Excavations in El Salvador

By: William R. Coe

Originally Published in 1955

View PDF

One facet of the Museum’s long-standing interest in Mesoamerica is its present concern with the prehistory of El Salvador. Interest has centered primarily on the Pre-Classic horizon, a cultural period amply evidenced in such adjacent countries as Honduras and Guatemala, but, until now, subject to little controlled excavation in El Salvador itself. The later Classic and Post-Classic horizons in the country are of course very much better known, almost entirely through the work of Stanley H. Boggs at the extraordinary sites of Tazumal and San Andrés.

Intensive work by the Carnegie Institution of Washington in the nearby Guatemalan highland region has yielded a long and fascinating cultural sequence of Pre-Classic occupation, beginning perhaps as early as the second millennium B.C. and ending during the first few centuries of our era. Typical Pre-Classic traits there include a lack of cut stone architecture, dependence on adobe in construction, highly developed monochrome and bichrome potteries, one hallmark being the curious decorative technique known as “resist,” also quantities of handmade figurines, remarkably little chipped stone, and few graves except for an occasional fantastically rich tomb. Elaborate ritual, a stratified social structure, and theocratic control were features already present, being essentially based on the surplus from maize agriculture. As its name indicates, this far-from-simple horizon was preliminary to the florescent Classic period that followed, developing and anticipating much that culminated in that later era. The Museum’s work at Piedras Negras and recently at Caracol illustrates many of these Classic accomplishments.

In an attempt to appreciate something of the Pre-Classic as it might be represented in the country, the relatively small site of El Trapiche was selected for work. Dr. A. V. Kidder, long active in the Guatemalan highlands, had visited the site with Mr. Percy C. Madeira, Jr., and, on the basis of surface sherds, had decided that it had seen Pre-Classic occupation and was thus a good place to start.

Four people digging out dirt from between clay walls.
Removing the dirt between the clay parts. Later the structure was removed meter by meter to clarify details of construction. (Photographs by the author.)
Image Number: 243714

Accordingly, I arrived in the country the first week of February, 1954, and, as soon as possible, “camp” was set up in Chalchuapa in a pleasant but often boisterous hotel in the far western part of the country. Chalchuapa, besides being a town, forms a rather large archaeological zone, the most important site of which is Tazumal, one of the few tourist attractions of the country. All these various sites are concentrated in a well-watered, richly soiled valley, at an altitude of about twenty-three hundred feet. The whole area now is very profitably devoted to coffee cultivation. The only active volcano in the range separating this broad valley from the Pacific is Izalco, frequently heard but not seen from Chalchuapa itself. As might be expected, the winter months are delightfully dry but with strong dusty winds in the afternoons. April usually sees the first of the rains and May is quite intolerable, the rains lasting until November. Work this season continued till the end of May and I learned that accuracy and enthusiasm tend to wane under drizzles and thunderous rains.

Our excavations were carried out under the most pleasant conditions possible. For one, the mounds at the site had been cropped clean by cattle. The owner of the site, Don Abram Perdomo, was most generous with his coconuts, oranges, bananas and pineapples. One mound in particular was topped by a magnificent shade tree which, when supplied with a hammock, made the custom of siesta quite understandable. And, across the road from the site was a river-fed swimming pool of Olympic proportions-a project of a former president of local origin. The ten workmen appeared every morning at the site around seven o’clock to begin the day’s work under the direction of a most able and conscientious foreman. Due to peculiarities of the hotel I rarely was able to arrive myself before eight o’clock. Mr. Arthur Stitzinger, a Museum member, joined me for two weeks early in the season and helped greatly by scouring the site’s surface for interesting objects. Later my wife arrived to help with the by then heavy problem of numbering and cataloguing the excavated material.

