Society for American Archaeology

Ecology and Ritual: Water Management and the Maya Author(s): Vernon L. Scarborough Source: Latin American Antiquity, Vol. 9, No. 2 (Jun., 1998), pp. 135-159 Published by: Society for American Archaeology Stable URL: http://www.jstor.org/stable/971991 . Accessed: 10/05/2011 21:13

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=sam. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to Latin American Antiquity.

http://www.jstor.org ECOLOGYAND RITUAL:WATER MANAGEMENT AND THE MAYA

VernonL. Scarborough

How the ancient Maya of the central YucatecanLowlands managed their water and land resources remains poorly knownX although crucial to an understandingof ancient political economy. Recent archival research and field data suggest the wide- spread use of artificially altered, natural depressionsfor the collection and containmentof water, bothfor potable consump- tion and agricultural ends. During the Classic period (A.D. 250-900) several of the principal cities in the Maya area constructed their largest architecture and monumentsat the summit of hills and ridges. Associated with these elevated cen- ters-"water mountains"-were sizable, life-sustaining reserroirs quarried into their summits. The effect of this town-plan- ning design was the centralization of a primary and fundamental resource. Although elite managers controlled the water source, other decentralizingforces prevented anything similar to Wittfogel's"total power." However, by ritually appropriat- ing the everyday and mundane activities associated with water by the sustaining population, elites used high-performance water ritual as manifest in the iconography to further centralize control. The significance of modifying the urban landscape in the partial image of the ordinary water hole defines the extraordinaryin Maya ritual.

El tema de como los mayas antiguos de las tierras bajas del centro de Yucataonadministraron sus recursos de agua y tierra permanece raqufticamenteexplicado, aunque esto se considera muy importantepara llegar a entender la economfa polftica antigua. Recientes investigaciones de los archivos e informaciondel campo insinuan el extenso uso de las depresiones natu- rales modificadas artificialmente,para la recoleccion, y el represamientodel agua, para los consumos domeosticosy agrfcola. Durante el perfodo Cla'sico(250-900 d. C.) varias de las ciudades principales en el aoreaMaya construyeronla arquitectura mayor y los monumentos en la cima de los y lomas. En asociacion con estos centros elevados-"las montanas de agua"-habfa grandes depositos de agua, las reservas de sostenimiento de vida excavados en sus cimas. El efecto de este diseno de planificacion de asentamientos fue la centralizacion de un recurso principal y fundamental. Aunque los admi- nistradores de la eolitecontrolaban la fuente del aguaXotras fuerzas descentralizadoras prevenlan cualquier cosa que fuera similar al "poder totalizador" de Wittfogel.Sin embargo, al apropiarse ritualmentede las actividades cotidianas y mundanas asociadas con el agua por parte de la populacion a la que mantenfan, los administradoresde la eoliteusaron ritos acuaticos sofisticados, como se manifiesta en la iconograffa para controlar centralmente matsy mats.El sign ificado de la modificacion del paisaje urbano a travetsde la imagen del pozo ordinario de agua define lo extraordinariodel ritual maya.

When processual archaeologists waive their right to The MayaLowlands of the greaterYucatan include cosmology and ideology in their reconstruc- Peninsulawere a difficultenvironment in tions, one gets the kind of dichotomy which we have seen in Mesoamerica: anthropologists writing about whichto makea living. The semitropical settlement and subsistence, while humanists write landscapewas definedby thinkarstic soils, sea- about religion and cosmology. And the humanists, for sonaland erratic rainfall, and a markedabsence of the most part, do not have the ecological and evolu- permanentnatural water sources. In keepingwith tionary perspective of the anthropological archaeolo- a tropicalecosystem speciesdiversity was abun- gists [Marcus and Flannery 1994:55]. dant but concentrationsof any single speciesat Shamanism is his medium; he can see "faraway"by any one location were minor preventingthe going into a trance and looking into a mirroror a con- largenatural harvests associated with moretem- tainer with water [Covarrubias1937:349]. perateenvironments. Nevertheless, one of the ear-

Vernon L. Scarborough * Departmentof Anthropology,University of Cincinnati,Cincinnati, OH 45221

LatinAmencan Antiquity,9(2), 1998, pp. 135-159. Copyright C)by the Society for AmericanArchaeology

