Analysis of Archaeological Artifacts from Cotzumalhuapa, Guatemala

Annex I.  Preliminary Report on the Artifact Analysis of the Obsidian Workshop at El Baúl
by Edgar Carpio Rezzio

Presentation

This report presents the most relevant aspects of the analysis of obsidian artifacts recovered from the excavations conducted in 2002 at the site of El Baúl, specifically in operation EB9 where the remains of a probable large workshop involved in the production of obsidian tools were explored. Works were initiated in March 2003 and concluded in March 2004.  During the year, the classification of artifacts was carried out according to the established criteria and with the participation of students from the Universidad de San Carlos de Guatemala working under my supervision. The attributes of each artifact were written down in a classification sheet. Simultaneously, a database using the Microsoft Access program was created, so that students could enter the information contained in the classification sheets. With this database a number of tables were elaborated to show the attributes of the sample and the more significant trends of the material.

Methodology

The first step consisted in designing the methodology for the classification and analysis of the several thousand artifacts recovered from the obsidian dump at El Baúl. To this purpose, a preliminary revision of the sample was accomplished to establish the variety of artifacts, the possible provenience of raw materials, and other aspects related to their form and function. Based on this preliminary analysis, a manual including the major variables to be analyzed and their respective codification was elaborated. The established variables were as follows:

Case number
Operation
Lot
Source
Type
Segment
Use
Retouching
Length (cm)
Width (cm)
Weight (grams)

This classification was based on the one used previously for other samples of the south coast, modified according to the characteristics of the material. We tried to outline the technological aspects inherent to the artifacts, which may help to reconstruct the production sequence. The complement was the creation of the classification sheet that included the annotation, case by case, of each one of the artifacts tested. In addition, a notebook was also used to write down and integrate any additional information concerning the artifacts that exhibited special traits and to elaborate the relevant drawing.

The analysis began with the classification of the sample recovered through shovel tests, each of which contained a large amount of artifacts. We soon noted that the dominant types were those associated with carving remains, represented mainly by refuse chips with a maximum size of 0.5 cm.  Another type also very frequently observed was the refuse of bifacial flakes, which differently than common refuse, exhibit a special, rounded aspect with a curved distal, a pronounced bulb and a transversal back. This refuse is associated with the carving of projectile points and other bifacial objects. We were surprised to observe the large amount of such artifacts and the presence of complete points recovered from excavations and surface collections, and they were given special attention to define the possible production level of projectile points at the site.

The objects were weighed, by using a digital Acculab scale of 0-200 g.  A magnifying glass with 10 magnifiers and a white light lamp were also used, to help appreciate features such as texture, color, erosion, modifications, and other interesting traits that some of the artifacts exhibited.

For the analysis of obsidian sources we used visual techniques, complemented with the analysis at the lab of 20 relevant selected samples. The samples were tested by Dr. Fred Nelson at Brigham Young University, using the method of X-ray fluorescence. These tests revealed a moderate presence of materials originating in the source of San Bartolomé Milpas Altas, which was not initially considered in our work, and was not taken into account in the analysis.

To reinforce our efforts, we visited the obsidian sources of El Chayal, the area of La Joya, the Ixtepeque volcano and the site of Papalhuapa, wherefrom a large amount of samples was collected. Finally, a sampler was created with materials from the most representative sources, namely: El Chayal (several outcrops), San Martín Jilotepeque and Ixtepeque. The results of the artifact analysis indicate that the source of El Chayal was the most represented one, followed by San Martín Jilotepeque.

Complementary Activities

With the purpose of familiarizing the analysts with the prehispanic technology, several replicative experiments were carried out using the carving technique of direct percussion. In this way, several flake cores were carved and their refuse recovered and compared with the carving refuse of the sampler prepared for the obsidian workshop; as a result, similarities were found with certain types of flakes. Work was accomplished as well with bifacial carving through direct percussion, which generated bifacial flakes that when compared with the flakes from the analyzed sample, helped to fully establish this technological type. With the material produced through the experimental carving, a sampler was also created to help establish comparisons, and to be didactically used in the future.

A graphic record of several lots and artifacts was elaborated with digital photographs using a Sony 2.0 mega pixels camera, with slides, and finally with drawings made on sheets of millimeter paper, transparent paper and the sketches drawn up in the notebooks used by all analysts.

With the help of Dr. Chinchilla, a database using the Microsoft Access program was elaborated, with separate tables for samples recovered from shovel tests and objects recovered from excavations, specifically operation P31.  Special artifacts were included in an additional database, composed of a collection of 181 mostly complete objects from the excavation of P31, representing the technological types that were being produced and carved at the workshop. These included projectile points of varied styles, prismatic blades of the final series, exhausted cores, some of which were reused, retouched macro blades, etc.

The classification work was concluded in March 2003, with the simultaneous completion of the data entry in the databases referred to earlier in this work. Presently, the statistical analysis is underway with the creation of tables, charts and graphics whose interpretation will be crucial in the reconstruction of production activities connected with obsidian. It should be said that this work could finally be accomplished thanks to the remarkable input of the students, whose hard work allowed for the classification of over 35,000 artifacts, one of the largest collections ever studied in the archaeology of Guatemala and in the overall Maya area.

