reading response

Tikal, Guatemala and Mesoamerican Urbanism Author(s): William A. Haviland Source: World Archaeology, Vol. 2, No. 2, Urban Archaeology (Oct., 1970), pp. 186-198 Published by: Taylor & Francis, Ltd. Stable URL: http://www.jstor.org/stable/124131 Accessed: 22-06-2016 16:35 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

http://about.jstor.org/terms

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].

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World Archaeology

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism

William A. Haviland

Since the formulation of the concept of an 'urban revolution' by V. Gordon Childe (1951, orig. 1936), it has come to be accepted that urbanism occurs only in the context of civilization. Yet, it is not accepted that civilization necessarily implies urbanism. No- where do doubts on this score seem so well founded as is in the case of the Precolumbian

civilizations of America:

'American instances clearly show that high culture and the state type of political organ- ization can develop with or without city life' (Armillas, in Sanders and Price 1968: x)

The lack of urbanism in some American civilizations is generally linked to environ- mental differences. In Mesoamerica, urban civilizations are seen to have arisen in the arid

highlands where irrigation ('hydraulic') agriculture provided for high population densities. In the tropical lowlands, by contrast, a dependence on slash-and-burn ('milpa') agriculture kept population density low, with the bulk of the population scattered in small rural hamlets. A particularly clear statement of this point of view has recently been presented by Sanders and Price (1968: io), who go on to suggest that non-urban lowland civilization came about as a response to pressures set up by the hydraulic, urban civiliza- tion (Sanders and Price 1968: 203-6).

Given this, an examination of some of the evidence pertinent to the highland-lowland contrast in Mesoamerica seems useful. Since space is limited, I shall briefly summarize the current conception of Teotihuacan, the earliest example of highland urbanism. I shall then dwell at length on the lowland Maya site of Tikal, to see how it compares with Teotihuacan, and whether or not it fits the lowland pattern as described above.

Teotihuacan is a large archaeological site in a valley a short distance north-east of Mexico City. In the past several years both the site and the surrounding valley have been subjected to much research both by Mexican and United States archaeologists. Final, detailed reports of this work are not yet available. However, a broad general summary, from which my own is drawn, will be found in Sanders and Price (x968), who also pro- vide the references to various preliminary reports.

At its height, about A.D. 600, Teotihuacan is described as fully urban. At this time, the site covered an estimated 19 km.2 with a population of some 85,ooo people. It had a planned central core, with an unplanned peripheral zone. The core contains such monumental construction as the enormous pyramids of the 'sun' and 'moon', and the

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism I87

'citadel' and 'market'. Here also are large complexes of rooms arranged in compounds around courtyards separated from one another by narrow 'streets' in a gridded pattern. These are presumed to have housed composite families, and a population density of about 5,200 persons per km.2 is estimated for this zone.

The peripheral zone, of approximately I2 km.2, consists of apparent houses arranged in a haphazard manner. The estimated population density is on the order of 3,750 persons per km. 2 Monumental public architecture comparable to that of the central zone is lacking. Beyond the peripheral zone was open country, for the bulk of the population of the valley seems to have resided in the 'city' itself.

Prior to 300 B.C., the situation in the Teotihuacan valley was far different. The overall population seems to have been small, dispersed over the valley. Then there was an acceleration of population growth, with a shift to settlement into two fair-sized towns; one at the end of the valley and the other more centrally located near springs. By A.D. I, the latter settlement of Teotihuacan had grown to a size of i2 km.2 and included perhaps 50% of the valley population. These changes are linked to the inferred development of hydraulic agriculture. Following this, the history of Teotihuacan is one of expanding size, increasing socio-economic diversity, and expanding political influence up to its collapse c. A.D. 750.

A word of caution is appropriate here, for until final detailed reports are available, there is no opportunity for evaluation of current assertions as to the nature of Teoti- huacan. Currently, there are in print population estimates which run anywhere from 60,00o to 300,000 (Sanders and Price 1968: ISI). How are these conflicting claims to be evaluated? Beyond this, how was it determined that a structure was, in fact, a house rather than a kitchen, storehouse, governmental office, monastic cell, or some such other possibility? How was it determined what percentage of houses were lived in at one time? On what basis was average size of household determined?