135 LATIN 136 AMERICANANTIQUITY [Vol. 9, No. 2, 1998

Figure 1. Map of Maya area with boundaries for the Southern Lowlands. liest tropical/semitropicalstates known evolved in conspicuous consumption of luxury goods, the this setting.The apogee of political and economic conspicuousmanipulation of monumentalarchi- development occurred during the Classic period tecture, and charismatic rulers. But the funda- (A.D. 25s900) in a Maya heartlandextending mental need for water and the food that it over , central and northernGuatemala, and allows preparingand maintainingthe earth for adjacent portions of and the plantingor the daily requirementsfor drinking southernMaya Lowlands (Figure 1). Given the water made water management in a fragile, difficultiesin accessing and centralizingresources water-stressed environment another powerful fromthe semitropicalsetting, how did the ancient Organlzlng:. . -orce.. Mayaorganize a labor force to harvest the land- The construction and maintenance of water scape? One window into their political econ- systems in the towns and cities of the ancient omy how a society employs power relationships Mayaconcentrated water in a quantityand qual- between groupsto organizethe use of resources- ityunavailable naturally. By placing water and its isthe management of land and water. managementapparatus in the center of their ele- As a critical and scarce resource during the vatedClassic-period communities, the Maya per- lengthydry season, water was politically manipu- mitted a controlling elite to manipulate the latedby a Maya elite to centralize and control resource.Nevertheless, the dispersed settlement powerduring the Classic period. Clearly, other andland-use design of the ancient Maya required meanswere used to marshall authority in the theelite frequently to draw the attention and MayaLowlands: overt warfare, acquisition and activitiesof the sustaining population into the Scarborough] WATERMANAGEMENT AND THE MAYA 137 public centers, in part by incorporatinga wide Ecological Possibilism and variety of ritual. In spite of the complexity of Hydrological Accomplishment Classic-period cities that functioned like "water mountains" gravity-flow, reservoir systems Water and land use during the Classic period of positioned at the summit of naturalhillocks and the Maya was a complex and sophisticatedset of ridges a support population could opt for systems and techniques adaptedto the semitropi- another environmental and political situation, cal microclimatic rhythms and topographiccon- given the weakly developed, centralizingforces at straints of the karstic Upper Central American work in a Maya center.Elite managersare posited lowland setting. Because the YucatanPeninsular to have symbolically appropriatedthe everyday environmentis characterizedby markedseasonal tasks of a sustaining population associated with scarcity of permanent water sources be they water use by promotingwater-related activities in springs, open lakes, or perennial streams popu- high ritual performance.By locating tanks near lations invested in creative landscaping adapta- the largest civic architecture structuresused in tions to collect and store surfacerunoff duringthe public theater the Maya elite employed ritual six to eight months of abundantannual precipita- acts derived from water use and availability. tion. As I have suggested elsewhere (Scarborough The paper begins with an extended introduc- 1993a, 1994), it was the limitationsof these envi- tion to the physical hydraulic systems in the rons that partially stimulatedthe accretionaryor southern Maya Lowlands. Reference is made to incrementalmodifications made to the landscape the centralizingeffects of the water-management by the Maya. design duringthe Classic period in spite of a dis- Much has been made of the dispersedland-use persed sustaining population. The principal case patternsof the ancient Maya. In spite of truly siz- study for the presentationis , representingin able populations demonstrable throughout the microcosm aspects of the many urban environ- lowlands-estimates comparable to the highest ments established throughoutthe southern low- known for preindustrialstates (Rice and Culbert lands. Next, the argument for the ritual 1990) the lack of densely concentratedpopula- appropriationof mundaneand everyday activities tions (and by extension, resources, as the argu- by an elite is articulatedusing ethnographicand ment goes) within their largest cities implies a ethnohistoric examples drawn principally from less complex social orderthan is apparentin other Bali, Madagascar,and highland Chiapas. Each primary state systems. V. Gordon Childe (1950) case study represents a tropical or semitropical set the tone for definitions of the state with his complex society employing water as a physical traitlist emphasizingurbanism. His greatestchal- and symbolic tool for organizingand maintaining lenge was what to do with the Maya. "Hence the social order. Where social hierarchicaldivisions minimum definition of a city, the greatest factor are pronounced, water is manipulated by elite common to the Old World and the New will be managers to secure and maintainpower through substantially reduced and impoverished by the high performanceritual. Finally, iconography is inclusion of the Maya" (p. 9). The association of examined to reveal the significance of water sys- dense urbancommunities of scale with thresholds tems and ritualacts by the controllingMaya elite. cresting 5,000 people/lun are frequently equated Tikal is again the principal context for this dis- with state levels of social complexity; societies cussion, which attemptsto link water imagery to with somethingless become suspect, less compli- the physical hydraulic features there, although cated the largest Maya centers being nearly a other iconographic sets from a greater Meso- full magnitude less dense (Sanders and Price american traditionare included. The practice of 1968; Sanders et al. 1979; cf. Whitehouse and elaborate ritual expression emphasizing the Wilkins 1986). importance of water systems by way of their What Childe and others have neglected are the functional and symbolic architecturalmanifesta- ecological constraintsinfluencing a tropical state. tion within the great centers-countered the nat- Because specific biological resources were never ural centrifugalforces pulling a dispersedsupport concentrated in extensive plots naturally, early populationaway (cf. Farriss 1978). settlement adaptationswere dispersed-designed LATINAMERICAN ANTIQUITY [Vol. 9, No. 2, 1998 138 to harvest best a wide variety of resources. nomic organization that pushed, invented, and Ethnographicand ethnohistoricdata indicate that reinvented their environmental parameters to before Spanish colonization of the region gar- accommodatea level of complexity differentthan dening and milpa productionwere not as depen- some other better understood social systems. dent on sizable maize surpluses (Atran 1993; Controlling water access and the manner by Nations and Nigh 1980; Schwartz 1990)- which land was developed to accommodate the although maize was always a staple. Polyculture use of water typifies aspects of Maya political included a number of tropical domesticates as economy throughtime.' well as several cultivarsthat were as healthy with Water and Land Development or without human propagation.Frequently crop- ping was done in the tangle of the jungle canopy, During the initial colonization of the Maya area some crops conserved and harvested best in a (> 1000 B.C.), the coastal plain and the few, managedforest setting (Marcus 1982; cf. Groube small, perennialdrainages leading to the sea were 1989:301). The significance of these cropping the only zones that could supportsedentary popu- strategiesillustrates the productivityof a tropical lations for the long term. Settled fisherfolk teth- landscape constrained by thin soils and erratic ered to coastal riverine resources probably rainfall.Early in Maya prehistorythe best adapta- experimented with domesticates initially better tions to the environment were those that mim- suited to highland environments (MacNeish icked it tricking the ecosystem into producing 1964; MacNeish and Nelken-Terner 1983; more for sedentary human populations. Larger Pulestonand Puleston 1971). Because of the envi- agriculturalplots supporting maize or comple- ronmentallimitations associated with the interior mentarybeans or perhapssquash functionedwell of the Yucatan Peninsula, colonization lagged in highlandMexico, but the constraintsof a semi- perhapshalf a millenniumbehind incipient seden- tropical setting were less conducive until the tism elsewhere in Mesoamerica (Andrews and Spaniardsarrived. Hammond 1990; Flannery and Marcus 1983; The organizingparameters for the Maya were Hammond 1991; MacNeish 1964; Niederberger put in place at an early date; by 600 B.C. sizable 1979; Sanderset al. 1979). With increasingpopu- investments in monumental architecture were lation, however, pioneer groups pushed inland made (Hansen 1991), and intensive forms of following the minor streams and associated swamp agriculture were initiated (Pohl et. al swamp margins adapting to the four-month-long 1996; Scarborough1993a). Because the environ- dry season by living near naturaldepressions that ment was always a constraint, labor skills and might hold water for an extended period. scheduling, coupled with the kinds and amounts Evidence for formalfood storagepits a char- of land and water available, were carefully acteristicassociated with long-termresidency at a blended to accommodatebest the forces and rela- site and a common feature in most early tions of production. As argued elsewhere Mesoamerican villages is seldom identified (Scarborough1993a, 1994), the environmentwas until the Late Preclassic period (400 B.C.-A.D. never a constant or fixed-in-form setting for the 250) in the Maya area. The lack of this kind of Maya; they accretionally altered it producing a storage for food may indicate a cropping strategy highly productive setting for humans. Maya eco- that allowed for adequateswamp-margin agricul- nomic organizationoperated within the ecologi- tural productionby the first interior villages on cal parameters of the semitropical setting, the peninsula. High humidity and temperatures preventingthe kind of resource concentrations- conspire to prevent simple storage adaptations, biological assets as well as some physical ones makingother economic adjustmentsnecessary for such as water found in more temperateor semi- long-term sedentism. One adaptationmay have aridenvironments where most otherarchaic states been to the elevated water table in swamp set- developed. The significance of Maya resource tings-even in the dry season that may have and settlement dispersion is not that it was less allowed two crops annually.When supplemented centralizedthan defined in semiaridstate systems, with natural forest foods and fibers, year-round but that the Maya developed a political and eco- residential occupation was possible given low Scarborough] WATERMANAGEMENT AND THE MAYA 139 populationdensities (Scarborough1994). articulatedby BarbaraFash at Copanusing icono- By the Late Preclassic period sizable popula- graphicreferences (Fash 1996), truewater control tions were established throughoutthe seasonally of scale is best manifest at the huge site of Tikal variable lowlands. Communities and sometimes (Scarboroughand Gallopin 1991). Togetherwith regions adapted to the microclimatic and topo- Copan (Fash 1991; Fash and Andrews V 1998), graphic variation apparentfrom one end of the Tikal continues to represent one of our best- YucatanPeninsula to the other.Years of trial and reportedlarge communitiesin the southernMaya errorin modifying the landscapeto raise its natu- Lowlands.2 rally limiting carrying capacity resulted in an What sprang from the cut-and-fill operations engineered environmentthat was extremely pro- in establishing the tiered building foundationsin ductive. Some sites and site areas constructed the core of the largest Classic centers were tower- their centers near the base of shallow natural ing pyramids, massive acropolises, and open depressions to take advantageof the slope runoff plaza pavements architecturallyknit together in generated into these settings, conserving the dramaticenterprises. Seen from kilometers away, water resourcein carefully alteredreservoirs next the greatest centers were beacons to thousands to the growing investments made in monumental occupying the densely settled rain forest. architecture. Sites as small as Cerros Although some of the fill necessary for the con- (Scarborough1983, l991a) and as large as struction of the standing architecturewas taken (Matheny 1976, 1978) or even from the initial leveling operationswithin a cen- (Dahlin 1984; Matheny 1986; Matheny et al. ter (Scarboroughet al. 1994), much more was 1980) positioned their centers to receive runoff obtainedby formal quarryingactivities. This was from a large, but essentially natural, watershed. a planned process in the constructionof a center, These communities used "concave microwater- with the location of quarryfill and the depression sheds" to direct the water and, by extension, the that resulted a deliberate act, possibly as impor- use of the land resource (Scarborough 1993a, tant to the built environment as the pyramids 1994) (Figure 2). Nevertheless, the Late themselves. It was the conversionof these depres- Preclassic adaptationcan be viewed as a passive sions into tanks and formal reservoirs that pro- form of water managementin that it representsa vided the water necessary for both the logical extension of the earliest sedentists' use of constructionof the center as well as its mainte- naturaldepressions for agriculturalends. nance. Naturalswales were sometimes mined and By the Classic period a significant water-man- widened for fill but closed off during the con- agement variation occurred at several sites struction process, with basal surface exposures (Figure 2). Throughout the southern lowlands, sealing naturallyby way of clay in runoff suspen- cities selected elevated ridges and hillocks to sion or by deliberateacts of plastering.From the position their most grand architecture.Through a onset of construction,water was at a premiumin process of moving centers away from immediate a large Maya site. Water was requiredin mortar and predictablewater sources and relocatingthem preparationfrom the cementing of veneer stone to to the summits of karstichills, the Maya chose to the paving and plasteringof floors and walls. The excavate deliberateand sizable reservoirsin prox- work gangs themselves needed potable supplies, imity to their elevated and massive architectural furthersuggesting that the first order of business works-especially pronounced at the enormous at a Maya constructionsite was the excavation of cities of Tikal and (Folan et al. 1996; a tank system. In most cases, the location of the Scarboroughand Gallopin 1991). By cutting and elevated tanks was in immediate proximity and filling the naturallyundulating and craggy sum- below the most grand constructionprojects at a mit surfaces, the foundationsfor a grand concen- site. tration of engineered landscaping occurred. Although well-defined reservoir systems are Tikal as a WaterMountain demonstrable at slightly smaller communities Tikal illustrates the archetypicalmanagement of such as La Milpa (Scarboroughet al. 1995) and water within a center (Figure 3). Located away (Scarborough et al. 1994), and recently from permanent natural sources of water, Tikal

Scarborough] WATERMANAGEMENT AND THE MAYA 141 was dependenton the collection of seasonal rain- to the swampy terrain surroundingTikal were fall. Precipitation rates today range from four sizable reservoirs positioned loosely in the 1,350-2,000 mm per year in this portionof north- cardinal directions. These "bajo-margin"reser- central Peten, , with a four-month voirs, located at the base of four of the six catch- period of annual drought (Scarborough and ment areas, were designed to captureand recycle Gallopin 1991). By the Late Classic period (A.D. "used"or "gray"water from the upslope residen- 600-900), Tikal consisted of extensive zones of tial settings into postulated swamp-marginagri- paved surfaces plazas, courtyards,platforms- cultural plots (Figures 4 and 5). On the scalar on which temples, pyramids,and palacelike struc- order of the central-precincttanks but located tures were built. The immediate impact of this away from the dense population aggregates far- architectureand its surfaces was to dramatizethe ther upslope bajo-marginreservoirs had a com- power of those responsible for the construction bined storage capacity of 50,00>175,000 m3, and maintenance of the center (Trigger 1990); runoff collected from a set of catchment areas however, these same pavementsprovided a delib- largerthan the centralprecinct itself. erate set of impervious catchment areas subtly Although we will likely never know the designed to seal the underlying,naturally porous amountof food made available to the residentsof limestone and direct rainy season runoff into the Tikal, we can posit the area watered by the bajo- sizable tanks within the central precinct of the margin tanks. Near present-day Tehuacan, site. Mexico, Wilken (1987: 159) indicates that a Six catchment areas or microwatersheddivi- canalized spring allows each memberof the com- sions were engineeredon the principalhillock on munity at Chilac at least one turn of water or 1.3 which Tikal rests (Figure4). Five catchmentareas x 106 liters (1 m3= 1,000 liters), enough water to occupied the densest portion of residentialspace, irrigate one hectare of green corn, garlic, and positioned on the gentle slope but originating at tomatoes. If we use the mean average storage or near the elevated central precinct. The most capacity for the tanks, then approximately85 ha completely altered catchmentwas at the summit could be agriculturallysupported at Tikal by the or central-precinctcatchment covering an area of bajo-margintanks a figure believed low given 62 ha and enclosing much of the monumental the elevated water table associated with the sea- architectureat Tikal. Withinthis catchmentalone, sonally inundated bajos, the lower evaporation more than 900,000 m3 of water could be poten- rates found in the Maya Lowlands, and the likeli- tially collected (based on 1,500 mm of annual hood that many of the crops grown by the ancient rainfall)by way of slightly canted pavementsand Maya were less water demandingthan those pro- subtle diversion weirs directing runoff into six duced today in the cash-cropeconomy of Chilac. central-precinct reservoirs. Where excavations With water available year-round, two or three were conducted in the Palace Reservoir (Figure crops could potentially be produced annually. 4), Harrison(1993:84) reports that the tank was Although the actual identificationof f1eldsbelow deliberately sealed by liner stones and imported the bajo-marginreservoirs remains conjecture, clays. The reservoirs had a combined storage field systems are recorded ethnographicallyand capacity of 100,000-250,000 m3 and allowed the archaeologically in similar settings elsewhere in planned release of water during the dry season the Maya Lowlands (HaxTisonand Turner 1978; through posited sluice gates located under the Puleston 1983; Wilk 1985). Such agriculturalfea- causeway system, especially well deE1nedon the tures are notoriously difficult to identify because easternmargins of the site (Figure 5). of the obvious naturalprocesses affecting season- Within the densely occupied residential zone ally inundatedswamps and the influence of allu- immediatelybelow the summitof the site, smaller viation and colluviation during the last thousand tanks for domestic use were apparent.Although years since Terminal Classic (A.D. 800-900) considerablysmaller in capacity,these reservoirs abandonment. were likely recharged through the scheduled Tikal representsour best-documentedexample release of water from the central-precincttank of a complete water-managementsystem in the system. Near the foot of the hillock and adjacent Maya area, though segments of the system out- bSm:^ a ; ' 0x:D' .t!' i SS8 od - C.\. . Z _

-

j

}R

ven

- : j 4 - 9 + - f r

wtySeSt f

z * l

&J #t- m :D

-

v OZ J s o:y 2 .

