cases here is preceded by the sign for smoke or fire. We suspect this god parallels a central Mexican deity, Huehueteotl, the old fire god (Nicholson 1971b: 412-13). Apparently, the Maya rituals kindled fires with a lashed staff, perhaps a ceremonial fire-drill. One image, on Naranjo Stela 30 [ILLUSTRATION FOR FIGURE 8C OMITTED], carries several records of such events as they are linked to calendrical rituals.

Another deity impersonated by Maya lords and ladies is named with the glyphs in FIGURE 9. Unlike the Jaguar God of fire, this deity seems to be aquatic, represented as a serpent with a water-lily bound to its head. Previously the name of this 'water serpent', as we shall call it, has been erroneously interpreted as a blood-letting expression (Schele 1982: [ILLUSTRATION FOR FIGURE 50 OMITTED]). The name glyphs of the deity include the term for 'snake' or 'snake-house', as well as, in the case of male impersonators, the enigmatic term yax chit (?), perhaps a mythological referent.

Another important supernatural may be impersonated in a portrait on a sculpted door jamb from Xcalumkin, Campeche [ILLUSTRATION FOR FIGURE 10 OMITTED]. Here the inscription identifies something known as '18 Ubah Kan' with the exotic costuming usually linked with the great Mexican metropolis of Teotihuacan. The name 18 Ubah Kan occurs in other inscriptions in association with Teotihuacan symbolism, and may refer to some appropriated deity from that site. Maya lords often linked themselves to Teotihuacan, long after its decline as a major force in Mesoamerica (Stone 1989; Taube 1992b; Freidel et al. 1993: 308-12; Stuart 1994).

The hieroglyphic commemorations of such ritual impersonations are common; they clearly provide important evidence conclusive to questions of divine kingship. Much like modern Maya in highland Chiapas (Vogt 1993: 116), Classic Maya lords episodically adopted the names and costuming of particular gods and performed rituals appropriate to those deities, such as fire-drilling. The similarities to better-documented practices in central Mexico (Hvidtfeldt 1958; Klein 1986) are sufficiently striking to suggest that Maya impersonations were not simply mummery and costumed drama. Rather, rulers and certain non-regnal figures shared in some manner the divinity of those gods. The costuming offered not so much a theatrical illusion as a tangible, physical representation of a deity. Significantly, these impersonations were not reserved for high kings. As we have seen, royal women and high-ranking nobles also assumed these roles. Conceivably, those divine qualities we might otherwise associate with kingship were distributed more widely among members of the Maya elite. Alternatively, instead of diluting the singular divinity of rulers, impersonations by political subordinates may have been cast in terms of mythological subordination. In some images on carved royal thrones, high-ranking nobles are shown in mythical guises as supports of the king's seat, just as certain gods sustain the earth or heavens.

The question remains why certain gods were selected for impersonation. Why was it deemed necessary, for example, for a noble to assume the identity of the Maize God or for the 'Water Serpent' to undertake certain rituals? Lost details of mythic narratives once held some of the answers, no doubt. Impersonators, in any event, may have been considered recurring manifestations of deities who 'participated' in repeating ritual cycles.

As gods be our witness

Aside from rulers 'possessing' gods and assuming their identities on certain occasions, we find in the Maya texts how deities could otherwise become participants of sorts during royal ceremonies. Both iconographic and textual sources reveal that gods were invoked or summoned by various means to witness certain rituals. Several depictions exist of supernaturals floating in clouds near or above rituals, possibly emanating from incense or burnt paper [ILLUSTRATION FOR FIGURE 11 OMITTED]. In the inscriptions, the act of conjuring spirits or deities is rendered by the term tsak, literally 'to conjure clouds' in Yucatec Mayan. The hieroglyph for this event is the so-called 'fish-in-hand', read TSAK, occasionally spelled syllabically tsa-ka. In the inscription of the Tablet of the Cross at Palenque, we read that the ruler K'inich Kan Balam 'thrice (?) conjures his god(s)', apparently in reference to the three deities of the Palenque Triad [ILLUSTRATION FOR FIGURE 12 OMITTED]. Through such royal acts of conjuring, deities were somehow manifested to become participants or witnesses in ceremonies.

Many ritual activities are said to occur 'in the company of' or 'in the sight of' (y-ichnal) deities [ILLUSTRATION FOR FIGURE 13 OMITTED].(6)

An inscription from Piedras Negras, for instance, records the 'receiving' of a war helmet by a ruler 'together with his god(s)'. A triad of deities is named, perhaps constituting a set of localized supernatural patrons, as found at Palenque and other sites. Later in this same text we read that the 'Holy Lord conjured the gods', a phrase which tells us that the deities were invited to participate and sponsor the ceremony through the direct solicitation of the ruler.(7)

This evidence from the inscriptions shows that 'gods' operate not as distant creator beings, coupled exclusively with incidents in remote time and space, but rather participate as ritual sponsors, particularly at moments when rulers receive key regalia under the authority of gods. Moreover, even though some of the rites described are approximately the same (the reception of regal emblems), the emphasis on certain gods varies from site to site. In the case of Palenque, the triad of GI, GII and GIII are of special interest only to the local dynasty, suggesting the existence of tutelary gods in purely local association with dynasties. This is apparently also true of other centres, where the same deities may appear in different aspects as the foci of distinct, localized cults. The records of sites in the Petexbatun region of Guatemala, for example, contain consistent references to 'GI-K'awil', perhaps a deity pair or, alternatively, a hybrid form of two entities, rather like those described above in some royal names [ILLUSTRATION FOR FIGURE 14 OMITTED]. GI-K'awil owned a stela, according to one text, and perhaps also took the form of a cult effigy that was erected or dressed. According to another inscription, an enemy may even have destroyed the god's 'banner' in an act which presumably humiliated the dynasty connected to this god or god pair.

