HomeBiographyWritingsPhotosContactAncient Maya Settlement Patterns and Environment at Tikal, Guatemala: Implications for Subsistence Models Dennis Edward Puleston A Dissertation in Anthropology, University of Pennsylvania, 1973

Chapter Eight: Subsistence Aspects of the Cultural Ecology of the Lowland Maya


The Invalidation of the “Sustaining Area” Concept
Alternative Subsistence Models


The Invalidation of the "Sustaining Area" Concept

Ursula Cowgill (1962:277) has estimated that slash-and-burn agriculture can sustain 39-77 persons/km2 (100-200 persons/mi2) in the southern lowlands as a whole. Sanders' (1962-3:237) estimate of 30-60 persons/km2 or 40-50 persons/km2 for the northern Peten as a whole, with 100 persons or more/km2 in "more favorable areas" (Sanders 1962:96), is approximately of the same magnitude. Clearly a significant inconsistency exists between the carrying capacity of the assumed slash-and-burn subsistence model and the settlement data from Tikal. Within the huge area of more than 120 km2 defined by the North and South Earthworks, the calculated population density of 538 people/km2 on all types of terrain (see Table 6), 325 people/km2 if we use Haviland's (1972) figures, exceed carrying capacity by a factor of ten or more. This aspect of the traditional model's unacceptability was realized before the present study became a reality, and it was exactly this problem which led to the formulation of the concept of "sustaining area." As revealed by the data presented here, however, the supposed sustaining area was not an unoccupied food producing area which balanced out the "overpopulation" of the heavily settled 120 km2 of Tikal itself. The intersite areas were also heavily settled and, in fact, probably sustained a population two to five times the carrying capacity of the locally available land under slash-and-burn maize agriculture (148 people/km2 in Late Classic times, see Table VI). Certainly then the intersite areas could not have served the subsistence needs of the even more densely settled areas within the huge site areas enclosed by the earthworks.

The slash-and-burn model, with "maize as the single major crop" (Tourtellot and Sabloff 1972:128) must be rejected. We are forced to accept the fact that there was no immediately adjacent "breadbasket" or "sustaining area" of sufficient size to produce the required maize by this method.

Alternative Subsistence Models

If slash-and-burn agriculture could not sustain the indicated population, the ancient Maya must have been sustained by some more intensive means of cultivation. Lundell, with considerable perception, observed four decades ago that During the time of the ancients, milpa agriculture in its present form was doubtless used in outlying areas of sparse population where there was    an abundance of land for this wasteful system. It is evident, however, that the Maya as a whole, or at least those that built up the great     Southern Culture, made advancements in agriculture beyond the primitive milpa type that survives today [Lundell 1933:76].

What alternatives were available? Though I have provided an initial discussion on the evidence for the use of the ramon in the previous chapter I wish to comment on a number of other possibilities.


It is often assumed that terracing serves simply to prevent soil erosion. This function is perhaps less than hal the story., however, for the construction of these features gives the farmer considerable control over soil conditions as well.

Terracing, where it occurs, is almost certainly a sign of intensive cultivation. The larege labor investment required for the consturction and maintenance of terraces argues strongly against the possibility that they were used for shifting cultivation as Sanders (1962:92) suggests. Small bush terraces may have been used fo fallowing agriculture, as they are by the Kofayar (1968), when the labor investment is small. Such a practive does not seem to be consistent with the data, however, in those portions of the Maya Lowlands where terraces occur. As discussed by Wilkin (1971), Thompson (1931:228) found "practically every hillside from Arenal to Mountain Cow" to have high stone terraces on it. Lundell (1940:9) "estimates that terraced region of eastern British Honduras ([Belize] covers 400 km2. " The quantity and size of these terraces certainly represent a large labor investment, particularly when maintenance is considered. Constructed on hillsides, they allow the farmer to create an "artificial soil profile" with properties not found on the orginal slope. This includes "control over drainage, moisture capacity, and fertility" (Hester 1954:89).

Terraces are generally faced with rough blocks of limestone and vary in width with the slope of the hill they are on. The only excavations of terraces I am aware of are those carried out by Saxe (ca. 1965). In the Maya Lowlands ancient terraces "occur al across the mountainous southern fringe of the Maya Southern Lowland subarea, from British Honduras in the east, across the mountain valleys of Guatemala to Chiapas on the west" (Hester 1954:86). They do not appear to occur in the region of Tikal, however. Even on the steep El Palmar ridges in the vicinity of Uaxactun there is no evidence of terracing. Why is terracing absent in this area where evidence of intensive cultivation would be expected? Perhaps because environmental conditions differed in some significant way, perhaps because other forms of intensive cultivation were more effective.