- \: :E

PS 9K i !

l

_ o -

D Figure3. The landscapeat Tikal. (Reprintedwith permissionfrom Research and Exploration 10:188*Copyright C) 1994 NationalGeographic Society) z o

-

D D Bajo Residential 1 l Catncthal prtecinct ...... [. .;;. Tikal catchment,

Scarborough] WATER MANAGEMENT AND THE MAYA 143

g;7/E/X3 Central precinct ,[^.*.* .1 Bejucal catchment .< . ..2 Residential catchment ^ Ruins

200- Contour reseNoir Contour inteNal I§ f [} Perdido catchment 25m s<\\\N Bajo-margin Floodgate [L 1S\\\\X] reseNoir Corrientalcatchment o ...... soo m

Figure 4. Main catchment areas at Tikal. (Reprinted with permission from Scierace 251:659. Copyright t 1991 American Association for the Advancement of Science) lined above are known at other centers a king or ruling elite. But the functional accessi- (Scarborough1993a). Unlike the Late Preclassic bility of the water source represents only one adaptationbased on a passive, concave microwa- aspect of the overall importanceof the water sys- tershed water system, Late Classic systems tem; that a set of elite managers was capable of entailed considerablygreater control of the water altering a limestone hillock or outcrop into a resource based on elevated site selection and an "water mountain," to produce a water source associated "convex microwatershed"adaptation where there was noneS also permitted the sym- (Figure 2). The significance of this control and bolic appropriationof this fundamentalyet mun- the centralizingforces associated with the manip- dane natural resource by the elite. It is the ulation of the summit central-precinctcatchment proximity of reservoirsto some of the largest and area suggest the economic and political power of most complicated architectureat Tikal-all cen- ", C-" ral proehbctrourtoUr _ Dlroethdraln^o o Intor^edltocontour

144 LATINAMERICAN ANTIQUITY lVol. 9, No. 2, 1998

s " ""&" > Rr°&"t§°ftl°° -- Indox contour

*&1_"rtin z lulo / r"rtoF O soo . . m Cor , sm

Figure 5. East margins of central Tikal. (Reprinted with permission from Science 251:660. Copyright O 1991 American Association for the Advancement of Science) tral-precinctreservoirs within 100 m of at least ciated with the largestcenters. Little to no rainfor one major civic-ceremonial building complex- two or three years-symptomatic of either the that helps link water to ritual. natural regional variation in precipitation or something more climate altering (Harrison1993; Scale, WaterAvailability, and Ideology Scarborough 1993b) would result in abandon- What made the paved, convex microwatershed ment of a household tank system, but not at a site system so effective was precisely its scale. Tank like Tikal. systems away from the great centers were likely Nevertheless, the technology associated with much more ubiquitous and dispersed than pre- reservoirmanagement at a less sophisticatedscale sent-day archaeologicalsurveys can identify in was known and practiced throughout the low- keeping with the naturalseparation of biological lands outside the immediate political sphere of resources in the semitropics.Small housemounds many large centers (Beach and Dunning 1997; are difficult enough to record; the few shallow Matheny and Gurr 1983). Tikal did have control tanks nearbyare even less visible after one to two over a scarce resource by way of the huge tanks millennia of abandonmentand siltation. But the located in its elevated core, and the immediate inability of the known small tanks to hold enough sustainingpopulation residing in proximity to the water through more than one dry season after a release channels was heavily invested in the gen- not uncommon extended period of drought pre- erationsof land modificationsassociated with this vented these household water holes from provid- engineered landscape. Although functionally ing the necessary risk reduction when compared sophisticated, the Classic water-control system to the truly huge catchmentsand reservoirsasso- was never an example of "total power" a la Scarborough] WATERMANAGEMENT AND THE MAYA 145

Wittfogel (1957); a support population had the between humans and their landscapes are funda- option to relocate, given a work force large and mental and that the daily activities and decisions coordinated enough to construct a smaller but of a group within this context strongly affect the sustainabletank system elsewhere. However, the kind and degree of ritual performance. Unlike social and economic costs of moving from an beliefs, symbols or myths, and the "conceptual accretionally altered landscape constructed over blueprints"for ritual, rituals themselves are the centuries (Scarborough1993a, 1994) would be a habitual actions necessary for publicly defining strong incentive to remain (cf. McAnany 1995), an ideology (Bell 1992:19). Rituals are the pre- since building a water system as elaborateas that scribedactivities that regularlyprovide aspects of at La Milpa or Kinal let alone Tikal would be meaning for religion or ideology within society. extremely expensive. Still, the option was there. Ritual is performed at all hierarchicallevels of Elites in the great centers recognized the fluid- society but frequently draws its influence from ity and fragility of lineage alliances, and not only the mundane, grounded in everyday experience within the ranks of the nobility (Haviland 1992; (Reynolds and Tanner1995:8). It is the everyday cf. Hendon 1991). The support population routines within a society that foster significant required periodic and substantial"proof" of the aspects of religious ritual, everyday tasks based ruling body's right and legitimacy to govern. on subsistence-level activities and dependent on Much of the literaturedevoted to Maya ideology the immediatedisposition of the landscape. focuses on the methods that the nobility The role of ritual among the ancient Maya is employed to capturethe minds and souls of their difficult to project without knowing more about subjects (Demarest 1992; Freidel et al. 1993). the structureof ritual as revealed by its influence Demonstratingdirect descent from a greatking or over other societies: societies at similar levels of set of deities or even a powerful lineage was most socioeconomic and sociopolitical complexity and importantto the ruler and the ruled; the role of those for which a rich ethnographicand ethnohis- blood sacrifice in establishingthat bond in public toric literatureexists. Because the Maya them- theater has received disproportionateattention selves did not record the kind of information when comparedto the more mundaneappropria- necessary for a complete understandingof their tion of economic resources by an elite (cf. political economy especially the role of ritualin Montejo 1991) the latter generally assumed attractingand maintaininga supportpopulation- necessary to control and order statecraft else- an examination of three well-reported groups where in the world (Adams 1992; Carneiro1992). occupying similarsemitropical to tropicalsettings However, anotherinterpretation of the iconog- to those of the Maya follows. Each group has a raphy from the great centers suggests that an well-developed set of water rituals, in part condi- attemptwas made to link the economic and eco- tioned by constraintsin access to water. logical underpinningsof the Maya to the elite Madagascar trappingsof control. Waterritual illustratessome of the ways that economic power was manipu- In an importantstudy treatingthe definitionof rit- lated by the elite in . It is the linkage ual and the mannerby which it allows and legiti- between the functionalnecessity of occupying an mates power, Bloch ( 1987) examines the engineered landscape, allowing households and late-nineteenth-centuryroyal bath ceremony of lineages to make a living, and the evolution of an the Merinain Madagascar.Here the influence and elite ideology to accommodateand subsequently authority of the grand royal ceremony were appropriatethat reality that directs the remainder derived from a similar, but less complex and of this presentation. involved, ritual performed widely at the local level. Waterblessings were splashed from father Ritual to child in each household at the beginning of the The ecological underpinningsof ritual are based New Year.All fathersin the kingdom acted as the on the ordinary,day-to-day actions and routines spiritualmedium in accessing and delivering the that humans carry out. They assume that the blessings of their immediate ancestors to their underlying relationships and energy exchanges children by way of ancestraltomb visitations the 146 LATINAMERICAN ANTIQUITY [Vol. 9, No. 2, 1998