Another deity is closely linked to the Copan dynasty and especially the unfortunate ruler 18 Ubah K'awil, who was taken captive by the ruler of Quirigua. Quirigua Stela I describes this deity as 'the god of 18 Ubah K'awil' just a few days before the captive lord was decapitated [ILLUSTRATION FOR FIGURE 15 OMITTED]. The action recalls the Aztec practice of seizing effigies of enemy gods and then housing them 'as spiritual hostages in a special building, the coateocalli' (Townsend 1992: 91). By this means, the Aztecs absorbed and usurped the cults of the vanquished, undermining their claims to an independent spiritual identity.

We believe such patterns profoundly affect our understanding of Classic Maya divinities. Rather like central Mexico and even Classical Antiquity, Maya deities display complex localized aspects and political associations (see Weber 1978: 413-14; MacMullen 1981: 1-7). There may be a K'awil or a 'GI' venerated at several sites, but it is not so much a single god as multiple, distinctly conceptualized versions forming a deity 'complex'. Such deities also suggest something of the extraordinary complexity of Classic Maya theology. There is no one set of gods codified and venerated by all Classic Maya. Rather, there are localized cults. A god revered at one site may partly share the name of a god at another, but we cannot presume an identity of ritual roles, meanings, or history of development. A 'creation' event at Dos Pilas, Guatemala, indicates the participation of local gods at an event usually interpreted in pan-Maya terms (cf. Freidel et al. 1993: 64-75 and Houston 1993: [ILLUSTRATION FOR FIGURE 4-4 OMITTED]). Future studies of Classic religion must take this variety into account and avoid using one site, especially Palenque, as a paradigmatic model for beliefs elsewhere in the Yucatan peninsula (cf. Freidel & Schele 1988).

Godly images

It would be a mistake, in our view, to assume that the 'participation' of deities in royal ritual was an abstract ideal, induced through hallucinogenic visions and 'conjuring'. Rather, we suggest that, much like in Postclassic Central Mexico, Classic Maya courts possessed abundant images of gods comparable to the Aztec teixiptla, described above. A possible glyph for such images occurs in the jumbled stucco inscription of Temple 18 at Palenque, Chiapas [ILLUSTRATION FOR FIGURE 16A OMITTED]: U-wi-ni-BAH, u-winba, 'effigy, image' in Yucatec Maya. Another glyph of the same meaning occurs on Dos Pilas Stela 15, and glyphs for k'oh, 'mask' or 'image', have been found in several inscriptions (Freidel et al. 1993: 65). In artistic representations, as well, cult effigies are commonly depicted. At Palenque an image of a god is presumably unwrapped from an enclosing bundle [ILLUSTRATION FOR FIGURE 16B OMITTED]. According to Macri (1988: 116-17), the ritual dressing of effigies of the Palenque Triad gods is a major thematic focus of the texts within the Temple of the Inscriptions at Palenque.

The physical remains of Classic period cult effigies are understandably rare, apparently being manufactured of wood and stucco for the most part. A large Preclassic figure of the rain god Chaak was discovered within a ritual cave by Ian Graham (see Stuart & Stuart 1977: 53). Archaeologists at Tikal, Guatemala, found disintegrated wooden effigies of the god K'awil - very similar to those bundled figures depicted in the Palenque tablets - in Burial 195, a royal burial near the centre of the city ([ILLUSTRATION FOR FIGURE 16C OMITTED]; Coe 1990: [ILLUSTRATION FOR FIGURE 198 OMITTED]). Accordingly, we have textual references to effigies as well as their physical remains.

We should mention one final category of image. El Zapote Stela 1 [ILLUSTRATION FOR FIGURE 17 OMITTED], a monument dating to the Early Classic period, depicts the image of a Maya deity, a variant of Chak, the rain god. The text on the back of the stela clearly specifies that the stela (u-lakam-tun-il) belongs to this god, who in turn is the deity (u-k'u-il) of a local lord. Later, the text apparently establishes an equivalence between the god and his monument, as though both were one and the same. Effigies may also have taken the form of vessels, such as the GI cache vessels of the Early Classic period, that received the same sort of animating force attested for later Lacandon Maya 'pot gods', lak-il k'uh (McGee 1990: 51-2). Like teixiptla, the 'pot gods' are periodic receptacles for divine force and the tangible medium through which gods consume offerings made to them.

To claim the Classic Maya made use of cult effigies or 'idols' may state the obvious, but some debate exists on how Maya religion may have changed during the transition between the Classic and Postclassic eras. Indeed, this issue has received far less attention than deserved. According to much received wisdom, derived largely from native Maya chronicles composed during colonial times (see Tozzer 1941: 23), the 'collapse' of numerous Classic-period kingdoms was followed by a period of intense 'Mexican' influence, where the veneration of rulers in the 'stela cult' gave way to idolatry and, according to at least one source (Seler 1898), 'bloody sacrificial rites'. As Taube (1992a) has demonstrated, however, the gods of the Postclassic era were closely linked to Classic period antecedents. Although the infusion of central Mexican culture into northern Yucatan in the Terminal Classic period did involve the adoption of new deities, 'cults' and new iconographic ideals, the essential 'Maya-ness' of the religion encountered in the conquest of Yucatan cannot be denied. Certainly sacrificial rites were as old as Mesoamerican religion itself. This is why the later traditions of Central Mexico and Yucatan are reasonable models for many aspects of Classic period religion, including the veneration of cult effigies. The chief disjunction between the Classic and the Postclassic religious paradigms concerns the changing nature of royal oversight of ritual activity, or at least in the way this was presented in the hieroglyphic writing and art. With the political disintegration of many lowland kingdoms in the 9th century, some royal ancestral cults no doubt foundered; while the role of kingly ritual underwent drastic transformations, the underlying nature of deities and the means of representing them remained startlingly similar.