Best adapted to riverine locations where moving water can be channeled into areas of cultivation, irrigation does not seem to have been a significant aspect of intensive cultivation in the central Peten. Channels discovered by Bronson in Square 5F at Tikal appear to have been portions of a dendritic water-collecting system, rathe than a water-distributing system associated with the Tikal Reservoir. Canals and ditches which served alternate functions o drainage and irrigation have recently been found in association with ridged fields in the Candelaria Basin (Siemens and Puleston 1972) and northern Belize. Nothing comparable to this has shown up in aerial photographs or on the ground in the Tikal region.


Closely related to irrigation systems, chinampas and chinampa-like drained fields have been mentioned as a possible means of intensive cultifation in the Peten (Palerm and Wolf 1957:28; Wolf 1959:78). Wilken (1971:438) feels that the possibilities are "exciting" and supports his optimsim by citing evidence of lakeside ditched fields of the "lacandon," described by Squite (1858:556), and "the Mayan penchant for building cities near lowland swamps." With respect to the first item, the great difference between tintal or logwood bajo and a lakeside situation does not seem to be appreciated by Wilken. The often highly acidic soils of bajos pose one important barrier to the cultivation of many crops in them. Poor drainage and the seasonal alteration between saturation because of long term flooding, and drying, with deep cracking, pose another. The lack of any evidence for ridged fields like those found along the Candelaria River further removes the possibility that they were used around Tikal from the realm of likelihood.

High Performance Milpa

Wilken (1971:442) discusses vaious ways in which milpa can be made more productive. The main drawback to this, however, is the necessity of fallowing. Except in riverine areas where annually flooded levees can be kept in permanent cultivation, it is hard to get around this problem. Though Wilken points out that as many as four maize crops can be sown on a different piece of land resulting in no increment in the productivity per unit area. Some increase can surely be brought about by hand pulling of weeds and other means of more intensive care, but it is doubtful that these techniques are hgihly significant. The only really substantial altenative here is Bronson's (1966) hypothesis that root crops were intercropped with maize. This contibution represents an important step away from the virtually unquestioned assumption that maize was the primary staple of the Classic Maya.

In light of the indicated population density, Bronson concludes that a very large portion of the Maya diet was derived from root crops. With an overall population density in the Tikal National Park of more than 300 persons/km2 in classic times, a figure four to ten times the overall carrying capacity with straight slash-and-burn maize cultivation, root crops would have had to play a very important role. IN terms of nutritional considerations, Bronson's hypothesis is weak, as indicated by Sanders and Price (1968:93) who point out that in the Nigerian example he overlooks the role of intensive interregional trade in foodstuff by train and highway. This trade, sustained by a highly eeloped commercial economy involving cacao and palm nut, actually permits a more varied diet than would be indicated by study of what comes out of the gardens. the potein content of a diet limited to root crops is notoriously low, and, in view of the vigor and longevity of ancient Maya civilization, primary reliance on root crops seems unlikely. A secondary role for root crops, however, is certainly possible if not probable (O. Puleston 1971:11). Other aspects of Bronson's model have been questioned by Sanders and Price (1968:92-92) and U. Cowgill (1971). All point to the lack of evidence, documentary, archaeological, and palaeoecological, "that roots were ever staples anywhere in Mesoameica." Cowgill further assersts that soil conditions were largely unfavorable for extensive cultivation of potentially one of the most important root crops, manioc.

Dooryard or Kitchen Gardens

Dooryard or kitchen gardens have received very little attention in studies of Maya subsistence, both ancient and modern. Even in the work of those who have described and comented on them, whose contributions will be discussed below, consideration of these areas as significant food producing units has been totaly eclipsed by the image of maize. Morley, who exalted the role of maize, sets the tone of his classic work, The Ancient Maya, with a quotation from a 16th century manuscript that reads:

If one looks closely he will find that everything these Indians did and

talked about had to do with maize; in truth, they fell little short of making a god of it. And so much is the delight and gratification they got and still

get out of their corn fields, that because of them they forget wife and

childred and every other pleasure, as if their corn fields were their final

goal and ultimate happiness (Morley 1946:2).