day before. It is by elaboratingand publicly artic- Nevertheless, Bali is also the origin for another ulating this household ceremony within the state interpretationof complex society grounded in palace that the annualroyal bath of the king legit- economy and the environment. imized his ancestralright to govern and declared Much has been written with respect to his fatherly dominationover his constituency. Balinese water systems. Geertz's (1980) notion of The political significance of this fact cannot be over- a negara, or theater state, operating indepen- stressed. On the one hand the ritual of the royal bath dently of the water and land base has received is the same as the ordinary rituals of blessing by considerable attention in cultural which each and every Merina assures and represents studies. More the reproductionof his family and himself. As such recently, Lansing (1987, 1991) has shown that a the royal ritual is a large scale descent group ritual more topographically and regionally sensitive and the whole kingdom appears as one large family. assessment provides a more complex interplay On the other hand, the ritual of the royal bath is a rit- between the theater state and the water-temple ual of the pre-eminence of the king, represented in system. His skillful assessment of the the ritual as the violent conqueror and absorber of water-tem- ple cattle and the Vazimba. Within this idiom the king is system and its interdependencywith agricul- by implication a violent being whose violence, ture and ideology is perhapsthe best presentation whether directed towards outsiders or subjects, is jus- of Rappaport's "ritually regulated ecosystem" tified. These two sides of the representationare in the theme available for a highly complex society- ritual quite inseparable and as a result subjects can significantly more complicated celebrate their own subordinationas though it were than Rappaport's their own reproduction[Bloch 1987:296]. own study of the New Guinea Highlands. As in the Maya area, Balinese water and its To draw from Roy Rappaport(1968), we are availability is of critical importanceand is regu- describing an aspect of his "ritually regulated lated by water districts or subaks.Laborers from ecosystem" here the term "ecosystem"is recast different villages and village districts (bandjar) as the complex social relationsbetween individu- farm within the subakdistrict, forming significant als and subgroups influenced by the availability political associations. Although unstated, such a of cattle and related environmentalconstraints- widespread social network between village dis- a system that promoteshierarchy and manifests a tricts may allow ready access to additionalwork- controlling elite. The political economy is par- ers drawnfrom a broadlabor pool at peak activity tially regulated by a state ideology that incorpo- times associated with the intensive demands of a rates a small-scale, lineage-specific ritual by rice economy. Unlike Geertz'sview of subakdivi- appropriatingits immediate significance to each sions as independent and autonomous, Lansing household and reinventing its meaning to place has demonstratedthat they are hierarchicallyorga- the supportpopulation in a subordinateposition. nized, with a dendriticplan mappedonto the bifur- Ideological regulation occurs when all subordi- cating system of streams and canals. A series of nates follow a calendar of prescribed events decision makers priestsccupy water temples established by the state and set in motion by the located at the juncture of downward-flowing actions of the king. Ritual and elite controls in diversion channels. Ultimate authority over the this context define themselves by way of reaching system is positioned with the High Priest of the into the base structure of society, examining Temple of the CraterLake, who resides near the aspects of routine behavior that reflect ordinary apex of the island a location viewed by the activity, and exaggerating and formalizing ritual Balinese as the source of their irrigationwaters. in a widely public context. In this way, the ordi- Many rhythms drive Balinese society, but a nary is made extraordinaryand used by the ritual centralelement is the tempo set by the water-tem- specialists to confirm their deep and compelling ple priests and their ability to regulate and sched- understandingof the root and essence of society. ule water for each subak. Lansing shows that Bali water amounts are finite, and allocation at each diversiondam depends on a complicated variety Bali is the source of one of the more recent and of informationmade most intelligible within the pervasive models incorporated, both explicitly watertemple. Planting and harvestingcycles are andimplicitly, in Maya studies the theaterstate. defined by consulting calendars both a sacred Scarborough] WATERMANAGEMENT AND THE MAYA 147 and secular seasonal round as well as the imme- Classic period (cf. Marcus 1993). diate conditions reported by the farmers them- Perhaps the best informationabout reservoirs selves during their regular devotional visitations and social structurein contemporaryMaya soci- to the water temples. Lansing illustrateshow pest ety is presented by Evon Vogt's (1969) classic damage is curtailed by preventing contaminated study of the communityof Zinacantanin highland paddy fields from spreadingweevil infestations, Chiapas. By way of direct historical analogy, forcing necessary but extended fallow conditions water ritual in Zinacantanreflects less the drama on portions of a subak.I would suggest that the performed in Classic Maya civic centers, and structureof the subaksystem, with farmerscom- more the deeply rooted and conservative, every- ing from several bandjars,spreads such risks and day activities of a sustaining population from allows the irrigationof anotherset of paddies. antiquity.Zinacantan ritual is not viewed as a bro- Because of the strong Balinese-Hindu ideol- ken down set of ritual survivals threads and ogy pervadingthe island and the role of the water- patches of high rituals performed by kings and temple hierarchy, considerable ceremony and nobility during the Classic period but as an ritual are invested in the success of the rice crop adaptiveset of indigenouspractices with antiquity (Barth 1993). With each new crop, "holy water" probablyas great as the ancientkings themselves. is obtainedfrom the Temple of the CraterLake at According to Vogt, settlementand social rela- the summit of the water system and carrieddown tionships in Zinacantan are significantly influ- to each water temple to bless the predicted enced by the location of the water hole. A water bounty.Lansing describes the activity and empha- hole group consists of clustersof dispersedneigh- sizes the role of purityin obtainingwater from the borhoodsor snas,the latterdefined as "residential "source"and the domain of the High Priest. units composed of one or more localized patrilin- Although coming from halfway around the eages" (1968:158-159). A collection of water world, the Balinese case study provides a window hole groups forms a hamlet, or paraje, 11 of into a previously engineered landscape that has which together define Zinacantan when the long disappearedin the Maya Lowlands. Water cabecera or ceremonial and political center is imagery and ritual remain fundamentalto con- included. Vogt indicates that "waterhole groups temporaryMaya groups, as evidenced by the fre- cut across paraje lines," which may suggest a quently described Cha-Chaac ceremonies similar economic and social risk-sharingstrategy performed in the field to avert drought (Farriss to that of the subaklbandjarassociation noted in 1984:292; Freidel et al. 1993). But as Farriss the Balinese case. Regardless,"the waterholes are demonstrates,these ceremonies represent a sig- highly sacred, and there are myths about each of nificantly altered survival of precolonial public them describing the circumstances under which ritual formercommunity ritual conducted in the the ancestorsfound the water and how the water- grand civic spaces of ruined cities, today per- hole acquiredits distinctivename. The waterholes formedon agriculturalplots away from towns and are also the focal points for special ceremonies, churches. called k'inkrus, performed by their neighborhood settlements"(Vogt 1968:158). Chiapas Highland Because water holes run dry seasonally, water The Maya areahas been studiedethnographically, hole groups regularlyconsolidate aroundthe few ethnohistorically,and linguistically for nearly a more permanent sources in the dry season but century. But as Farriss suggests, much changed extend back out to the nearest seasonally filled structurallyin society following the Classic and source duringthe rainy season. The k'inkrus cer- Early Postclassic periods (> A.D. 1200), making emony functions to clean and maintainthe water clear comparison to the Classic past from con- hole twice a year but also appearsto purify ritu- temporaryMaya populations difficult. Although ally the Zinacantecosthemselves (Vogt 1968:167, less integrated,less complex, and much reduced 1969:147). in scale, present-day Maya groups still reveal Zinacantan water holes indicate the impor- valuable infwormationabout their landscape,water tance of water ritualand shamanismamong small use, and associated rituals that is pertinentto the populations of living Maya. Their water-control 148 LATINAMERICAN ANTIQUITY [Vol. 9, No. 2, 1998 system is undeveloped and functionally passive, control over their supportconstituency, in partby not at all as sophisticatedas the landscape engi- employing elaboratewater ritual. For the Classic neered by the ancientMaya. In spite of the ethno- Maya, however, ethnographyis absent, forcing a graphic record's richness in describing the reexaminationof a thirdresearch arena iconog- water-holegroup in Zinacantan,it is the water use raphy, the last major source for assessing ritual and ritual of the Balinese that may best explain elite propagandaand the importanceof water. the engineered landscape and iconographic dis- Because of the importance of water to play of the ancient Maya at the summits of their Mesoamericansocieties, a portion of this section great water-temple mountains. Nevertheless, refers to iconography outside the Maya area. Zinacantandemonstrates the role of reservoirsin These dataindicate ways in which waterwas used coordinatingdaily activities in a rural Maya set- and reveredin neighboringstates, or chiefdoms in ting and the influence of ritual in elevating the case of the Olmec. Tikal will represent the water's fundamentallife-sustaining propertiesto focused context for well-provenienced imagery the extraordinary;it probablyillustrates the kinds from a single site in an attempt to link best the of adaptationsmade by the smallest aggregatesof functional hydraulic system outlined above with ancient Maya away from their cities. ritual display though supportive iconographic The above examples of water's role in ritual expressionfrom elsewhere is mentioned.The evi- emphasize water's daily use and how its eco- dence presentedwill challenge the sensibilities of nomic significance can coordinate social activity some readers. Nevertheless, the argumentsdraw at many different levels. In the case of from the writings of respected Mesoamericanart Madagascar,power is institutionalizedby appro- historians and symbolic anthropologists who priating water ritual based on ancient traditions have, in ways, recently redirectedtheir subdisci- defined at the lineage level, a design projectedfor plines. In keeping with the epigraphat the begin- the Classic-period Maya. At similar complex ning of the article, initial forays into water institutional levels, the high priest and his imagery by ecologically oriented archaeologists acolytes in Bali, located at the summit of the are necessary but in dialogue with those already island, command the same kind of sway over the treatingthe subject matter. water source as envisioned for the elevated rulers at Tikal; neither is in a position to coerce control GreaterMesoamerican Water Traditions severely, but throughritual manipulationis capa- and the Maya ble of directingand appropriatinglabor and social An importantlink in integratingreservoir activi- activity furtherdownslope. Ritual in this context ties into the limited corpus of iconographic functions as a conduit for the exchange of agri- images featuring reservoirs is their association cultural or economic information. Aspects of with caves, springs, and mirrors. Unfortunately, water ritual maintain a conservative presence in direct historical or epigraphic allusions connect- Zinacantan, ritual continuing as survivals from ing tank systems to these other images are rare the more ruralpractices of the Classic Maya non- and subject to interpretation.This is especially elite precisely the kind of practices that were true in the Maya area where the human-made appropriatedby the nobility. convex microwatershedsof the Classic-period- evolving from the less centralizedPreclassic sys- Iconography and Water tem indicate the control necessary to order The ethnographicrecord shows the significance society's water needs, supplying it to at least a of water ritual and how some elites acquire ele- portion of the supportpopulation in proximity to ments of power by appropriatingeveryday water the great centers. Unequivocal evidence for elite activities in developing a state-level ideology. manipulation of water ritual is not easily mar- These social anthropologicalexamples together shalled empiricallyand requiresfrequent associa- with the documentationof the ancient, physical tions to other zones of Mesoamerica. water system and its resource centralizing The hillock-positioned cities of the Classic- design suggest how the Maya elite could have period Maya were derived from a slowly altered Scarborough] WATERMANAGEMENT AND THE MAYA 149 landscape with deep ancestralroots of perhaps a Caves and Springs millennium. The passive water systems of the Preclassic assumed the presence of water and Much of what is linked to water imagery in invested in sustaining and extending its natural Mesoamericais associated with cave symbolism; availability, not reinventing the landscape to caves frequentlydefine the springheadfor an oth- accommodate true water control by creating a erwise scarce water supply. Where precipitation "source."This lack of immediate environmental was slight at such early sites as Chalcatzingo, or economic control did not stimulateless icono- cave and water associations are clear (Grove graphic imagery or excite a less-developed, 1987). Nevertheless, even here recent evidence underlying water ideology than did a later bur- for holding tanks (Angulo 1993; Grove 1987) geoning Classic elite expressing itself in the con- suggests an additional connection between the trol of water and the associated social relationsof iconography and reservoirs. The fierce-looking production. Evidence for a long-lipped earth Relief IX (Figure 7), with its symmetrically monsteris widespreadfrom the Pacific Coast into squarelikeor quatrefoilmaw, need not be a cave the interior of the Peten, the latter at sites like opening given Reilly's novel claims. Below I sug- and Cerros before the Classic period gest that during the Classic period in the Maya (Tate 1982). At Abaj Takalik and , on the Lowlands the quatrefoil represented reservoir Pacific side, the long-lipped trait is associated imagery as well as cave associations. Relief IX with celestial water givers at an early date as well may accent the importance of Reilly's plazas (Parsons 1972; Quirarte1977). and/or the role of reservoirs in storing water in Some of the most evocative interpretationsof proximityto the core areaof some Olmec centers. water imagery are associated with Olmec The gaping maw may be a plaza surface or tank antecedents that appear incorporated into later opened wide to receive or belch forth those pass- Classic Maya waterritual much of it understood ing between worlds. as evolved shamanisticritual (Freidel et al. 1993). At Teotihuacan the significance of spring In a recent paper, Reilly (1994) reveals several imageryis most profoundwith the Pyramidof the intriguing hypotheses for the symbolic function Sun constructed over a caverned water source of Olmec plazas. At La Venta, specifically, he (Heyden 1975, 1981). Springswere the city's prin- suggests that the buried serpentine mask or cipal source of potable water and a considerable Massive Offering 2 in Plaza A was an expression amount of its agriculturalsupply (Sanders et al. of the wateryunderworld in which a key rulerwas 1979). In a graphicimage from the Tlalocanmural placed some 4 m above the offering. Reilly's in the Tepantitlacompound (Stone 1995; Taube interpretationhas Monument 6, the basalt sar- 1983) (Figure 8), water mountains were the cophagus carved in the image of a swimming sources of springs the Pyramidof the Sun may crocodilian and preparedfor an uninterredruler, have been the material, human-madesymbol of floating but teetering into the underworld. these mountains.Water mountains were function- Over a millennium later at Tikal a set of four ally understoodto provide springwater associated carved long bones found in Tomb 116 associated with cave entrances.Among the Aztec, Brodaindi- with Temple 1 and RulerA (Hasaw Ka'an K'awil) cates that the word in Nahuatlfor village or com- suggests a similar scene (Schele and Miller munitywas altepetl,which translatesas "mountain 1986:Figure7.1) (Figure 6). Here a canoe load of of water"(Broda et al. 1982; Fash 1996).3 animal spirits (naguals) and the accompanying Less metaphorically,the Maya actually con- Paddler Gods escort the ruler into the watery structedfunctional water mountains where there depths. In two of the carvings the canoe is shown were none. Ironically the illustration from the dipping into the watery world, with the back of Tlalocan mural may best depict the site of Tikal, the ruler's wrist pressed to his forehead.As early althoughno known springsexist at Tikal, and few as the Preclassic, or Formative, period, water are identified at ancient lowland cities. Cave and symbolism was well developed and associated spring imagery was used to emphasize the import with the heart of an architecturalcenter. of water use both functionally and ritually- n o Stingray Paddler Jaguar Paddler