There are other variables when relative amount of social differentiation, possible craft specialization by ward, and irrigation are considered. More in the way of positive, rather than indirect, evidence for irrigation is particularly desirable, since this is posited as a prime causative factor in the development of Teotihuacan. But with these cautions in mind, a reconstruction of Teotihuacan as fully urban does not seem unreasonable, taking

large population, nucleation and socio-economic diversity as urban criteria. As Teotihuacan was the largest centre of its day in highland Mesoamerica, so Tikal

was one of the largest lowland Maya centres. Located some 300 km. north of Guatemala City, Tikal was built on a broad limestone eminence in a medium rain-forest setting. Currently, an average of I35 cm. per year of rain falls at Tikal, mostly between June and December. There is no reason to assume that this, or the flora and fauna, have changed significantly over the last 2,000 years. Here the Maya settled in the last millennium B.C.,

and their civilization flourished until approximately A.D. 870. Since I956, Tikal and the surrounding region have been the subjects of concerted investigations under the auspices of the University Museum of the University of Pennsylvania and the Guatemalan government. Few of the final reports have been completed, but two summaries are available (Coe I965; 1967). Other publications have been issued on limited aspects of Tikal, and most are listed by Coe (I967). My discussion here is based on his summaries, my own work on the Tikal settlement pattern and burials (Haviland I965; I967, I969),

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

i88 William A. Haviland

_ -=I ;XX 4^~ 51 - j .2 v ) > ~ A ~ .. 4c '*. 1 i ) ' .4' G' 22 . 4a 220 ; Ti a:m )a "~ ) / a -'-~ ~L ,

z -H

'- v =5 X 0 --A" ' *-H; '? ^J IN

r -- - - e 7 -- g . - \ 'A \Vi ' ' _ \9

:, - - r. - , I _-- ---I, ~ -I-I.-

\ u\ \ f, -S \ 'if. V ( a 4 e 41 / V Ru; g ; - - 2d rr , f ' ' , "

/19 L"-'.elie-' 4 G ' -

t1 ( \ '- ~ I 4 t"--N;; . . . . yTi

>I/25O {A - ? / . . ' .

AgI IaSi

17ii 7'L5 T ^ T ^'--

i12 a

i/ C~~~~~~~~~ ?,~~~~~~~~~~~~~~~~~~~

,! . . ......,: <~ :~~ ,

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism 189

4 /

a

I' f t 51

1 *

pI Is

-,

21 , so

/ 9 r^ \\\

~-/ :.~

'4 A 1

A. A

.4 -;3L

I'l .

Al.

Al

Au A

Figure ig Map of the cetral i6 km.2 of Tikal, with the Great Plaza

at the centre.

Each square is 500 m. on a side

L . > / 1 .\ " I1"

r7 ] /* * </}-

! . i :, l ,I - K '. _I.-- ____ I '- < . ~ <> v ,.A -I_ __ I k. _ . \

?= V \ - I v^ 31 ,4J 14 I / * / , A A \

*;3 i^ i- e \ r- Wt I f 4 A

u A-kj , 1< A 0! , ^rT^^lN s ^ij 1iL '',- f / A -'

= r *3 , 1/ r^ y . 1' 1 a w l r \ ( i/; \\;i J /'i ( , /r

5*t'-," I t % ;' r'^ \ \ j. ! C ,, s.

* V ; f i .1. u /

^ ij f /I -

t. I e J-v ^^ - ,;

rcJ /9r c-~rI

8~ ~ ~~~~~? I /r /~ ~ 1

i

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

190 William A. Haviland

and other sources which will be specifically cited. Where necessary, I have updated these materials from my own files and those of Coe.