The practice of using cult effigies continues to be common, in one form or another. Today, ritual processions take place among the highland Maya of Chiapas, Mexico, where movement 'consists of a group of ritualists walking single-file, in fixed rank order, from one shrine to another' (Vogt 1993: 42). A few such processions, often stretching over days, are known for Classic Maya rulers (Stuart & Houston 1994: 90-92). Remarkably, these resemble ritual movements of modern Maya in that they involve east-to-west movements (Gossen 1972: 147), which set out 'from the direction of the rising sun and flow[ing] along its path' (Vogt 1985: 488, 489; 1993: 43). More broadly still, they recall the triumphal progress of charismatic rulers in the Old World, who may employ processions to 'locate the society's center and affirm its connections with transcendent things by stamping a territory with rituals signs of dominance' (Geertz 1977: 153). Whether or not one accepts Geertz's notion of exemplary rulers - and there are grounds for questioning its reductionistic view of monarchy in the region where he defined it (Barth 1993: 221-4) - these processions served as more than casual outings.

The custom of carrying god images during the Classic period may contribute in part to the current practice in some Maya communities, which meld syncretic images of Catholic saints with concepts of local sovereignty. When images sally forth in seasonal processions, they survey the boundary of their domain and function 'as principal participants [in ritual] rather than mere . . . objects of devotion' (Watanabe 1992: 72). '[V]isible, familiar, and generally predictable', these saintly images insist on clothing, feeding and processing (Watanabe 1992: 75). Their relationship with the community presupposes reciprocal obligations, the saint to be tended, the community to be protected and sanctified by its presence (Watanabe 1992: 75). Visits of saintly images to other towns may 'preserve an earlier ritual of obeisance by patron deities, through which regional hierarchical integration was expressed and reinforced' (Farriss 1984: 152).

We do not suggest that the 'cult of saints' is more Maya than folk Catholic, yet the ritual behaviour connected with them does have indigenous parallels. Often, modern Maya saints indulge in amorous intrigue and possess vengeful, bilious temperaments. Their human petitioners demand more than revere (Madsen 1967: 381), and Maya have been known to suspend uncooperative saints upside down in trees (John Monaghan pers. comm.). Some of these surprising actions occur in medieval Europe as well (Alan Kolata pers. comm.). But, at the least, we can see in the profound social bonds between saints and particular communities a reflection of the tutelary gods, some tangible as carved images, that existed throughout ancient Mesoamerica. In Postclassic Mexico, these images supplied leaders with political authority as 'god-bearers' or 'god-protector lords' (see Schroeder 1991: 122-3, 142, 172-3). Embodiments of the community, images were the targets of raids by antagonistic groups, just as saintly images are sometimes stolen or spirited away today (Nicholson 1971b: 409). In all likelihood, Classic Maya architecture of the elite cannot be understand solely in terms of mortuary or residential architecture - the prevailing mode of interpretation - but as places where these images were housed in the splendour and cosseted privacy due them (Houston in press).

Divinity and rulership, authority and belief

This article began with a list of devices to link rulers and divinity, all of which are present among the Classic Maya. We have evidence of:

1 declarations that living lords were god-like, alongside stronger claims for the divinity of deceased rulers in the ascending generation;

2 royal epithets representing rhetorical, analogical claims to divinity;

3 narratives or iconography likening royal behaviour to the exemplary behaviour of gods;

4 intermittent possession by godly force through costuming and behaviour appropriate to certain deities;

5 summons of gods to witness, presumably with favour, the rituals of living rulers; and

6 the possession and tending of god images which, most sources indicate, served as receptacles of divine force.

This last, involving the custodianship of images, argues powerfully for the assertion of legitimate rule through royal possession of 'ethnic' or 'tribal' gods that prefigure the saints' cults of the historic Maya region. The strategies employed by Classic Maya rulers resemble general patterns documented for ancient Egypt, where the king 'is marginal to the world of the gods, yet through him they rely on this world and on human efforts to sustain them and the cosmos' (Baines 1995: 11).

Armed with several recent decipherments, we can return to the apparent paradox that Maya kings were at once 'gods' and men. Through their close interactions with deities, extending to the point of apparently exerting control over them through acts of conjuring, kings and high nobles could claim an obvious qualitative difference from those of lower social and political standing. The ritual acts that displayed and presented such roles were certainly important in defining the category of ahaw, 'lord', yet they were perhaps not the exclusive domain of high rulers. What set rulers apart, then, defining them as true divine rulers? We have suggested that a more 'quantitative' distinction may be crucial, specifically with regard to the relative strength or heat of one's life essence or soul, widely known as ch'ulel (literally 'holiness'). The widespread title of high kings, k'ul ahaw or 'Holy Lord', may intimate that kings, through their close connexions to the spiritual realm, were the 'hottest' of all living people.

A compelling theological framework for understanding the Classic Maya idea of divinity, especially as resident in masks, clothing and effigies, comes from near-by central Mexico. General comparisons, once common in Mesoamerican studies and evident in this paper, are unfashionable; research tends now to focus on comparisons within sub-regions and language groups. That approach is defensible analytically in that scholars preserve control over context and historical setting. But there is also a cost: broader patterns become less clear, and students lose sight of the essential fact that Mesoamerica remains a region of intense interaction and homologous, tandem development. In our view, the 'Mesoamericanist' and 'sub-regionalist' approaches are neither wrong nor right. Both remain essential to a balanced perspective on Mesoamerican antiquity, and the tension between the two defines a crucial zone of research.