s 0QZ1191gs1SgeE

dead parrot Kankin king o hiS passing

Ja PaddW m :n

z

z

-

Jaguar Paddler

canoe

|| "aood"

Figure 6. Incised bones from Burial 116 at Tikal (from Schele and Miller 1986). (Drawing courtesy of Linda Schele and the Kimbell Art Museum)

o

.

y

z o

lv y

co Scarborough] WATERMANAGEMENT AND THE MAYA 151

Figure7. Relief 9 at Chalcatzingo(from Stone 1995).(Courtesy of the Universityof TexasPress and AndreaStone) throughoutMesoamerica; in the Maya area much symbolism to elevate their worth and control over of this iconographywas probablyrelated to reser- society. The sometimes reflective, sometimes voir activities too. Essentially,reservoirs were the transparentquality of water may have predis- primarysource of predictablypotable waterin the posed it as a source of both contemplation and southern Maya Lowlands during the Classic purity.Large, human-made bodies of water,when period making most iconographic allusions to located within the core of a sizable center sur- water referencesto reservoirsas well. rounded by towering structures,functioned par- tially as reflective surfaces. Saunders's (1988) MirrorImages and SurfaceTensions recent survey of mirrors and mirror imagery in Of special import is the influence of mirrorsand the greaterMesoamerican lowlands suggests the reflections suggested by shamanistic ritual and implicationsof creatinga reflective reservoirsur- the role of reservoir surfaces. The associated face. water symbolism incorporatedby Mesoamerican The manufacture of magnetite and ilmenite groups indicates the empoweredrole of elite spe- mirrors as early as Olmec times at San Jose cialists in evoking and controllingbroadly under- Mogote and their movement into the Olmec stood knowledge derived from deeply rooted Veracruzand Tabascoheartland persuasively doc- traditions,with more subtle and esoteric aspects ument the sociopolitical significance of these sta- of ritualused to mystify, once these basic associ- tus markers(Flannery 1968; Pires-Ferreira1976). ations to water's use were made. Kings and nobil- Their use was more than socioeconomic, as illus- ity drew on the significance of water and water tratedon the many pieces of Olmec and Maya art symbolism to all Maya and then appropriatedthat (Saunders 1988). Frequent association with 152 LATINAMERICAN ANTIQUITY [Vol. 9, No. 2, 1998

Ancient Maya WaterRitual Although many iconographic elements allude to water, e.g., waterlilies, fishes, etc. (Puleston 1977; Schele and Miller 1986), high-performance waterritual is best discernablein scenes capturing aspects of water's use. Ethnographyas well as on- the-groundhydraulic systems help in interpreting the significance of these depictions. By examin- ing the following scenes principally from Tikal we are allowed a glimpse into how the Maya elite may have transformedand appropri- ated mundaneeveryday activities into ritual per- formance and power. Their associations with Figure 8. Watermountain from the Tlalocan Mural at reservoir imagery, possible acts of quarrying, Tepantitla,Teotihuacan (from Stone 1995). (Courtesyof even the passing into and out of the watery world the Universityof TexasPress and AndreaStone) of afterlife,reveal aspects of everyday life that an rulers' attire suggests shamanistic connections entire society faced and questioned. (Freidelet. al 1993; Schele and Miller 1986), with A repeated set of images found at several mirrorsreflecting the other parallel worlds from southernMaya Lowland sites includingTikal- which curers are able to marshall supernatural and associated with both water and earth is the forces. Taube (1983: 112-119, 1992: 186-189) Cauac, or Witz, Monster and the WaterlilyMon- shows the significance of scrying ("crystal gaz- ster (Fash 1996; Schele and Miller 1986; Spero ing") with mirrorsand watery surfaces through- 1986; Tate 1980; cf. Stuart 1987) (Figure 9). Tate out Mesoamerica. The Tikal bones and the La indicates that the Cauac symbol "was a place, the Venta sarcophagusillustrate the passing of rulers earthor ancestralabode, for the transformationof throughsuch watery "surfacetension." matter into energy" (1980: 111 in Spero At Tikal (Carrand Hazard 1961; Scarborough 1986:186).5 Perhaps most illustrative was the and Gallopin 1991), Calakmul(Folan et al. 1996), Cauac Monster identified with frequent images (Folan et al. 1983), (Beetz and referencing water, but defined by a zoomorphic Satterthwaite1981; Chase and Chase 1987), Rio head with a cleft or "fontanel"(Schele and Miller Azul (Adams 1990), La Milpa (Scarboroughet. al 1986) connectingor opening to an underworld.As 1995; Tourtellotet al. 1993), and many otherprin- a half quatrefoil, Fash (1996) associates similar cipal Maya cities, reservoirs rest immediately imagery at Copan with caves or waterholes.The below the most grandtemples and palaces. From Cauac Monster is frequently positioned as the the temple and palace forums, high-ritualperfor- ground on which an ascending ruler stands, pro- mance by an elite overlookingthe reservoirsreen- viding the symbolic supportof the "earthor ances- forced the temple dramato those in proximity to tral abode"in legitimizing sovereignty.Given the the tanks. The thin reflective reservoir surfaces connotations,the Cauac Monster and its related defined the tension between this world and the manifestationsmay represent the stone-quarried next, with the mirroredritual actions of the elite chasms cut into the summits of many large Maya strengtheningtheir associationto and controlover cities; the Maya unearthedthe building blocks for water. That these great mirrored surfaces were pyramids and palaces converting "matter to also the metaphorical source for the most pre- energy" then filled their quarriedchasms with cious of economic resources a terrestrialwater water in deE1ningthe reflective surface between source created through societal energies with this world and the next. The quatrefoilenclosure is the same resource acting as a potential passage- argued as "an opening between cosmic realms, a way to other worlds, ritual activity in association yawningchasm in the earth'scrust that leads to the comes as no surprise.4 underworld"(Tate 1980:47 in Spero 1986:186). Scarborough] WATERMANAGEMENT AND THE MAYA 153

Figure9. CauacMonster from Stela 1 at (from Stone 1995).(Courtesy of the Universityof TexasPress and