Surveys show that, at its height (after A.D. 550), Tikal covered a minimum of 123 km.2 (this corrects my previous calculation). The north and south boundaries are defined by artificially constructed earthworks, located at 4'5 km. north, and 8 km. south of the nucleus of Tikal (cf. Puleston and Calleder i967). These run between two extensive areas of seasonal swampland, devoid of habitation, which delimit the site on the east and west. Within these boundaries is a peripheral zone, where settlement was sparse, of roughly 60 km.2, and a central zone of roughly 63 km.2 Here settlement was far denser(fig. 19).

The heart of the central zone of Tikal was the 'Great Plaza', a paved area of about one hectare (plate 8). Concentrated around this are most of the 300 or more major structures, which include large ceremonial structures, ballcourts, and 'palaces'. My primary concern here, however, is with the thousands of probable house units which constitute the bulk of the structures in the central zone. These probably include most, but not all, of the major 'palaces'. That most of these were elite-class residences is indicated by five features. First, their overall appearance and arrangement suggests large-scale versons of smaller known residence groups. Indeed, there is an architectural continuum from the simplest, pole-and-thatch houses based on platforms of one, two and three levels, through small 'palaces' with walls partly of masonry, partly of pole-and-thatch, to the major all-masonry 'palaces'. Second, household trash clearly indicates that many of them were lived in immediately after the collapse of Tikal. Although this could represent people moving into previously non-residential structures (as they clearly did in some instances), in at least one case a series of similar small-scale modifications before and after the collapse suggests continuity of occupation from earlier times. In some other cases, such 'palaces' were preceded at the same locus by smaller houses, and again continuity of occupation is suggested. Third, at least one midden of domestic trash, dated

between A.D. 550 and 650, has been found in clear association with a major 'palace'. The reason such discoveries are not common may be that trash was certainly in high demand for use as construction fill, and so was not allowed to build up before it was hauled away. Fourth, burials have been found in some large 'palaces' which have their closest analogues in ordinary household burials. Finally, cached offerings and other indications of cere- monial activity are usually absent in these contextso

While perhaps most 'palaces' were residential, not all were constructed for this purpose. One marked exception is just east of the Great Plaza, in the form of outer and inner quadrangles of palace-like buildings, but which lack many features of presumed residential 'palaces', such as benches. The whole assemblage is suggestive of a large market, but although there is no evidence to contradict this hypothesis, neither is there positive evidence for it. Also non-residential are the long, narrow, single-room 'palaces' of the ceremonial twin pyramid groups (fig. 20).

Over I25 smaller structures of central Tikal have been excavated, but not all were houses. The exceptions include kitchens, family shrines, and what were probably administrative buildings. Domestic and associated out-buildings were usually arranged around small rectangular plazas in numbers of from two to five. These groups were probably extended family households.

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate i Dorestad near Wijk bij Duurstede. View of the

excavation in progress, with the post-holes of houses and fences marked by pegs

Plate 2 Amsterdam. Waste

wood of ships stacked under one of the posts of a medieval house at the harbour

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate 3 Dordrecht. Wooden revetment built against a 4 m. high embankment (left) in the riverbed (right). Each separate part of the revetment represents the width of a house-plot. The posts of older platforms can be seen. To the extreme right are the posts of a pier built against the revetment. Above this pier there was a stone paved street

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate 4 Siraf. General view of the site, looking west from the recent fort Photo. G. Sansbury

Plate 5 The largest cemetery at Siraf Photo. P. Farnworth

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate 6 Siraf. Site F. General view of the residential quarter, looking south Photo. G. Sholl

Plate 7 Site G. The shrine Photo. G. Sholl

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate 8 Air-view of the Great Plaza of Tikal, with east and west flanking temples and the North Acropolis (centre left)

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate 9 Excavation of Str. 4F-3 at Tikal, a small residence built and occupied between A.D. 550 and 870

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate o1 View from the temple at Mnajdra, Malta, looking south to the island of Filfla. The temples are now at least as early as the Pyramids of Egypt

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Plate ii The temple at Mnajdra, Malta - a decorated entrance