We have also touched on the role of divinity and kingship in even broader terms. In authorizing royal action, appeals to divinity appear to come 'from a realm beyond history, society, politics, beyond the terrain in which interested and situated actors struggle over scarce resources' (Lincoln 1994: 112). This transcendent quality of divine or divine-centred rule is fundamental, for it contravenes 'massive challenges' to a king's decisions in virtue of his linkage to godly force (Bendix 1978: 17). Yet saying that one rules with divine authority is not the same thing as exercising absolute rule. Far from it: '[w]herever authority rests on religion, any slip in authority can be interpreted as a withdrawal of divine favor, and the charter of religious failing can be strategically used in a campaign of de-authorization' (Lincoln 1994: 207). Supernatural mandates, laden with possibilities, are also fraught with risk; the prudent lord favours routinized ritual over unpredictable and falsifiable divination (David Webster pers. comm.). The same point was made long ago in Weber's (1978: 1114-15) study of charismatic domination and legitimacy. Once charisma and its supernatural supports were made routine and institutionalized, 'charisma and charismatic blessing [could be transformed] from a unique transitory gift of grace of extraordinary times and persons into a permanent possession of everyday life' (Weber 1978: 1121). Whether such efforts are successful is another matter.

For scholars, a difficult issue is the nature of the strategies documented for the Classic Maya. Were these held by the entire society, and how did they shift through time? We have described a system of legitimation predicated on dynastic assertions of divinity and monopolistic attempts to control divine mediation. These efforts may have met with variable success, and almost certainly shifted subtly as elites began to play a more ostentatious role in sculpture and writing towards the middle of the Late Classic period (Fash 1991: 160-61; Houston 1993: [ILLUSTRATION FOR FIGURE 5-4 OMITTED]). Following Giddens (1984: 29), we see royal 'power' not as an abstraction, to be linked inflexibly to certain titles, or as a set of static propositions, but as the tangible consequence of interaction between flesh-and-blood actors. Power derives from social and political discourse involving assertion, on the one hand, and acceptance or rejection by persons for whom that message is intended, on the other. As with any institution, divine kingship involves, as Baines (1995: 6) contends for ancient Egypt, human mortals fulfilling divine roles that have to be 'continually renegotiated and redefined'. The system of beliefs about Maya kings studied here is only one part of that equation. Whether it was widely held, whether it was believed firmly by the larger population, is another. A system of rule does not exist in the abstract, divorced from people. To operate, that system must have people who believe in its validity on an unreflective or subjective level (Weber 1978: 33).

The views expressed in Classic Maya art and inscriptions were royal views of the universe and its divine orderings, yet this is not to say that such ideas were simply imposed upon the ancient societies at large by a controlling elite. The presentation of royal power, divine authority and the convergence between the two certainly can be interpreted (sometimes too simply) as forms of political propaganda (Marcus 1992), but the underlying notions of divinity and power likely derived from more ancient, pervasive concepts that explained and personified the natural world (Freidel et al. 1993: 58). Notions of rulership and divinity described here coalesced within an idiom of an animating, godly force and represented its compelling extension into the realm of political authority. The grafting of ever-changing ideas about political power on to more broadly held concepts about the nature of the universe probably made those notions more compelling to royal subjects. Interestingly, some ancient notions of divine authority and its manifestations may have survived to modern times, despite even the removal of native elites at the conquest. The persistence of ritual presentations, terminologies and even some royal titles suggests ideas that continue over long periods. But the emphasis on continuity, on the basic premisses that underlie Maya existence, past and present, risks neglecting important patterns of local variation. In this essay, we have identified elements, strategies and probable objectives of divine kingship among the Classic Maya, and stressed their relation to a complex, localized theology. Still unwritten, but urgently needed, is a study of how rulership changed through time - this it surely did - and how the rhetoric of rulership and divinity, our focus in this article, corresponded to the political realities of the Classic period.

Acknowledgements. This paper has several ancestors, including a version presented by Houston at the University of California, Riverside, and another presented by him at Michael D. Coe's retirement symposium, sponsored by Yale University. John Clark, Cecilia Klein, John Monaghan, Bridget Hodder Stuart and Karl Taube helped throughout with good advice, moral support, and references to relevant literature, supplied also by Peter Heather. Most of the research here benefitted from support provided by the National Endowment for the Humanities through a Collaborative Projects grant awarded to the authors through Brigham Young, Yale and Harvard Universities, and from funds generously awarded by Dean Clayne Pope of Brigham Young. The NEH, an independent Federal Agency, takes no responsibility for our comments, nor do other friends who have contributed ideas and bibliography to this manuscript: Jose Miguel Garcia Campillo, Tom Cummins, Miguel Civil, Nick Dunning, Bill Hanks, John Hawkins, Alan Kolata, John Robertson, Evon Vogt and David Webster. Kerr numbers correspond to roll-out images in the Justin Kerr archive that he has most generously shared with us. Superb drawings by Ian Graham and his colleagues also illustrate this article.

1 For glyphic notation we use a system advocated by George Stuart for his 'Research Reports' series: bold indicates literal glyph transcription, italic the probable rendering in Classic Mayan.

2 Taube (1992a) provides an excellent discussion of specific Maya deities, and Bottero (1992: 211) gives a comparable emphasis on theo-anthropomorphization in Mesopotamia. Guthrie (1993) offers a broad discussion of anthropomorphization in all religious thought.