It is on such a basis that almost a century o Mayan studies has been predicated.

But a closer look at the record reveals a far more complex picture involving many cultigens and even transplanted "wild" species. I have discussed and presented a small portion of the available evidence in another place (Puleston 1968:59-67). While maize was certainly the principle crop even as far back as Conquest times, the role of other crops including trees was hardly insignificant. Landa (1941:198-200) lists many species of fruits and vegetables cultivated by the Maya generally in the immediate vicinty of their houses. Evidence for the importance of these gardens at that time, and particularly the trees, comes from numerous references to the Spanish stratagem of destroying village orchards and kitchen gardens as a means of discouraging potential returnees in populations that were victims of forced resettlement schemes.

In accordance with Governor Figueroa's instructions the ruit trees at Tzuctok and Ichbalche were cut down and the houses burned to discourage the Indians from returning to their old locations (Scholes and Royes 1948:288).

Use of the same technique is described in the Relacion of Dzonot (Tozzer 1941:72).

And this Tomas Lopez was responsible for the moving and for the death of so great a number of people as have died, because the Indians say that since they ordered them to move by force and burned their houses and cut down their fruit trees which they owned, it aroused such a great sadness in their hearts that they died, . . .

The Relacion de los Pueblos de Chuaca y de Chechimula with a similar account confirms the supposition that these trees were in the context of kitchen gardens rather than orchards outside the village: " . . . Likewise he ordered them to set fire to all the furit trees which they had behind their houses in the said town: (translated by Wauchope 1938:133).

For the Peten we have at least one fairly certain reference to kitchen garden arboriculture in the 17th century. Father Avendano (ca. 1910::110) provides this important datum in the description of his departure from Yalain, a town apparently west of Lake Peten, " . . . an indian . . . guided us to the other farms, half a league from there, which , from the abudance of the furit, appeared an orchard." In this passage it is farily clear that it is the "farms" rather than the half a league of trail that was like an orchard. The overall effect of kitchen gardens on settlement patterns is fortunately recorded in Lopez de Gomara"s brief but lucid sixteenth century description of the Chontal Maya town of Potonchan which he says, "covered a large area, for the houses were separated from one another by gardens" (Scholes and Roys 1948:37). this was not an isolated instance, as is revealed in a letter Montejo wrote to the King of Spain where he stated that ". . . all the towns are [veritable] fruite gardens" (chamberlain 1953:66).

Preconquest archaeological evidence for kitchen gardens, apart from that provided by settlement patterns and the modern distribution of the ramon is thin but at least consistent with the picutre that emerges from the ethnohistorical record. Perhaps the best example is that provided by the famous murals of Temple of the Warriors at Chichen Itza. Here on the huge mural of a "Maya seacoast village" houses are widely spaced and interspersed with what may be up to five different kinds of tress (Morris 1931).

The details of Maya kitchen gardening and arboriculture do not emerge until we get into modern ethnographic data. Even here treatment is almost always sparse, probably because anthropologists tend not to be botanists or horticulturalists themselves and therefore have a minimal basis for communication on these subjects with their informants. They are easy to overlook, as Edgar Anderson (1969:137) points out in his classic description of a kitchen garden he studied in Guatemala.

It was covered with a riotous growth so luxuriant and so apparently planless that any ordinary American or European visitor, accustomed to the puritanical primness of north Europena gardens, would have supposed (if he even chanced to realize that it was indeed a garden) that it must be a deserted one. Yet when I went through it carefully I could find no plants. which were not useful to the owner in one way of another. There were no noxious weeds, the return per man-hour of effort was apparently high, and I came away feeling that as an experienced vegetable gardner (I am on of those strange people who would rather hoe vegetables than play golf or fo to the movies) I had gotten more new ideas about growing vegetables than from visiting any other garden anywhere.

Despite these difficulties kitchen gardens are reported for most of the Maya area (Wilkin 1971:441) and much of the rest of Mesoamerica from Mexico (Hester 1954:95ff), down through Costa Rica (Wagner 1958:215), to Panama (Gordon 1969).