Plate 12 Decorated stone blocks at the temple of Hal Tarxien, Malta, with the legs of a colossal statue (at right rear). On the new chronology this is one of the earliest monumental sculptures in the world

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism

Figure 20 Sketch plan by Peter D. Harrison of the 'palaces' of the Central Acropolis of Tikal. Over i *6 ha. of largely late classic (A.D. 550-870) buildings are depicted. Scale x :I500

An attempt to estimate the population of Central Tikal demands control of certain variables. The first group of these is time: how many houses were actually lived in at once? Combining data from small structure excavation with that from a programme of ceramic test pits, there is available chronological information for about Izo residential groups (exclusive of major 'palaces') which appear on the map (Carr and Ilazard I96I) of the central 9 km.2 Since dates are dependent on ceramic data, a possibility which must be considered is that houses may have been used, abandoned for a significant length of time, and then rebuilt (Adams I969: 23; Sanders and Price I968: I65). This hypothesis is not consistent with four lines of evidence. First, structures were frequently altered during spans of time with no apparent change in ceramics. For example, the small Str. 6E-26 was built and then modified five times, all within the two centuries during which Imix ceramics were in vogue. This case would suggest alteration every thirty or so years. There are other structures which suggest time intervals between alterations of ten to forty years. Second, when a small structure was altered, portions of older floors and walls commonly remained in use. Had the structure stood abandoned for any significant length of time, exposed to the rigour of a hard climate, the component walls and floors would have deteriorated and not been in any condition to be reused. Third, there are a number of middens which show continuous deposition. Had the associated architectural de- velopments stood abandoned for any length of time, soil would have accumulated over the midden prior to further deposition. This would leave as stratigraphic evidence layers in the middens, which in fact they lack. Finally, the rarity of disturbed burials in structure platforms, some of which contain as many as eleven, implies remembrance of their location on the part of the Maya over several generations. Such remembrance would not be expected in the absence of continuous occupation.

All mapped residential groups for which there is chronological information were in use FA

I9I

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

192 William A. Haviland

in the period of time between A.D. 55o and 870, even though the history of some reaches further back into antiquity than others. The sample is sufficiently large to state with confidence that all mapped small residential groups were in use then. However, this is a span of some 320 years, and it is important to know if these groups were in use at one point in this span of time. Apparently, not all groups were in use prior to c. A.D. 770. Conversely, some groups were abandoned before others. The excavated sample does indicate that at one point in time, roughly A.D. 770, all mapped groups were in use. By A.D. 870, some of them were abandoned, as were all by A.D. 1000.

It is of cardinal importance to know approximately how many of the individual structures within the mapped groups were in use c. A.D. 770. Excavation revealed that four mapped ruins were not structures at all, while five unmapped structures were dis- covered; a net gain of one. Of these, 95 % are judged to have been in use c. A.D. 770.

A complicating factor is the existence of a number of other structures which do not appear on the site map and which were not parts of mapped groups. Twenty-six such 'hidden' structures are now known, and it seems obvious that there are more to be found.

There is, unfortunately, no way to estimate how many. Without exception, these known 'hidden' structures were abandoned by A.D. 770. But these findings raise the question: does A.D. 770 represent the high point for Tikal in terms of small structure density as previously thought? Certainly, structures were far more numerous then than in Pre- classic times (prior to A.D. 250). It is also clear that structure density increased in Early Classic times (A.D. 250-550); although how much is uncertain. While it is impossible to estimate just how many Early Classic 'hidden' structures remain to be discovered at Tikal, this is offset by the fact that not all such structures were in use at one time. How- ever, the indications are that almost as many groups of small structures were in use between A.D. 550 and 650o as by A.D. 770. Therefore, contrary to my statements in earlier articles, it appears that structure density reached a peak by A.D. 550. Then, for just over two centuries, the abandonment of existing structures was offset by construction of new

ones, so that overall density remained constant (plate 9). The second important variable which must be controlled is function: how many small