3 In our opinion, Freidel & Schele's (1988: 348, 363) discussion of Classic Maya rulers as 'conduit(s) of supernatural power and direct divine inspiration' goes too far in connecting divine or supernatural power with the ahaw ('lord') title. Apparently restricted to the royal family, the title does not in itself connote divinity, but may rather supply the Mayan equivalent of the central Mexican term, tlahtoani, 'speaker': note proto-Cholan *aw, 'shout' (Kaufman & Norman 1984: 116), resulting possibly in *aj-aw, 'he of the shout', 'shouter'. The rhetorical connection with 'Big Men' is obvious. Moreover, Freidel contends that the title of 'divine lord' (in fact, 'holy [place] lord') came into being as an 'institution' and 'definition of central power' by AD 199 (Freidel 1992: 119). We believe that common use of this title took place far later, c. AD 500.

4 Stela 6, a monument from Caracol, Belize, may record that a deceased ruler witnessed or 'saw' (yi-IL-a-hi, y-il-ah-i, 'he was seeing it') a ritual performed by his successor (Beetz & Satterthwaite 1981: [ILLUSTRATION FOR FIGURE 7B OMITTED], glyph B20). This would seem to represent a textual description of the 'floating ancestor' motif.

5 The banner-like sign is identical to the motif Michael Coe (1978: 106) first identified as the 'number tree' or 'computer print-out' in Maya art. The element often extrudes from beneath the arms of scribal gods (Reents-Budet 1994: [ILLUSTRATION FOR FIGURE 2:27 OMITTED]), and one sculpture of Early Classic date shows a youthful deity writing on similar vegetation (Berjonneau et al. 1985: plate 364). We suspect that the Maya thus depicted a perishable medium for more casual script, rather like the palm-leaf employed in India and southeast Asia (Gaur 1992: 40, 50-51). The fact that only numbers occur on such vegetation suggests its typical content: rapid accountings unaccompanied by explanatory, linguistic glosses. Theoretically, these notations underline an important point made by Piotr Michalowski (n.d.: 11, 14) for Mesopotamian script, namely, that we are dealing not with one, unitary notational system, but with many, each potentially of different origin and developmental trajectory.

6 The term -ichnal (spelled yi-chi-NAL) perhaps appears in modern Yucatec Mayan as -iknal, an inalienably possessed noun with two possible meanings: 'home or habitual place' or the perceptual 'inner space [that] can be encompassed in a single visual field and is in practical reach of any adult within it' (Hanks 1989: 91-2). Linguistically, the term refers to the corporeal field of one person and as such suggests a more precise understanding of the related glyphic expression: the gods appear as witnesses and not, properly speaking, as direct participants in ritual. When the y-ichnal expression involves two human beings, the second name corresponds to someone of higher status who sponsors the event.

7 One of the gods at La Mar, Bolon Yokte K'u, plays a role in many texts, but the most enigmatic completes the inscription of Monument 6 from Tortuguero, Mexico. Here is recorded a calendrical event in the early 21st century AD, at which time, apparently, the god may 'descend' ye-ma, y-emal (there are some technical problems with this translation). The reference is notable for its uniqueness. Prophecy forms an important body of colonial Maya literature but is poorly represented in Classic Maya texts, where future statements relate almost exclusively to impersonal temporal events that are safely predictable (e.g., the 13 baktuns will be finished at 13.0.0.0.0 in the Maya Long Count).

References

BAINES, J. 1995. Kingship, definition of culture, and legitimation, in O'Connor & Silverman (1995b): 3-47. Leiden: E.J. Brill.

BARTH, F. 1993. Balinese worlds. Chicago (IL): University of Chicago Press.

BARTHEL, T.S. 1952. Die Morgensternkult in den Darstellungen der Dresdener Mayahandschrift, Ethnos 17: 73-112.

BEALS, R.L. 1945. Ethnology of the western Mixe. Berkeley (CA): University of California Press. University of California Publications in American Archaeology and Ethnology 42(1).

BEARD, M. 1990. Priesthood in the Roman republic, in Beard & North (ed.): 19-48.

BEARD, M. & J. NORTH (ed.). 1990. Pagan priests: religion and power in the ancient world. London: Duckworth.

BEETZ, C.P. & L. SATTERTHWAITE. 1981. The monuments and inscriptions of Coracol, Belize. Philadelphia (PA): University Museum, University of Pennsylvania. University Museum Monograph 45.

BENDIX, R. 1978. Kings or people: power and the mandate to rule. Berkeley (CA): University of California Press.

BERJONNEAU, G., E. DELETAILLE & J.-L. SONNERY. 1985. Rediscovered masterpieces of Mesoamerica: Mexico-Guatemala-Honduras. Boulogne: Editions Arts.

BERLIN, H. 1963. The Palenque triad, Journal de la Societe des Americanistes 59: 107-35.

BLOCH, M. 1987. The ritual of the royal bath in Madagascar: the dissolution of death, birth and fertility into authority, in Cannadine & Price (ed.): 271-97.

BOONE, E.H. 1989. Incarnations of the supernatural: the image of Huitzilopochtli in Mexico and Europe. Philadelphia (PA): American Philosophical Society.

BOTTERO, J. 1992. Mesopotamia: writing, reasoning, and the gods. Chicago (IL): University of Chicago Press.

BURGHART, R. 1987. Gifts to the gods: power, property, and ceremonial in Nepal, in Cannadine & Price (ed.): 237-70.

BURKHART, L.M. 1988. The slippery earth: Nahua-Christian moral dialogue in 16th-century Mexico. Tucson (AZ): University of Arizona Press.

CANNADINE, D. 1987. Introduction: divine rites of kings, in Cannadine & Price (ed.): 1-19.

CANNADINE, D. & S. PRICE (ed.), 1987. Rituals of royalty: power and ceremonial in traditional societies. Cambridge: Cambridge University Press.