Altar D from the North Acropolis at Tikal 269). It is in this context that the governing elite (Jones and Satterthwaite1982) (Figure 10) shows legitimized their rightful association to the four repeatedviews of God N, with his water-tur- actions of their ancestorscarved at the summitsof tle carapace seated in a cartouchelikequatrefoil, their water mountains.By metaphoricallypassing extendinghis left arrnand hand throughthe frame into the watery realm, the king was associating to offer tobacco, perhapsa flint, and even maize. himself with the immortalityof deities like God N Andrea Stone (1995) suggests that the enclosing and the transformativeeffects of water.The canoe border aroundGod N is a cave, his gifts associ- scenes from Burial 116 carryingRuler A into the ated with his emergence from the underworld. underworld lie within 100 m of the Palace Given the absence of well-defined caves and the Reservoir,Temple 1 acting as the cenotaph over- size and abundance of reservoirs at Tikal as looking the reflective surfaces of the central- well as the population'sdependency on these lat- precinct reservoirs at Tikal. Although the ter features anotherinterpretation of Altar D is iconography associated with water management more likely: God N's water-adaptedattire may harbors expected ambiguity, it demonstratesthe represent his relationship to the great reservoirs significance of water ritual to the Maya elite. within the central precinct of Tikal, and his holy Water containment and consumption for offerings the kinds of gifts proffered(see Note 4). domestic and agriculturalpurposes away from the God N's ability to pass though this world and into central precincts of the largest cities was an the next suggests the transformativerole of reser- everyday act, but one dependent on the sustain- voir symbolism. ability and longevity of the centers. From the Schele and Miller (1986) indicate that two principal cities, it is posited that Classic Maya principal metaphorsdefined death for the Maya rulersinfluenced their constituencies by inventing elite. The first was representedby a fall or sinking a state ideology that emphasizedthe ritual appro- into a watery underworldor into the open maw of priationof water's use. The power of Tikal and its an earthmonster, the second, a canoe tripthat car- many subordinatecapitals, for example, encom- ried the ruler into the underworld."The Xibalba passed a political domain or regional state of of the Classic period was different in one way approximately 2,000 km2 (Culbert et al. from the Popol Vuh version of Hell. It was a 1990:117)6-Tikal manifesting a rich record of watery world that could only be entered by sink- water imagery. Copan represents another center ing beneathwater or by passing througha maw in with abundantand well-documented water sym- the surface of the earth (p. 267) .... A second bolism (Fash 1996). As mentionedearlier, during metaphorof death was derived from the watery periods of prolonged drought farmers could nature of the Underworld and from the major migrate to the margins of the great cities to sus- vehicle of water transportation,the canoe" (p. tain themselves or at least this was the implicit 154 LATINAMERICAN ANTIQUITY [Vol. 9, No. 2, 1998

Figure10. God N fromAltar 4, Tikal(from Jones and Satterthwaite1982). (Courtesy of the Universityof Pennsylvania

message symbolized by the architectureof the specialistdeposits water from these locations into the grand catchment areas and tanks. More likely, streamsat theirlow wateryshrines. "This water both purifiesthe springand adds information from the car- centers became the locus for elaborateceremony, dinaldirection, which are calledupon in divinations constructed to make the mundune and common and prayers on behalf of clients with illness" the uncommon. Acknowledging the great water (Tedlock1982:139). This information provides direct mountains as the source for much of the water linksbetween the wateryshrines, the sacredlakes of made available to the hinterlandsis analogous to the community,and the four seas at the horizon [Bassie-Sweet1996:70]. the Balinese view that the CraterLake at the sum- mit of the island is the source for all water to the The significance of this water ritualis its similar- subak systems hydraulicallyan impossibility. ity to those performedin Bali and the possibility In a recent study by Bassie-Sweet (1996), a that the Quiche example is a survivalof an earlier good case for water shrines in the ancient Maya practiceassociated with the Classic Maya. Lowlands is developed. Drawing on the Popol Tikal, as the principalcase study for this pre- Vuh and ethnographicwork, she argues that the sentation, manifests four low-lying bajo-margin Maya geographicallyand topographicallydefined reservoirs (Figure 3 and 4). Located approxi- high-water shrines and low-water shrines. Low- mately in the cardinal directions, these features lying cenotes, lakes, springs, and pools of water are equidistantfrom the summit of the civic cen- in caves are low-water shrine sites, while moun- ter and primecandidates for the kind of low-water tains and ridges are potential high-water shrines. shrine areas identified in the ethnographicrecord. She further states that in Highland Guatemala Is it possible that the high-water shrines associ- today, some communities take water samples ated with the central-precinctreservoirs were the from high-water shrines elevated volcanic source of "holy water" offered to the carefully lakes and add them to their own low-water positioned bajo-margintanks? Such a ritual dis- shrines in an act of purification. play within the core of the Tikal community- scheduled at opportuneperiods of the year by the In the Quiche area the four seas of the horizon appear in the local landscape as four sacred lakes, each elite may have influenced much of the regional located in one of the cardinaldirections. After return- capital's sustaining population. Given that the ing from a pilgrimage to the sacred lakes, the ritual bajo-marginreservoirs were designed to extend Scarborough] WATERMANAGEMENT AND THE MAYA 155 the agriculturalproductivity of the area, high rit- analysis proposes a causal link that endeavors to ual conducted in the context of a water "cosmo- explain how elites are able to maintaincontrol of gram" (cf. Ashmore 1989) of the site was a a convincing ideology, one that persuades and transparentway of appropriatingthe mundane accommodatesthe needs of a subordinatesupport activities of the hinterlands by the ruling elite. population. Not unlike the structurallysimilar Balinese case, The research arena associated with water and it also allowed a monitoring of agriculturalpro- land management in the Maya area remains ductivity by the Maya elite. sorely understudied.Until more time and energy are invested in the study of landscapes and their Conclusion implicationsfor the political economy, ambiguity Maya political economy was based on the fragile, will continue. Through controlled excavation, sometimes inhospitablesurrounds of a semitropi- survey, and artifactual analysis of reservoirs, cal jungle. The immediate abundanceapparent in catchmentzones, and drainages,we will begin to some other archaic statecraftexperiments associ- appreciatethe complexity of the Maya physical ated with permanentwater sources principally system. Further,ethnohistoric and ethnographic high-volume rivers with rich sediment loads- studies need to examine water systems as a prin- was not an option. Nevertheless, diverse biologi- cipal componentin evaluatingthe organizationof cal resources were harnessed by incrementally labor and land management an orientation changing the landscape to capture some of the already taken in many other parts of the world many variant and alternative energy pathways (Scarboroughl991b). defining a tropical ecosystem, harvesting those This presentation has attempted to integrate pathways that accommodated humans. two researchorientations that have grown apart- Centralizingthe reservoir systems at the summit ecological and economic studies vs. ritual,icono- of many of the great centers during the Classic graphic, and ideological examinations. The period was one of the ancient Maya's most sig- division between these two orientations is par- nificant evolutionaryadaptations. tially an academic one based on institutionaldis- However, constructingthe great "watermoun- ciplines and how we are trained to evaluate tains" of the Classic period was not enough to information.However, if more time and energy is permitkingly power; the political economy of the not invested in "crossingover," vast data sets are Maya requiredaggressive warfare,sizable luxury sure to be neglected. Anthropologicalarchaeolo- surpluses, and charismatic leadership. It also gists are intellectually positioned to engage both required a persuasive set of state-level rituals, spheres. many appropriatedfrom the everyday activities of a sustainingconstituency. The physical and sym- Acknowledgments.The genesis of this paperwas an invitation bolic investments associated with water manage- from Elin Danien to participate in the University of ment were key elements in centralizing a Pennsylvania/UniversityMuseum's Maya Weekendon March naturallyless-centralized society, one modeled on 29-3 1, 1996. The title of the conference was "Stone Mountain, Sacred Stone: The Face of Maya Ritual," forcing a different set of environmental and organiza- an economically and ecologically oriented archaeologist to tional expectations than some other early exam- assess ideological issues governingMaya decision making. In ples of statecraft. additionto Elin and the staff responsiblefor the grandgather- The ultimate effects of specific rituals in ing, I would like to thankWendy Ashmore and JeremySabloff ancient Maya society remain unclear, but by for the mannerin which the museumextended itself. The orig- inal draft of the paper was submitted to Latin American examining the function of ritual in other complex Antiquity at the invitation of Gary Feinman and together societies, as well as among contemporaryMaya, with comments and reviews from Joyce Marcus, Peter a better understandingis possible. This approach Harrison,Dorie Reents-Budet,Barry Isaac, and Boyd Dixon, to elucidating ancient Maya ritual permits a I was encouragedto revise the manuscriptfor publication.The groundingin ecology and economy, the base sub- manuscript benefited from the constructive exchange pro- vided by Pat Culbert and his invitation to discuss this paper structureon which society depends.The everyday with his students in his graduate Maya seminar at the activities of a group stimulate much of the cre- University of Arizona. Andy Hofling read versions of the ativity for what is ideologically possible. This paper and offered incisive commentaryat the outset. Barbara LATIN 156 AMERICANANTIQUITY [Vol. 9, No. 2, 1998