structures were actually houses? Criteria for judging small structures to have been houses include: abundance; resemblance to known modern and historic houses; presence of utilitarian and domestic objects in occupation context; presence of midden deposits, ideally containing bone and ash; presence of non-ceremonial burials; association with out-buildings (structures too small for living purposes, or with some evidence of special- ized function, such as kitchens); and the lack of evidence for any other function. Diffi- culties arise, first, from the fact that the modern Maya build structures that directly resemble houses, but which are used for a variety of other functions (for example, store- houses). The Classic Maya may have done the same. Second, the Classic Maya drew on household refuse for fill in the construction of major ceremonial structures. Third, animal bones are scarce, but this can be explained in three ways: meat was evidently not common in the diet, dogs probably scavenged any bones which lay around, and many bones must have decayed with time. Ideally, one would wish to find a ruin with fireplace and household gear left where the owners commonly kept them in the house. However, people do not usually leave serviceable items behind when they move, unless the house burns down or a similar catastrophe occurs. Fortunately, in one instance part of a

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism 193

kitchen assemblage was found in a small, square, out-building, and a few good middens have been identified. These serve to strengthen the diagnosis of residential function for most small structures. Indications of ceremonial activity permit others to be identified as small shrines. All things considered, I estimate that approximately I6% of all small structures at Tikal were residential adjuncts, rather than bonafide houses.

Consideration of all these matters together permits an estimate of the peak population of Tikal. Allowing 5 6 persons per house, on the basis of present-day household statistics in modern Maya communities in Yucatan, this gives a figure of about Io,ooo persons for the central I6 km.2. Since this is based on small structures alone, and excludes the major 'palaces', this must be regarded as a conservative estimate. For the entire central zone of Tikal, population is estimated at about 39,000, an overall density of 600-700 persons per km.2 Considerable uninhabitable swampland is included here, so density per usable land would be much higher (fig 19).

Peripheral Tikal was never as densely settled as Central Tikal. The evidence from the strip surveys, coupled with excavations in and around 104 of the structures, indicates much more of a tendency to shift residence from one locale to another. A density of Ioo persons per km.2 for this zone raises the estimate of the peak population of Tikal to about 45,000. Here, it must be emphasized that the data on which this figure is based will ultimately be thoroughly reassessed, but the figure at least indicates the magnitude of a final population estimate for Tikal.

The implications of the evidence summarized so far are quite clear: starting from a small, dispersed population in the last millennium B.C., the population of Tikal swelled to large proportions by A.D. 550. While it may not have equalled that of Teotihuacan, it compares favourably with R. M. Adams' (i966: 69) estimate of 50,000 for ancient Sumerian cities. At first, there was some tendency to shifting residence, but by the time the peak population was reached, Tikal had achieved a degree of nucleation which was maintained for at least two centuries. Although not so compact nor carefully planned as Teotihuacan, the density of Central Tikal was six times that of the surrounding region, and nucleation is, after all, a relative thing.

Interestingly, the tendency to nucleation is seen on a smaller scale in the surrounding

region (Fry I969). Here, however, the development was different from that of Tikal, where nucleation seems to have come about through steady population growth. Beyond the periphery, around A.D. 550, there was a redistribution, freeing from habitation the land with the best agricultural potential, with concentration of population into small nucleated centres and the fringes of swampland. This change had an abruptness not seen in Tikal.

Tikal certainly meets two of the stated criteria for urbanism: large population and nucleation. What, then, of the third: socio-economic diversity? Most would agree that the first inhabitants of the Maya lowlands were milpa agriculturalists, growing the traditional corn, beans and squash. The general assumption is that this remained the basis of Maya subsistence, as it is for the Maya today. The traditional point of view is that this does not provide for large and stable population concentrations, particularly of non-farmers (Haviland i966: 36). Recently, a more optimistic appraisal has emerged of the possibilities of milpa agriculture in the Maya lowlands. Cowgill (i962) now suggests that a population of 38-77 persons per km.2, some of them not farmers, might be sup-

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

194 William A. Haviland

ported in the Tikal region. Her argument, however, is challenged by Reina (I967), who demonstrates that in at least one native community south of Tikal a family has difficulty getting sufficient food to meet its needs.