CARRASCO, P. 1950. Los Otomies: cultura e historia prehispanicas de los pueblos mesoamericanos de habla otomiana: Mexico: Instituto de Historia, Universidad Nacional Autonoma de Mexico.

COE, M.D. 1973. The Maya scribe and his world. New York (NY): Grolier Club.

1978. Lords of the underworld: masterpieces of Classic Maya ceramics. Princeton (NJ): Art Museum, Princeton University.

COE, W.R. 1967. Tikal: a handbook of the ancient Maya ruins. Philadelphia (PA): University Museum, University of Pennsylvania.

1990. Excavations in the Great Plaza, North Terrace and North Acropolis of Tikal. Philadelphia (PA): University Museum, University of Pennsylvania. Tikal Report 14, Volume V. University Museum Monograph 61.

FARRISS, N.M. 1984. Maya society under colonial rule: the collective enterprise of survival. Princeton (NJ): Princeton University Press.

FASH, W.L. 1991. Scribes, warriors and kings: the city of Copan and the ancient Maya. London: Thames & Hudson.

FERRIE, H. 1995. A conversation with K.C. Chang, Current Anthropology 36: 307-25.

FOUGHT, J. 1986. Cholti Maya: a sketch, in M.S. Edmonson (ed.), Supplement to the handbook of Middle American Indians 2: Linguistics: 43-55. Austin (TX): University of Texas Press.

FREIDEL, D.A. 1992. The trees of life: ahau as idea and artifact in Classic Lowland Maya civilization, in A.A. Demarest & G.W. Conrad (ed.), Ideology and Pre-Columbian civilizations: 115-33. Santa Fe (NM): School of American Research Press.

FREIDEL, D.A. & L. SCHELE. 1988. Kingship in the late Preclassic Maya Lowlands: the instruments and places of ritual power, American Anthropologist 90(3): 547-67.

FREIDEL, D., L. SCHELE & J. PARKER. 1993. Maya cosmos: 3000 years on the shaman's path. New York (NY): William Morrow.

GARLAND, R. 1990. Priests and power in Classical Athens, in Beard & North (ed.): 75-91.

GAUR, A. 1992. A history of writing. New York (NY): Cross River Press.

GEERTZ, C. 1977. Centers, kings, and charisma: reflections on the symbolics of power, in J. Ben-David & T.N. Clark (ed.), Culture and its creators: essays in honor of Edward Shils: 150-71. Chicago (IL): University of Chicago Press.

1980. Negara: the theatre state in 19th-century Bali. Princeton (NJ): Princeton University Press.

GIDDENS, A. 1984. The constitution of society: outline of the theory of structuration. Berkeley (CA): University of California Press.

GOSSEN, G.H. 1972. Temporal and spatial equivalents in Chamula ritual symbolism, in W.A. Lessa & E.Z. Vogt (ed.), Reader in comparative religion: an anthropological approach: 135-49. New York (NY): Harper & Row.

1986. Mesoamerican ideas as a foundation for regional synthesis, in G.H. Gossen (ed.), Symbol and meaning beyond the closed community: essays in Mesoamerican ideas: 1-8. Albany (NY): Institute for Mesoamerican Studies.

GRAHAM, I. 1967. Archaeological explorations in El Peten, Guatemala. New Orleans (LA): Middle American Research Institute, Tulane University. Publication 33.

1978. Corpus of Maya hieroglyphic inscriptions 2(2): Naranjo, Chunhuitz, Xunantunich. Cambridge (MA): Peabody Museum of Archaeology and Ethnology, Harvard University.

1980. Corpus of Maya hieroglyphic inscriptions 2(3): Ixkun, Ucanal, Ixtutz, Naranjo. Cambridge (MA): Peabody Museum of Archaeology and Ethnology, Harvard University.

1982. Corpus of Maya hieroglyphic inscriptions 3(3): Yaxchilan. Cambridge (MA): Peabody Museum of Archaeology and Ethnology, Harvard University.

GRAHAM, I. & E. VON EUW. 1975. Corpus of Maya hieroglyphic inscriptions 2(1): Naranjo. Cambridge (MA): Peabody Museum of Archaeology and Ethnology, Harvard University.

1992. Corpus of Maya hieroglyphic inscriptions 4(3): Uxmal, Xcalumkin. Cambridge (MA): Peabody Museum of Archaeology and Ethnology, Harvard University.

GROVE, D.C. 1987. Comments on the site and its organization, in D.C. Grove (ed.), Ancient Chalcatzingo: 420-33. Austin (TX): University of Texas Press.

GRUBE N. & W. NAHM. 1994. A census of Xibalba: a complete inventory of way characters on Maya ceramics, in B. Kerr & J. Kerr (ed.), The Maya vase book: a corpus of rollout photographs of Maya vases 4. New York (NY): Kerr Associates.

GRUZINSKI, S. 1989. Man-gods in the Mexican highlands: Indian power and colonial society, 1520-1800. Stanford (CA): Stanford University Press.

GUITERAS-HOLMES, C. 1961. Perils of the soul: the world view of a Tzotzil Indian. Glencoe (NY): The Free Press.

HANKS, W.F. 1990. Referential practice: language and lived space among the Maya. Chicago (IL): University of Chicago Press.

HOCART, A.M. 1970. Kings and councillors: an essay in the comparative anatomy of human society. Chicago (IL): University of Chicago Press.

HOUSTON, S.D. 1989. Reading the past: Maya glyphs. London: British Museum Publications.

1993. Hieroglyphs and history at Dos Pilas: dynastic politics of the Classic Maya. Austin (TX): University of Texas Press.

1995. Killing death: mortuary beliefs of the Classic Maya. Paper presented at the symposium 'Death, Burial, and the Afterlife', San Diego Museum of Man, San Diego (CA).