Fash provided her excellent unpublishedCopan water-man- Traditional Societies, edited by D. Cannadine and S. agement manuscript for which I am grateful. Additionally, Price, pp. 271-297. Cambridge University Press, New Julie Kunen,an advancedgraduate student at the Universityof York. Arizona, provideda fine seminarpaper that broughtmy atten- Broda, J., D. Carrasco,and E. Matos M. tion back to the Balinese case studies. My wife, Pat Mora, 1982 The Great Temple of Tenochtitlan, Center and Periphery in the strengthened the presentation at several critical points. Aztec World.University of California Press, Berkeley. Revisions to this manuscriptwere completed during a sabbat- Carneiro,R. L. ical leave from the University of Cincinnati,a year-longleave 1992 Point Counterpoint: Ecology and Ideology in the made partiallypossible by a Taft Faculty Sabbatical Grant.I Development of New World Civilization. In Ideology am grateful further to the School of American Research in and Pre-Columbian Civilizations, edited by A. A. Santa Fe for a summerresidency during 1996. Demarest and G. W. Conrad, pp. 175-204. School of American Research Press, Santa Fe, New Mexico. References Cited Carr,R. F., and J. E. Hazard 1961 Tikal Report No. 11: Map of the Ruins of Tikal, El Adams, R. E. W. Petetn,Guatemala. Museum Monographs.University of 1990 Archaeological Research at the Lowland Maya Pennsylvania, Philadelphia. City of Rio Azul. Latin American Antiquity 1:23A1. Chase, A. F., and D. Z. Chase Adams, R. McC. 1987 Investigations at the Classic of Caracol, 1992 Ideologies: Unity and Diversity. In Ideology and Belize: 1985-1987. Monograph No. 3. Pre-Columbian Pre-Columbian Civilizations, edited by A. A. Demarest Art Research Institute, San Francisco. and G. W. Conrad, pp. 205-222. School of American Childe, V. G. Research Press, Santa Fe, New Mexico. 1950 The Urban Revolution. Town Planning Review Andrews, E. W. V, and N. Hammond 21 :3-17. 1990 Redefinition of the Swasey Phase at , Belize. Coggins, C. C., and O. C. Shane III American Antiquity54:570-580. 1984 Cenote of Sacrif ee: Maya Treasures from the Angulo V., J. Sacred Wellat . University of Texas Press, 1993 Water Control and Communal Labor during the Austin. Formative and Classic Period in Central Mexico. In Covarrubias,M. Economic Aspects of Water Management in the 1937 Island of Bali. Alfred A. Knopf, New York. Prehispanic New World, edited by V. L. Scarborough Culbert,T. P., L. Kosakowsky,R. E. Fry, and W. A. Haviland and B . Isaac, pp. 151-220. Research in Economic 1990 The Population of Tikal, Guatemala. In Anthropology, Supplement 7. JAI Press, Greenwich, Precolumbian Population History in the Maya Connecticut. Lowlands, edited by T. P. Culbert and D. S. Rice, pp. Ashmore, W. 103-122. University of New Mexico Press, 1989 Construction and Cosmology: Politics and Albuquerque. Ideology in Lowland Maya Settlement Patterns. In CurrentAnthropology Word and Image in Maya Culture: Explorations in 1996 Agency, Ideology, and Power in Archaeological Language, Writing,and Representation,edited by W. F. Theory. CurrentAnthropology 37: 1-86. Hanks and D. S. Rice, pp. 272-286. University of Utah Dahlin, B. H. Press, Salt Lake City. 1984 A Colossus in Guatemala: The Atran, S. City of El Mirador.Archaeology 37:18-25. 1993 Itza Maya Tropical Agroforestry. Current Demarest,A. A. Anthropology 34:633-700. 1992 Ideology in Ancient Maya CulturalEvolution: The Barth,F. Dynamics of Galactic Polities. In Pre-Columbian 1993 Balinese Worlds. University of Chicago Press, Civilizations, edited by A. A. Demarest and G. Chicago. W. Conrad, pp. 158-158. School of American Research Bassie-Sweet, K. Press, Santa Fe, New Mexico. 1996 The Edge of the World: Caves and Late Classic Farriss,N. M. Maya World View. University of Oklahoma Press, 1978 Nucleation versus Dispersed: The Dynamics Norman. of Population Movement in Colonial Yucatan. Hispanic Beach, T., and N. P. Dunning American Historical Review 58:187-216. 1997 An Ancient Maya Reservoir and Dam at 1984 Maya Society under Colonial Rule: The Collective , El Peten, Guatemala. Latin American Enterprise of Survival. Princeton University Press, Antiquity:20-29. Princeton, New Jersey. Beetz, C. P., and L. Satterthwaite Fash,B. W. 1981 The Monuments and Inscriptions of Caracol, 1998 IconographicEvidence for WaterManagement and Belize. University of Pennsylvania Press, Philadelphia. Social Organizationat Copan. In Bell,C. The Rise and Fall of the Classic Maya Kingdom of Copan, edited by W. L. 1992 Ritual Theory, Ritual Practice. Oxford University Fash and E. W. Andrews V. School of American Press, New York. Research Press, Santa Fe, New Mexico, in press. Ms. Bloch,M. 1996. 1987 The Ritual of the Royal Bath in Madagascar:The Fash,W. L. Dissolution of Death, Birth and Fertility into Authority. 1991 Scribes, Warriorsand Kings: The City of Copan In Rituals of Royalty: Power and Ceremonialism in and the Ancient Maya. Thames and Hudson, London. Scarborough] WATERMANAGEMENT AND THE MAYA 157

Fash, W. L., and E. W. Andrews V (editors) Jones, C., and L. Satterthwaite 1998 The Rise and Fall of the Classic Maya Kingdomof 1982 Tikal Report No.33, Part A: The Monuments and Copatn.School of American Research Press, Santa Fe, Inscriptions of Tikal:The Carved Monuments.Museum New Mexico, in press. Monograph No. 44. University of Pennsylvania, Flannery,K. V. Philadelphia. 1968 The Olmec and the Valley of Oaxaca: A Model for Lansing, J. S. Inter-Regional Interaction in Formative Times. In 1987 Balinese "Water Temples" and Management of Dumbarton Oaks Conference on the Olmec, edited by Irrigation.American Anthropologist 89:326-341. E. P. Benson, pp. 79-130. Dumbarton Oaks, 1991 Priests and Programmers: Technologies of Power Washington,D.C. in the Engineered Landscape. Princeton University Flannery,K. V., and J. Marcus (editors) Press, Princeton, New Jersey. 1983 The Cloud People: Divergent Evolution of the Lockhart,J. Zapotec and Mixtec Civilization. Academic Press, New 1992 The Nahuas after the Conquest. Stanford York. University Press, Palo Alto, California. Folan, W. J., E. R. Kintz, and L. A. Fletcher MacNeish, R. S. 1983 Coba: An Ancient Maya Metropolis. Academic 1964 Ancient Mesoamerican Civilizations. Science Press, San Francisco. 143:531-537. Folan, W. J., J. Marcus, S. Princeman, M. Dominguez MacNeish, R. S., and A. Nelken-Terner Carrasco,L. Fletcher,and A. Morales Lopez 1983 The Preceramic of Mesoamerica. Journal of Field 1996 Calakmul: New Data from an Ancient Maya Archaeology 10:71-84. Capital in Campeche, Mexico. Latin American Marcus,J. Antiquity6:310-334. 1982 The Plant Worldof the Sixteenth- and Seventeenth- Friedel, D., L. Schele, and J. Parker Century Lowland Maya. In Maya Subsistence: Studies 1993 Maya Cosmos: Three Thousand Years on the in Memory of Dennis E. Puleston, edited by K. V. Shaman's Path. William Morrow,New York. Flannery,pp. 239-273. Academic Press, New York. Geertz, C. 1993 Ancient Maya Political Organization. In Lowland 1980 Negara: The Theatre State in Nineteenth Century in the Eighth CenturyA.D., edited by Bali. PrincetonUniversity Press, Princeton,New Jersey. J. A. Sabloff and J. S . Henderson, pp. 111- 183. Groube,L. DumbartonOaks, Washington,D.C. 1989 The Taming of the Rain Forests: A Model for Late Marcus,J., and K. V. Flannery Pleistocene Forest Exploitation in New Guinea. In 1994 Ancient Zapotec Ritual and Religion: An Foraging and Farming, edited by D. R. Harris and G. Application of the Direct Historical Approach. In The C. Hillman, pp. 292-304. Unwin Hyman, London. Ancient Mind, edited by C. Renfrew and E. B. W. Grove, D. C. (editor) Zubrow, pp. 55-74. Cambridge University Press, New 1987 Ancient Chalcatzingo. University of Texas Press, York. Austin. Matheny,R. T. Hammond,N. (editor) 1976 Maya Lowland Hydraulic Systems. Science 1991 Cuello: An Early Maya Community in Belize. 193:639-646. CambridgeUniversity Press, Cambridge. 1978 Northern Maya Lowland Water-ControlSystems. Hansen, R. D. In Pre-Hispanic Maya Agriculture, edited by P. D. 1991 The Road to . Natural History, May:8-14. Harrison and B. L. TurnerII, pp. 185-210. University Harrison,P. D. of New Mexico Press, Albuquerque. 1993 Aspects of Water Management in the Southern 1986 Investigations at E1 Mirador, Peten, Guatemala. Lowlands. In Economic Aspects of WaterManagement National Geographic Research and Exploration in the Prehispanic New World, edited by V. L. 2:332-353. Scarboroughand B. L. Isaac, pp. 71-120. Research in Matheny,R. T., and D. L. Gurr Economic Anthropology, Supplement 7. JAI Press, 1983 Variationin PrehistoricAgricultural Systems of the Greenwich, Connecticut. New World. Annual Reviews in Anthropology Harrison,P. D., and B. L. TurnerII (editors) 12:79-103. 1978 Prehispanic Maya Agriculture. University of New Matheny,R. T., R. Hansen, and D. L. Gurr Mexico Press, Albuquerque. 1980 El Mirador, Peten, Guatemala:An Interim Report. Haviland,W. A. Papers No. 45. New WorldArchaeological Foundation, 1992 Status and Power in Classic Maya Society: The Brigham Young University, Provo, Utah. View from Tikal. AmericanAnthropologist 94:937-940. McAnany,P. A. Hendon, J. A. 1995 Living with the Ancestors: Kinshipand Kingship in 1991 Status and Power in Classic Maya Society: An Ancient Maya Society. University of Texas Press, Archaeological Study. American Anthropologist Austin. 93:894-918. Montejo, V. Heyden, D. 1991 In the Name of the Pot, the Sun, the Broken Spear, 1975 The Cave underneath the Pyramid of the Sun at the Rock, the Stick, the Idol, Ad Infinitum and Ad Teotihuacan.American Antiquity40:131-147. Nauseum: An Expose of Anglo Anthropologists' 1981 Caves, Gods and Myths: World-view and Planning Obsessions with and Invention of Mayan Gods. Paper in Teotihuacan. In Mesoamerican Sites and World- presented at the 90th Annual Meeting of the American Views, edited by E. P. Benson, pp. 1-35. Dumbarton AnthropologicalAssociation, Chicago. Oaks, Washington, D.C. Nations, J. D., and R. B. Nigh 158 LATINAMERICAN ANTIQUITY [Vol. 9, No. 2, 1998