The site is composed of eight featureless mounds and the whole is fairly compact. The largest, Mound 1, about eighty feet high, rests to the north and beside it to the west is Mound 6. Across from these, to the south, are three mounds (2, 3, 4) with Mounds 5 and 7 to the west of these. Mound 8 is the most distant from the group and archaeologically might be considered as part of the adjacent Finca Casa Blanca group. Mounds 3 and 6 were excavated as far as was possible, while Mounds 1, 2, and 8 were simply test-pitted. Also various stratigraphic testings were made around the area, and these were supplemented by extensive surface collecting. Almost the entire season was spent at this one site. Towards the end of May I was able to visit other sites in the zone, and eventually a two-day dig was made on the shore of a local lake.

The high point of the season was Mound 3. Hoping to locate a stairway or facing-in fact, anything to orient the digging-we started from the north and dug a long, broad trench into the near center of the mound. This, however, revealed nothing save a clamp earth fill. The mound, which at that point seemed to be utterly featureless, was found to rest on a thick layer of blue clay overlying a soft volcanic stone of unknown depth. The lack of constructional features in the trench was frankly disappointing. Still there was compensation in the finding of tremendous quantities of fine potsherds, broken figurines and other artifacts. Most surprising was the discovery deep inside the mound of two small carbonized corncobs. Burned earth patches here and there provided samples of carbon which I hope may be dated in the Museum’s carbon-14 laboratory. It was towards the middle of April that we began the last section of the trench, which, had it been completely dug, would have taken us to center of the mound. Digging down only a short way we came upon a thin blue clay line set vertically in the earth fill. It required a month and a half of further digging to follow out this lead. Because of the line’s position and soon baffling meanderings, we were forced to follow it by removing both the northwest and southwest quarters of the mound’s summit. Clearing of these areas first revealed a mass of formless adobe which later was found to overlie a deposit of boulders and fragmentary milling implements. Below this was a layer of earth, and, directly below, the clay lines themselves. These lines were confined entirely to the west half of the mound while the clearing of the southeast area of the mound disclosed the same thick layer of adobe (as well as a firepit) and very little else.

Mound 6 mid excavation, surrounded by trees, men working.
Mound 6 at El Trapiche, talken from the summit of Mound 3, showing work in progress. The north half of the mound was left unexcavated.
Image Number: 243712

Isolating and determining the relationships of these clay lines was made very difficult by their extreme delicacy, rain at any moment of the day, and the inconsiderate visits of local livestock. Large piles of corn stalks and palm leaves and yards of oilcloth were kept on hand, and each evening it was necessary to cover everything and surround the area with barbed wire. That this unique structure was ever cleared is due to the interest and patient care of the workmen.

A little over sixty feet long, on roughly a north-south axis, the structure was found to be composed of parallel clay channels varying in both depths and levels. Parts along its east side were rectangular. Serpentine patterns and complex junctures, basins, troughs, and even a flat spout on the north end, are major features of this strange construction. It is really necessary to depend on illustrations to give any substantial idea of it. The whole had been intentionally filled with a gritty earth that contained many sherds, pieces of obsidian and so forth. In excavating this fill a few broken vessels were found resting either in the fill or on the clay floors of certain portions. The structure was thoroughly cleared but to better understand how it went together it became necessary to remove meter sections of it, working from north to south, drawing and photographing all features in the process. This disclosed many complicated elements including a much narrower and shorter buried clay structure. The clay used throughout was for the most part blue, identical to the clay found beneath the mound; occasionally a cream-colored clay was used. Carbon was collected here and there but no portion of the structure actually appeared to have been burned. Beneath the whole was the same unremarkable fill which formed the remainder of the mound.

No obvious clue was found to explain the function of this perplexing structure. Exposed, without its peculiar fill, surely it could not have withstood many days of sun and rain. It apparently was built by the following steps: first, the surface of the earth mound was leveled, the earth was removed in long parallel furrows, large rectangular pits were excavated, and, finally, these excavations were thinly lined with clay. Then, or possibly later, everything was filled with earth and rubbish, pottery vessels having been placed in various parts of the structure. Eventually, the whole was buried beneath a layer of earth, a mass of boulders and broken grinding implements, and, finally, topped-off with a layer of adobe.