In press. Symbolic sweatbaths of the Maya: architectural meaning in the Cross Group at Palenque, Mexico, Latin American Antiquity.

HOUSTON, S.D. & D.S. STUART. 1989. The Way glyph: evidence for 'co-essences' among the Classic Maya. Washington (DC): Center for Maya Research. Research Reports on Ancient Maya Writing 30.

HVIDTFELDT, A. 1958. Teotl and Ixiptlatli: some central conceptions in ancient Mexican religion. Copenhagen: Andreassen.

JONES, C. & L. SATTERTHWAITE. 1982. The monuments and inscriptions of Tikal: the carved monuments. Philadelphia (PA): University Museum, University of Pennsylvania. University Museum Monograph 44.

KANTOROWlCZ, E.H. 1957. The king's two bodies: a study in mediaeval political theology. Princeton (NJ): Princeton University Press.

KAUFMAN, T.S. & W.M. NORMAN. 1984. An outline of proto-Cholan phonology, morphology, and vocabulary, in J.S. Justeson & L. Campbell (ed.), Phoneticism in Mayan hieroglyphic writing: 77-166. Albany (NY): Institute for Mesoamerican Studies, State University of New York at Albany. Publication 9.

KEIGHTLEY, D.N. 1978. The religious commitment: Shang theology and the genesis of Chinese political culture, History of Religions 17: 211-25.

KELLEY, D.H. 1965. The birth of the gods at Palenque, Estudios de Cultura Maya 5: 93-134.

KEMP, B. 1989. Ancient Egypt: anatomy of a civilization. London: Routledge.

KERTZER, D. 1988. Ritual, politics, and power. New Haven (CT): Yale University Press.

KLEIN, C.F. 1986. Masking empire: the material effects of masks in Aztec Mexico, Art History 9(2):135-67.

KUBLER, G. 1969. Studies in Classic Maya iconography. New Haven (CT): Connecticut Academy of Arts and Sciences. Memoirs 18.

LIEBESCHUETZ, J.H.W.G. 1979. Continuity and change in Roman religion. Oxford: Clarendon Press.

LINCOLN, B. 1994. Authority: construction and corrosion. Chicago (IL): University of Chicago Press.

LOCKHART, J. 1992. The Nahuas after the conquest: a social and cultural history of the Indians of central Mexico, 16th through 18th centuries. Stanford (CA): Stanford University Press.

LOPEZ AUSTIN, A. 1993. The myths of the Opossum: pathways of Mesoamerican mythology. Albuquerque (NM): University of New Mexico Press.

LOUNSBURY, F.G. 1980. Some problems in the interpretation of the mythological portion of the hieroglyphic text of the Temple of the Cross at Palenque, in M.G. Robertson (ed.), Third Palenque Round Table, 1978, Part 2: 99-115. Austin (TX): University of Texas Press.

MCANANY, P. 1995. Ancestors and the Classic Maya built environment. Paper presented at the Dumbarton Oaks Fall Symposium, 'Function and Meaning in Classic Maya Architecture'.

MCGEE, R.J. 1990. Life, ritual, and religion among the Lacandon Maya. Belmont (CA): Wadsworth.

MACMULLEN, R. 1981. Paganism in the Roman empire. New Haven (CT): Yale University Press.

MACRI, M. 1988. A descriptive grammar of Palenque Mayan. Ph.D dissertation, University of California, Berkeley (CA).

MADSEN, W. 1960. The Virgin's children: life in an Aztec village today. Austin (TX): University of Texas Press.

1967. Religious syncretism, in M. Nash (ed.), Handbook of Middle American Indians 6: 369-91. Austin (TX): University of Texas Press.

MARCUS, J. 1978. Archaeology and religion: a comparison of the Zapotec and Maya, World Archaeology 10(2): 172-91.

1983. Topic 97: Zapotec religion, in K.V. Flannery & J. Marcus (ed.), The Cloud People: divergent evolution of the Zapotec and Mixtec civilizations: 345-51. New York (NY): Academic Press.

1989. Zapotec chiefdoms and the nature of formative religions, in R.J. Sharer & D.C. Grove (ed.), Regional perspectives on the Olmec: 148-97. Cambridge: Cambridge University Press.

1992. Mesoamerican writing systems: propaganda, myth, and history in four ancient civilizations. Princeton (NJ): Princeton University Press.

MARCUS, J. & K. FLANNERY. 1994. Ancient Zapotec ritual and religion: an application of the direct historical approach, in C. Renfrew & E.B.W. Zubrow (ed.), The ancient mind: elements of cognitive archaeology: 55-74. Cambridge: Cambridge University Press.

MARKMAN, P.T. & R.H. MARKMAN. 1989. Masks of the spirit: image and metaphor in Mesoamerica. Berkeley (CA): University of California Press.

MARTIN, S.N.d. Tikal's 'star war' against Naranjo. Unpublished paper, London.

MATHEWS, P. 1991. Classic Maya emblem glyphs, in T.P. Culbert (ed.), Classic Maya political history: hieroglyphic and archaeological evidence: 19-29. Cambridge: Cambridge University Press.

MICHALOWSKI, P.N.d. On the early toponymy of Sumer: a contribution to the study of early Mesopotamian writing. Unpublished paper, University of Michigan, Ann Arbor (MI).

MOERTONO, S. 1968. State and statecraft in old Java: a study of the later Mataram period, 16th to 19th century. Ithaca (NY): Southeast Asia Program, Department of Asian Studies, Cornell University.

NICHOLSON, H.B. 1971a. Pre-hispanic central Mexican historiography, in Investigaciones contemporaneas sobre la historia de Mexico: memorias de la tercera reunion de historiadores Mexicanos y Norteamericanos (1969): 3881. Mexico (DF): Universidad Nacional Autonoma de Mexico.