One of the most useful anthropological studies of kitchen gardens is that done by Wisdom (1940) for the Chorti of Guatemala. His study reveals not only the geat variety to be found in Maya kitchen gardens but also many aspects of the complexity involved in their maintenance. The Chorti are constantly transplanting useful and ornamental trees into the kitchen gardens around theri houses. The technology involved can be quite sophisticated as in the case of the Tajin-Totonac who are adept at grafting (Kelly and Palerm 1952:141). This is not to say that all these skills are of New World origin but it does seem to confirm the supposition tha tkitchen gardening is a subsistence skill of long standing and considerable importance. The practically instantaneous acceptence and wild-fire spread of the use and cultivation of bananas and planains upon their introduction by the Spaniards (Tozzer 1941:199) seem to provide indirect evidence tha thorticultural skills were similarly well-developed at the time of the Conquest.

Wauchope in his book Modern Maya Houses (1938:129), includes a sketch plan of a typical Mya back yard in which the presence of fruit trees is indicated. He points out the Yucatecan's love of their furit trees and the fact that a wealthy man may have many trees on his property. This does not exclude the ramon. Lundell (1938:41) notes that , "In every village of the modern Maya in the Yucatan Peninsula it [the ramon] is one of the most conspicuous trees, being planted in dooryards, along fences, and in the streets."

On the basis of modern examples, we may imagine that ancient Maya kitchen gardens sustained a broad variety of trees, shrubs, annuals, and root crops growing in tumultuous profusion. A view from a temple top 1,200 years ago out over residential areas would probably have revealed a mantle of green not unlike the view today. Perhaps the only obvious difference would be the wisps of smoke and the occasional glistening roof comb of a taller temple.

The gardens themselves would have sustained many species. We may imagine an upper story of tall trees that proveded fruit, timber and other products--including the ramon (Brosimum sp.), chico zapote (Manilkara zapota), mahoganey (Swietenia macrophylla), Ficus sp., Sabal sp., and Protium copal. Many others would, of course, be represented. While some of these may have been allowed to grow into real giants up to 120 feet tall, the overall height of the upper canopy was probably nearer 50-80 feet because of the shortness of many cultivated species and the likelihood that trees would have been removed once they had passed their peak of productivity and had begun to pose serious potential threats as storm hazards.

Essentially, all these trees would have been useful in some way. No particular garden could have contained more than a small percentage of all the useful large trees (see Table X), and the distribution and utilization of certain species may have been regulated by social, economic, political and religious factors.

Beneath the upper story of taller trees there would have occurred most of the soft fuit producing species including the avocado (Persea), mamey (Calocarpum mammosum), the many species of Annona which include the cherimoya and custard apple, Lucuma, Pseudomedia and so forth (see Table X), as well as smaller specimens of species found in the upper canopy. Here also we could expect to find the smaller timber trees regularly used in construction poles for house frames and wall uprights. Roof construction could be obtained from specially maintained specimens of allspice (Pimenta officinalis), uapake (Dialium guiananse), habin (Piscidia piscipula), sapotillo (Lucuma sp.), sastante (Xylopis frutescens), and many others. Certainly many of these trees served more than one purpose by providing shade, furit, poles, perhaps a useful resin, bark that could be converted into coth or cordage, and, of course, firewood. Pruning probably produced a constant supply of material while increasing the vigo of the trees that were so cared fo. Useful materials could be obtained from areas of escoba bajo where the botan (Sabal sp.) which produces the guano leaves used for thatching grows. For an item in such high demand as the latter, utilization must have been regulated. The lack of any outlying hinterland other than the bajos would have resulted in a very different form of exploitation than exists today. When a new house was required, raw materials that could not be obtained from a family's own kitchen garden must have been obtained from that of someone else. When an older tree had to be cut down in someone's garden, it is unlikely that much of ti went to waste even if the immediate owner of the garden did not have use for such a quantity. Distribution of localy intermittent resources of this nature may have been partially served by a market system of some kind, but kin-based exchange relationships and systems of resource allocation within kin groups may have played a major role, too.

The degree of utilization of tintal, or logwood bajo also deserves consideration while we are on the subject of regulation and resource distribution. While it is improbable that these areas were used for agriculture, it is almost certain that they were exploited for firewood.


Go Back to Top of Page

PagesNotesPrefaceIntroductionCh 1Ch 2Ch 3Ch 4Ch 5Ch 6Ch 7Ch 8Ch 9ContactVoivode Multimedia