As is often the case, the truth probably lies somewhere between these extremes. While Reina's objections to Cowgill's argument seem valid, it also appears that the relative efficiency of milpa agriculture is influenced by the way people organize themselves (Haviland I968b). Apparently, the Preclassic Maya of Tikal overcame the organizational difficulties, as indicated by evidence for a steady population growth.

Although the bulk of the Preclassic Tikal Maya were very probably agriculturalists, a substantial segment of the population c. A.D. 550 was not. Then, within central Tikal, there was little more than a hectare of land for each house suitable for agriculture; certainly not enough to support a household through milpa agriculture. Moreover, some- one living, say, on the south edge of the site map would have had a 5 km. walk just to reach the peripheral zone. He could have done so, except that there were others living on the periphery who were more favourably located with respect to potential milpa areas.

On the other hand, the residents of the central zone did produce some of their own food. A significant association of Breadnut trees (Brosimum alicastrum) with house ruins indicates that each household had growing a number of such trees. This required very little expenditure of time and energy for a substantial return of highly nutritious food (Puleston I968). Therefore, ample time would have been available for other econ- omic activities. There are a number of clues as to what these activities were.

Evidence of trade in non-perishable items is clear at Tikal. Granite, quartzite, hematite, pyrite, jade, slate and obsidian were all imported, either as raw materials or finished objects. Marine materials came from coastal areas. Tikal itself is located on a source of abundant flint, which may have been exported in the form of raw material and finished

objects. The site also happens to be located between major river systems to the east and west, the Rio Hondo and Rio San Pedro, and so may have been strategically positioned on a major overland trade route between the two. Analysis of ceramics in peripheral Tikal and beyond indicates that utilitarian ceramics were produced in a few specialized

centres and traded locally, while decorated ceramics were traded widely (Fry i969). Distribution was probably through regional markets. Some pottery, sculpture, and three ceremonial platforms testify at least to the exchange of ideas between Tikal and Teoti- huacan, or a Teotihuacan-affiliated entrepreneur. What sort of flourishing trade went on in perishable goods such as textiles, feathers, salt, or cocao must remain unknown. All in all, it is hard to believe that there were no full-time traders among the Tikal populace.

In the realm of technology, the fine stonework and artistry speak for themselves; specialists seem required. Moreover, in some groups of dwellings in the central zone, but never in the peripheral zone, evidence of flint and obsidian workshops have been found. The massive architecture of the nucleus of Tikal also required specialists, at least to plan the work and direct its execution. The complex calendar required astronomers. A large population must have required some sort of bureaucracy. And so it goes. Moreover, there are no archaeological indications of such occupational specialists, who might have been part of the economic fabric, as weavers, butchers, curers, brewers, embalmers, tailors, and so on. Yet, such specialists may be found in small Indian communities in Mesoamerica today.

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism 195

Along with evidence of economic differentiation, there is evidence of social differenti- ation. An elite hereditary ruling class apparently emerged in the last century B.C. (Haviland I968a: 107). Stratification is suggested by the range of domestic architecture from simple to ornate already mentioned. Such a range is already apparent in Preclassic times, and becomes more complex in Early Classic times, when major 'palaces' appear. Analysis of human skeletal material indicates a more favourable nutritional environment for those individuals who were buried in tombs beneath the major temples. Over I20 burials from domestic contexts display considerable variation in the quality of grave, orientation, position, and the presence, quantity, or absence of pottery vessels. The burial accorded to any one individual very probably was determined in part by his rank in society.

The great quantities of excavated artefacts, some of them from middens, add to the theme of variability. There are problems here, though, as most household trash has been mixed with earlier material from structure fills through erosion. Moreover, like modern peoples, some of the inhabitants of Tikal were probably more careless than others and broke more objects, or took more care to clean up their living areas. Nonetheless, some people appear to have been able to afford luxury items, such as tools of imported obsidian rather than native flint alone. Obvious luxury items such as shell ornaments, metallic mosaic ornaments, and jade beads were found in varying quantities from one group to another. In general, the abundance and variety of artefacts tend to be greater in groups of larger structures and less in groups of smaller ones.