1971b. Religion in pre-hispanic central Mexico, in R. Wauchope (ed.), Handbook of Middle American Indians 10(1): 395-446. Austin (TX): University of Texas Press.

O'CONNOR, D. & D.P. SILVERMAN. 1995a. Introduction, in O'Connor & Silverman (ed): xvii-xxvii.

(Ed.). 1995b. Ancient Egyptian kingship. Leiden: E.J. Brill.

PEMBERTON, J. 1994. On the subject of 'Java'. Ithaca (NY): Cornell University Press.

PRICE, S. 1987. From noble funerals to divine cult: the consecration of Roman emperors, in Cannadine & Price (ed.): 56-105.

PROSKOURIAKOFF, T. 1965. Sculpture and major arts of the Classic Lowlands, in R. Wauchope (ed.), Handbook of Middle American Indians 2: 469-97. Austin (TX): University of Texas Press.

1978. Olmec gods and Maya god-glyphs, in M. Giardino, B. Edmonson & W. Creamer (ed.), Codex Wauchope: 11317. New Orleans (LA): Tulane University.

REENTS-BUDET, D. 1994. Painting the Maya universe: royal ceramics of the Classic period. Durham (NC): Duke University Press.

RINGLE, W.M. 1988. Of mice and monkeys: the value and meaning of T1016, the God C hieroglyph. Washington (DC): Center for Maya Research. Research Reports on Ancient Maya Writing 18.

SAHLINS, M. 1981. The stranger-king, or Dumezil among the Fijians, Journal of Pacific History 16: 107-32.

SCHELE, L. 1982. Maya glyphs: the verbs. Austin (TX): University of Texas Press.

1992. The founders of lineages at Copan and other Maya sites, Ancient Mesoamerica 3: 135-44.

SCHELE, L. & D. FREIDEL. 1990. A forest of kings: the untold story of the ancient Maya. New York (NY): Morrow.

SCHELE, L. & P. MATHEWS. 1979. The Bodega of Palenque, Chiapas, Mexico. Washington (DC): Dumbarton Oaks.

SCHELLHAS, P. 1904. Representation of deities of the Maya manuscripts. Cambridge (MA): Harvard University. Papers of the Peabody Museum of American Archaeology and Ethnology 4(1).

SCHROEDER, S. 1991. Chimalpahin and the kingdoms of Chalco. Tucson (AZ): University of Arizona Press.

SELER, E. 1898. Quetzalcouatl-Kukulcan in Yucatan. Zeitschrift fur Ethnologie 30: 377-410.

SJOBERG, A. 1957-71. Gotterreisen, in Reallexikon der Assyriologie und Vorderasiatischen Archaologie 3: 480-83. Berlin: Walter de Gruyter.

SPORES, R. 1967. The Mixtec kings and their people. Norman (OK): University of Oklahoma Press.

1984. The Mixtecs in ancient and colonial times. Norman (OK): University of Oklahoma Press.

STONE, A. 1989. Disconnection, foreign insignia, and political expansion: Teotihuacan and the warrior stelae of Piedras Negras, in R.A. Diehl & J.C. Berlo (ed.), Mesoamerica after the decline of Teotihuacan, AD 700-900:153-72. Washington (DC): Dumbarton Oaks.

STUART, D.S. 1984. Royal auto-sacrifice among the Maya: a study in image and meaning, RES 7/8: 6-20.

1993. Historical inscriptions and the Maya collapse, in J.A. Sabloff & J.S. Henderson (ed.), Lowland Maya civilization in the 8th century AD: 321-54. Washington (DC): Dumbarton Oaks.

STUART, D.S. & S.D. HOUSTON. 1994. Classic Maya place names. Washington (DC): Dumbarton Oaks. Studies in Pre-Columbian Art & Archaeology 33.

STUART, G.E. & G.S. STUART. 1977. The mysterious Maya. Washington (DC): National Geographic Society.

TAUBE, K.A. 1985. The Classic Maya maize god: a reappraisal, in M.G. Robertson (ed.), Fifth Palenque Round Table, 1983: 171-81. San Francisco (CA): Pre-Columbian Art Research Institute.

1992a. The major gods of ancient Yucatan. Washington (DC): Dumbarton Oaks. Studies in Pre-Columbian Art & Archaeology 32.

1992b. The Temple of Quetzalcoatl and the cult of sacred war at Teotihuacan, RES 21: 53-87.

THOMPSON, J.E.S. 1970. Maya history and religion. Norman (OK): University of Oklahoma Press.

TOWNSEND, R.F. 1979. State and cosmos in the art of Tenochtitlan. Washington (DC): Dumbarton Oaks. Studies in Pre-Columbian Art & Archaeology 20.

1992. The Aztecs. London: Thames & Hudson.

TOZZER, A.M. 1941. Landa's Relacion de las cosas de Yucatan. Cambridge (MA): Harvard University. Papers of the Peabody Museum of American Archaeology and Ethnology 18.

VOGT, E.Z. 1969. Zinacantan: a Maya community in the highlands of Chiapas. Cambridge (MA): Belknap Press, Harvard University Press.

1985. Cardinal directions and ceremonial circuits in Mayan and Southwestern cosmology, National Geographic Research Reports 21: 487-96.

1993. Tortillas for the gods: a symbolic analysis of Zinacanteco rituals. Norman (OK): University of Oklahoma Press.

WATANABE, J.M. 1992. Maya saints and souls in a changing world. Austin (TX): University of Texas Press.

WEBER, M. 1978. Economy and society: an outline of interpretive sociology. Berkeley (CA): University of California Press.