1980 The Evolutionary Potential of Lacandon Maya Mirrors. In Recent Studies in Pre-Columbian Sustained-YieldTropical Forest Agriculture.Journal of Archaeology, edited by N. J. Saunders and O. de Anthropological Research 36:1-30. Montmollin, pp. 1-39. BAR International Series 421. Niederberger,C. B. British Archaeological Reports, Oxford. 1979 Early Sedentary Ecology in the Basin of Mexico. Scarborough,V. L. Science 203: 131-142. 1983 A Preclassic Water System. American Antiquity Parsons,L. 48:720-744. 1972 Iconographic Notes on a New Izapan Stela from 1991a Archaeology at Cerros, Belize, Central America, Abaj Takalik, Guatemala. Atti de XL Congresso Volume111: The Settlement System in a Late Preclassic International degli Americanisti, pp. 203-212. Casa Maya Community. Southern Methodist University Editrice Tilgher, Genova, Italy. Press, Dallas, Texas. Pires-FerreiraJ. 1991b WaterManagement Adaptations in Non-Industrial 1976 Shell and Iron-Ore MirrorExchange in Formative Complex Societies: An Archaeological Perspective. In Mesoamerica, with Comments on other Commodities. Archaeological Method and Theory vol. 3, edited by M. In The Early Mesoamerican Village, edited by K. V. B. Schiffer, pp. 101-154. University of Arizona Press, Flannery,pp. 311-328. Academic Press, New York. Tucson. Pohl, M. D., K. O. Pope, J. G. Jones, J. S. Jacob, D. R. 1993a Water Management Systems in the Southern Piperno, S. A. deFrance, D. L. Lentz, J. A. Gifford, M. E. Maya Lowlands: An Accretive Model for the Danforth,and J. K. Josserand Engineered Landscape. In Economic Aspects of Water 1996 Early Agriculture in the Maya Lowlands. Latin Management in the Prehispanic New World,edited by American Antiquity 7:355-372. V. L. Scarboroughand B. L. Isaac, pp. 17-69. Research Puleston, D. E. in Economic Anthropology, Supplement 7. JAI 1977 Press, The Art and Archaeology of HydraulicAgriculture Greenwich, Connecticut. in the Maya Lowlands. In Social Process in Maya 1993b Introduction. In Economic Aspects of Water Prehistory: Studies in Honour of Sir Eric Thompson, Management in the Prehispanic New World,edited by edited by N. Hammond, pp. 448A67. Academic Press, V. L. Scarboroughand B. L. Isaac, pp. 1-14. Research New York. in Economic Anthropology, Supplement 7. JAI 1983 Press, Tikal Report No. 13: The Settlement Survey of Greenwich, Connecticut. Tikal. Museum Monographs. University of 1994 Maya Water Management. National Geographic Pennsylvania, Philadelphia. Research and Exploration 10:184- 199. Puleston, D., and O. S. Puleston Scarborough,V. L., M. E. Becher, J. L. Baker, G. Harris,and 1971 An Ecological Approach to the Origins of Maya F. Valdez, Jr. Civilization. Archaeology 24:330-337. 1995 Water and Land at the Ancient Maya Quirarte,J. Community of La Milpa. Latin American Antiquity 6:98-119. 1977 Early Art Styles of Classic Maya Art. In The Scarborough,V. L., R. P. Connolly, and S. P. Ross Origins of Maya Civilization, edited by R. E. W. Adams, 1994 The Prehispanic Maya Reservoir System at Kinal, pp. 249-283. University of New Mexico Press, Peten, Guatemala.Ancient Mesoamerica 5:97-106. Albuquerque. Scarborough,V. L., and G. G. Gallopin Rappaport,R. 1991 A Water Storage Adaptation in the Maya 1968 Pigs for Ancestors. Yale University Press, New Lowlands. Science 251:658-662. Haven, Connecticut. Schele,L., and M. E. Miller Reilly, K. F. III 1986 The Blood of Kings: Dynasty and Ritual in Maya 1994 Enclosed Ritual Spaces and the Watery Art. Kimbell Art Museum, Fort Worth,Texas. Underworld in Formative Period Architecture: New Schwartz,N. B. Observations on the Function of La Venta Complex A. 1990 Forest Society: A Social History of the Peten, In Seventh Round Table, 1989, edited by M. Guatemala. University of Pennsylvania Press, Greene Robertson and V. M. Fields, pp. 125-135. Pre- Philadelphia. ColumbianArt Research Institute, San Francisco. Spero,J. M. Reynolds,V., and R. Tanner 1986 Beyond Rainstorms: The Kawak as an Ancestor 1995 The Social Ecology of Religion. Oxford University Warrior,and Patron of Witchcraft. In Sixth Press, New York. Palenque Round Table, edited by V. M. Fields, pp. Rice,D. S., and T. P. 184-193, Culbert University of Oklahoma, Norman. 1990 Historical Contexts for Population Reconstruction Stone,A. J. in the Maya Lowlands. In Precolumbian Population 1995 Images from the Underworld: and History in the Maya Lowlands, edited by T. P. Culbert Traditionsof Maya Cave Painting. University of Texas and D. S. Rice, pp. 1-36. University of New Mexico Press, Austin. Press, Albuquerque. Stuart,D. Sanders,W. T., J. R. Parsons,and R. S. Santley 1987 Ten Phonetic Syllables. Research Reports 1979 The Basin of Mexico: on Ecological Processes in the Ancient Maya Writing, vol. 14. Center Evolution of a for Maya Civilization. Academic Press, New York. Research, Washington,D.C. Sanders,W. T., and B. J. Price Tate,C. 1968 Mesoamerica: The Evolution of a Civilization. 1980 The Maya Cauac Monster: Visual Random House, New Evidence for York. Ancestor Veneration among Saunders,N. J. Ancient Maya. Unpublished Master's thesis, University of Texas, 1988 Anthropological Reflections on Archaeological Austin. Scarborough] WATERMANAGEMENT AND THE MAYA 159

1982 The Maya Cauac Monster's Formal Development manner by which it feeds itself. All cultures have invented and Dynastic Contexts. In Pre-ColumbianArt History, creative ways of provisioning themselves and, in the case of edited by A. Cordy-Collins, pp. 33-54. Peek complex societies, ways of ensuring that some "agents"get Publications, Palo Alto, California. less than others, in terms of both quality and quantities of Taube,K. A. resources. From my perspective, these are organizationalor 1983 The Teotihuacan Spider Woman. Journal of Latin institutional American Lore 9:107-189. patterns that remain at least partially grounded 1992 The Iconographyof Mirrorsat Teotihuacan.In Art, in the limitations of an environment, an environment that is Ideology, and the City of Teotihuacan, edited by J. C. frequently engineered to produce more of what society Berlo, pp. 169-204. Dumbarton Oaks, Washington, wants. Some environs are naturally more productive than D.C. others, and technological adjustmentscan stimulate the mod- Tedlock, B. ification of a landscape to accommodate greater organiza- 1982 Time and the Highland Maya. University of New tional stability or resilience. For elites to marshall Mexico Press, Albuquerque. power the laborofothers, especially foodproducers they Tourtellot,G. III, A. Clarke, and N. Hammond probablyalso appropriatedideologies grounded in water and 1993 Mapping La Milpa: A City in NorthwesternBelize. Antiquity67:96-108. soil at both the local and regional level. A supportpopulation Trigger,B. G. in the decentralized archaic states of Mesoamerica could 1990 Monumental Architecture: A Thermodynamic "vote with their feet" if ideology did not have a "bottom-up" Explanation of Symbolic Behavior. WorldArchaeology component. 22:119-132. The insightful critique by Bruce Trigger (1996) of the 1996 Comment on "Agency, Ideology, and Power in above mentioned papers relates their shortcomings: "they Archaeological Theory." Current Anthropology tend to treat religion much as Voltaire did in the eighteenth 37:63-64. century as an exploitative deceit that elites practice upon Vogt, E. Z. the lower classes. No convincing argumentis advanced why 1968 Some Aspects of Zinacantan Settlement Patterns the and Ceremonial Organization. In Settlement lower classes initially should have accepted such claims Archaeology, edited by K. C. Chang, pp. 154-173. if it would have been contraryto their own interests to do so" National Press Books, Palo Alto, California. (p. 64). 1969 Zinacantan: A Maya Community in the Highlands 2. The focus of the presentationconcerns the collection and of Chiapas. Belknap Press, Cambridge,Massachusetts. deliberateallocation of water by the ancient Maya. Whitehouse, R., and J. Wilkins However, 1986 The Making of Civilization. Alfred A. Knopf, New during the rainy season, rainfall and runoff were probably in York. excess of the storage capacities of even the largest Classic- Wilk, R. R. period centers. The drainage system designed for controlled 1985 Dry Season Agriculture among the Kekchi Maya distributions during the dry season was as effective in dis- and Its Implications for Prehistory. In Prehistoric charging overflow during the wet season. Lowland Maya Environmentand Subsistence Economy, edited by M. Pohl, pp. 47-58. Papers of the Peabody 3. Joyce Marcus (personal communication 1996) has made Museum of American Archaeology and Ethnology Vol. me aware of Lockhart's (1992) fine study in which he inter- 77. HarvardUniversity, Cambridge, Massachusetts. prets altepetl as a territory and the organization of people Wilken, G. C. controlling that territory. 1987 Good Farmers. University of California Press, 4. Berkeley. Cenotes during the Postclassic period were the deposito- Wittfogel, K. A. ries for significant ceremonial offerings. Coggins and Shane 1957 Oriental Despotism: A ComparativeStudy of Total (1984) reinventoried and interpretedmany of the offerings Power. Yale University Press, New Haven, made at the Sacred Well of Sacrifice at Chichen Itza. Connecticut. Classic-period reservoirs in the southernMaya Lowlands are expected to yield similar findings, the posited origin for this Notes water ritual. Unfortunately,little excavation has yet to occur in the human-madetanks of the 1. A set of articles examining the importance of "Agency, lowlands. Ideology, and Power in Archaeological Theory," recently 5. Bassie-Sweet (1996) indicates that "the word chac is a published in Current Anthropology 37:1-86 (1996), con- cognate for cauac 'lightning, thunder, storms"' (p. 56), a vincingly articulates the interpretive role of ideology for clear association to water. archaeologists. The focus of the discussions is about a polit- 6. Calakmul, as the capital of a regional state, administrated ical economy based on "top-down"elite to elite alliance for- control over an area of 8,000 km2,according to Folan et al. mations and power relationships drawing on material (1996:310), a region four times larger than the Tikal state symbols of control. Clearly, this is an importantelement in and one-and-one-half times larger than Bali. understandingthe influence of ideologies in decision mak- ing, but it does not address the "economy" in political econ- Received May 16, 1996; accepted January 24, 1997; revised omy well. The base stratumon which any society rests is the February 17, 1997.