The logical explanation for all this architectural, mortuary and artefactual variation is that the population of Tikal was not divided simply into a ruling class and a peasant class. The existence of the former seems real enough, but for the greater part of the population more than a simple two-class system seems indicated. This is what might be expected in a large population with economic specialists of one sort or another.

In sum, there seems no reason to doubt the urban nature of Tikal, even though it may

not have approached the degree evident at Teotihuacan. To do so is to allow typological considerations to obscure important developmental processes in Maya society. Nor does it seem warranted at this point to attribute Tikal's urbanism to Teotihuacan influence. Evidence of some kind of Mexican influence does appear by the early fifth century A.D., first in the form of minor decorative elements on monuments, and thereafter as occa- sional decorative and stylistic motifs on pottery. In the fifth century, implements of green obsidian and clay ear spools were imported from Mexico. Aside from three examples of Mexican-style architecture, which might be two centuries later, outside influence subsided with the sixth century. There is certainly no indication of a Teoti- huacan 'takeover', in spite of the presence of a monument depicting as its principal subject the Mexican rain god (Stela 32). This monument is conspicuous by its exceptional character, and on another stela (No. 31) Mexican figures are clearly subordinate to a Maya. This all has the appearance of periodic foreign elements coming into a strong Maya cultural tradition, as opposed to a sudden, overwhelming influx. While this may

have played a part in the trend towards urbanism, this trend was, I think, already established by the fifth century A.D.

It is evident that the urbanism of Tikal owes nothing to hydraulic agriculture. Unfor- tunately, discussions of hydraulic systems of agriculture have so stressed waterworks

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

196 William A. Haviland

per se, that the fundamental point involved is often obscured: central control of important resources. This can occur in the absence of irrigation, which, I suspect, was the case at Tikal. While the final answer to the causes of Tikal's growth is not yet to hand, the follow- ing speculation is consistent with the evidence as now known. A strong religion among the early Tikal Maya was likely, in order to have countered the uncertainties of life. Here soils are thin, and there is no water save that which can be collected in ponds. Although the rain is abundant in season, the onset of the rains tends to be unreliable. Because of

this, the inhabitants of the region today are under tension at this time of year (Reina I967: 4-5). It seems evident that the Maya priesthood, with its great concern with calendrical matters, probably tried not only to placate the gods, but offered predictions as to the proper times to plant crops and other matters of concern to agriculturalists.

Given an efficient adaptation to the lowlands, a population increase developed. With milpa agriculture, this would have tended to have had a centrifugal effect on population distribution. Here, however, the Maya may have been loath to move far from Tikal on account of its religious significance, not to mention its supply of flint. The latter would have been valuable not just for tools, but as a commodity to exchange for jade, obsidian, and quartzite which were already being imported by 500 B.C. Such a situation could gradually have led to a shortage of agricultural land, with increasing development of local resources to supplement milpa agriculture. Experimental evidence indicates that chultuns, subterranean chambers, served to store the breadnut. These appear shortly before 200 B.C., so this food source was evidently being tapped by then. This may have made possible the formal plaster and masonry architecture which appeared by 2oo B.C.

Incidental by-products of increasing reliance on the breadnut would have been a further population increase, and an increase in tempo of population growth in the region does become evident by A.D. I, as well as the freeing of substantial numbers of people from full-time agriculture. Thus, there could be more occupational specialization, which most likely catered to the needs of, and was controlled by, the emergent upper class. The attendant redistribution of commodities would have been most efficient in a population centre, which would have tied specialists to Tikal. At the same time, milpa agriculture did continue, but was more and more relegated to the periphery. One may well imagine the need for control to mediate disputes over fields and to regulate the shifting of plots in a context of increasing land scarcity.

Perhaps, too, waterworks should not be discounted entirely. In Central Tikal water was provided in a number of artificially constructed reservoirs, which would have become more and more important as Tikal grew. Undoubtedly, there was control of this system in order to ensure a supply of water for everyday needs.