The curious thing is that Mound 3, where this clay construction was found, was quite unlike the other mound excavated, Mound 6. While the former was solidly composed of earth, Mound 6 produced a sequence of once smooth adobe floors, also a deeply buried adobe platform. No clay occurred in Mound 6. Both mounds shared identical pottery and other artifacts and neither mound, as far as they were investigated, contained burials.

Mound 3 with large trench dug in the middle, men working.
The last section of trench dug to the center of Mound 3. One edge of the clay structure has just appeared. In following it out we removed the whole upper right-hand portion of the mound.
Image Number: 243713

As regards the El Trapiche artifacts, we were fortunate enough to find them in quantities that I believe amount to an adequate, representative sample. About one hundred thousand sherds were collected from the actual excavations and about five thousand surface sherds were brought in for selection; some thirty thousand eventually were saved for transportation and study. No preliminary work, save for washing, numbering and cataloguing, was clone in the field. Since the collections have not as yet arrived for study, I hesitate to say very much about the El Trapiche material. On the whole the sherds and figurines belong to the late Pre-Classic and to the earlier portion of that division. Many wares distinctly relate the site to others in the vicinity of Guatemala City and less specifically with Pre-Classic sites to the north in Honduras. Very common among the sherds were ones decorated in a negative or resist technique, one which involves the selective coating of the vessel with wax or some such substance prior to firing. A number of interesting local wares also appeared. It is a strange fact that among the thousands collected, not a single complete clay figurine was found-simply heads, torsos and limbs. A number of these were of the movable-arm type, and one torso evidenced the former presence of both movable arms and legs, all connected to the body via perforations and strings. Every shovelfull of earth contained fragments of obsidian, only a small percentage of which were recognizable as implements. Strikingly rare was jade, for El Trapiche produced only a few tiny beads. Other artifacts include such items as legless metates and one-handed manos, cylindrical and flat stamps, bark-beaters of both the oval and monolithic (mallet) types, and fragments of flanged green stone bowls.

While it is true that the four months spent at the site produced nothing very spectacular, a lot of basic information was gathered on the nature of the local Pre-Classic. The architectural data are regrettably meager and most perplexing. While sharing a number of fundamental features with contemporary sites in Guatemala and elsewhere, the clay structure, if nothing else, gives the site a touch of singularity. As already noted, the late Pre-Classic in the western part of El Salvador appears to have had relatively few contacts with western Honduras to the north, an area equally rich in Pre-Classic remains. There are also striking differences between the El Trapiche figurines and those seemingly Pre-Classic further to the east in El Salvador. Then, there is the problem of accounting for the emergence of Classic Tazumal. Architecturally, we found little at El Trapiche that might have anticipated and led to that later and adjacent florescence. Beyond these tentative remarks, I would hesitate to say anything more without actually studying the material itself.

Plans had originally included a reconnaissance of at least the eastern part of the country. It quickly became clear that to do so would require cutting short the work in Chalchuapa, which seemed inadvisable at the time. Should the Museum continue work in the country, location and excavation of appropriate Pre-Classic remains to the east would be of great value. However, it should be noted that the early portion of the Pre-Classic horizon, fairly well established in parts of Guatemala and Honduras, remains a supposition in the Chalchuapa zone. Without a good complete western sequence to refer to, moving east in search of more of this ancient horizon might be somewhat risky. And deposits to fill this gap will be hard to find.

Cite This Article

Coe, William R.. "Excavations in El Salvador." Museum Bulletin XIX, no. 2 (June, 1955): 15-21. Accessed May 29, 2024. https://www.penn.museum/sites/bulletin/3991/

This digitized article is presented here as a historical reference and may not reflect the current views of the Penn Museum.

Report problems and issues to digitalmedia@pennmuseum.org.