Eventually, the pressure for food and land reached the point of no return, and so population growth stopped. This event is marked archaeologically by the pullback from prime land, the advent of nutritional problems as evidenced by the bones from burials, and perhaps by construction of the earthworks. In other words, a period of readjustment set in, which must have been directed by an already strong central authority. Activities then continued as before, minus further population growth, for another 300 years or so.

I do not wish to deny that the Tikal Maya were subjected to outside influences; the evidence is otherwise. However, these appear to have been secondary in importance to factors such as those I have suggested here. I envision a system of feedback, with these

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

Tikal, Guatemala and Mesoamerican urbanism I97

and other factors mutually re-inforcing one another. Hopefully, when analysis of all the Tikal data are complete, these things can be discussed with greater assurance. This will not be possible, however, if there is continued adherence to an outmoded typology which labels tropical lowland civilization as non-urban, and so obscures factors which in fact appear to have been of prime importance in the growth of that civilization.

I6.iii. 970 University of Vermont

References

Adams, R. E. W. I969. Maya archaeology I958-I968, a review. Latin American Research Review. 4 (2):3-45.

Adams, R. M. I966. The Evolution of Urban Society: Early Mesopotamia and Prehispanic Mexico. Chicago.

Carr, R. F. and Hazard, J. E. I961. Map of the Ruins of Tikal, El Peten, Guatemala. Philadelphia.

Childe, V. G. 1951. Man Makes Himself. New York.

Coe, W. R. I965. Tikal: Ten years of study of a Maya ruin in the lowlands of Guatemala. Expedition. 8:no. I.

Coe, W. R. 1967. Tikal: A Hiandbook of the Ancient Maya Ruins. Philadelphia.

Cowgill, U. M. 1962. An agricultural study of the southern Maya lowlands. American Anthropo- logist. 64:273-86.

Fry, R. E. 1969. Ceramics and settlement in the periphery of Tikal, Guatemala. Doctoral Disserta- tion, University of Arizona, Tucson.

Haviland, W. A. 1965. Prehistoric settlement at Tikal, Guatemala. Expedition. 7 (3):I5-23.

Haviland, W. A. 1966. Maya settlement patterns: a critical review. Middle American Research Institute. Pub. 26:21-47.

Haviland, W. A. 1967. Stature at Tikal, Guatemala: implications for ancient Maya demography and social organization. American Antiquity. 32:316-25.

Haviland, W. A. I968a. Ancient lowland Maya social organization. Middle American Research Institute. Pub. 26:93-117.

Haviland, W. A. i968b. Comment on 'Milpas and Milperos'. American Anthropologist. 70: 564-5.

Haviland, W. A. 1969. A new population estimate for Tikal, Guatemala. American Antiquity. 34:429-33-

Puleston, D. E. 1968. Brosimum alicastrum as a subsistence alternative for the Classic Maya of the central southern Lowlands. M.A. Thesis, University of Pennsylvania, Philadelphia.

Puleston, D. E. and Callender, D. W. Jr. I967. Defensive earthworks at Tikal. Expedition. 9 (3):4o-8.

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms

198 William A. Haviland

Reina, R. E. I967. Milpas and Milperos: implications for prehistoric times. American Anthropo- logist. 69: 1-20.

Sanders, W. T. and Price, B. J. 1968. Mesoamerica: The Evolution of a Civilization. New York.

Abstract

Haviland, W. A.

Tikal, Guatemala and Mesoamerican urbanism

A common assumption is that in Precolumbian times, an urban form of civilization developed in the highlands of Mesoamerica, while a non-urban form developed in the tropical lowlands. However, excavations at Tikal, one of the largest Maya sites, questions this assumption. The results of these excavations are summarized, and suggestions are offered which might help to explain the emergence of urbanism in the lowlands.

This content downloaded from 204.168.144.64 on Wed, 22 Jun 2016 16:35:19 UTC All use subject to http://about.jstor.org/terms