Finally, a revision was made of another obsidian sample recovered at that same workshop during the excavations that Sonia Medrano conducted in 1996, in a sector adjacent to the area excavated by the Cotzumalguapa Project. This was done with the purpose of establishing comparisons and of complementing the information on the collection of the workshop located at El Baúl. The types are consistent with those of the tested sample, and the trend was confirmed towards the production of prismatic blades and bifacial projectile points. The first, because of the presence of many of those artifacts corresponding to the final series, and the second, because of the large amount of flakes of bifacial refuses.

Preliminary Results

The tested sample was only a fraction of the total obsidian recovered at the excavations. This sample includes two components:

1. Shovel Tests (23,927 analyzed artifacts). This sample derives from the analysis of ten shovel tests, which include some of the richest in obsidian refuse.
   
   
2. Excavations (23,670 analyzed objects). This sample derives from one of the special obsidian deposits recovered in operation EB9A-P31.

Tables 1 and 2 summarize the results of the analysis of both components. Then, there are some preliminary comments derived from these results for each analyzed variables.

Source

According to data obtained from shovel tests, the most represented source was El Chayal with 69.19%, followed by San Martín Jilotepeque (SMJ) with 21.68%.  The materials tentatively identified as from San Martín Jilotepeque amounted to 4.09%, which together with SMJ would add up to 25.75% of the total. The materials originating in Ixtepeque hardly amounted to 0.54%, and they probably derive from the source of San Bartolomé Milpas Altas, a suggestion that considers the results of the X-ray fluorescence analysis. On the contrary, in the excavations, El Chayal represented 35.32% while SMJ only amounted to 1.67%.  This remarkable variation in percentages takes place because 61.21% (14,489) of the sample artifacts from the excavations were not evaluated in the aspect of the obsidian sources, as they were very small refuses that made it very difficult to conduct visual analysis.

In the table of excavations, we have included the "Not evaluated" source, which encompasses 61.21% of the artifacts in this sample. This category corresponds to micro-refuses recovered when materials were sifted through a 2 mm mesh. The minimum size of these materials made it impossible to accurately establish the sources through visual methods.

Because of the above, it is inferred that most materials come from the source of El Chayal, which, surprisingly, is located at a greater distance from El Baúl than San Martín Jilotepeque. The variety presented by the materials from El Chayal suggests that there was a supply from different deposits belonging to this source, mainly from La Joya.

The materials that could not be assigned to a source, which amounted to 4.5% in shovel tests and 0.05% in excavations, might be related to the source of San Bartolomé Milpas Altas, whose presence was confirmed through the X-ray fluorescence analysis conducted by Dr. Fred Nelson.

Results of the source analysis indicate that the production workshops at El Baúl were mainly supplied with materials originating at El Chayal, although it is not known whether the artisans traveled directly to the source to obtain preforms, or whether the entity that exploited the deposits in this source sent traders to distribute the necessary materials for the production of the most frequent types. The presence of materials from El Chayal in the southern coast of Guatemala has been documented as of the Preclassic period, and has experienced a strong increase along the Terminal Preclassic and the Early Classic periods.

Type

In this variable, the types most represented in the shovel test sample were the following: smash (40.37%), flake residues (37.41%), irregular blade (7.71%), prismatic blade (7.50%) and small flake (3.26%). As to the excavated sample, the results obtained were: smash (63.69%), flake residues (29.93%), flakes from bifacial (3.56%), irregular blades (0.63%).

As seen, the types associated with the carving of artifacts represent the majority in both samples, with flakes and residues occupying a position of privilege. These results point to an intense activity related to the production of obsidian objects at sector EB9 of the site. The residues are related to the production of prismatic blades and bifacial artifacts such as projectile points. We may say that the huge amount of such artifacts shows that the elaboration of obsidian objects was a major activity in the economy of El Baúl, one that probably involved a large number of artisans and was, possibly, a family activity.

Another type directly related to the production of prismatic blades, as is the exhausted core, is not significant in the sample, with a mere 0.5% for the shovel test sample. On the contrary, this type is represented in the collection originated in the excavations, but it was included in a different database denominated "special artifacts". In this type we observed reutilization and retouching for new purposes, indicating the reuse of available resources.

Reconnaissance activities at other areas in El Baúl have revealed the presence of other zones with a high density of obsidian objects which might indicate the presence of other artifact production workshops, suggesting there was a remarkable demand of such products that probably exceeded the boundaries of the urban sector, with objects that reached other localities in the surroundings or even farther away.

Segment, Use and Retouching

In both samples, these variables showed the behavior we anticipated for carving refuses. Over 90% presented no use and retouching. Regarding the segment, there is a small variation between the shovel tests and the excavations, as in the first case the complete artifacts represent 83.73%, while the excavated ones amounted to 97.03%. This is because in shovel tests, the amount of prismatic blades and associated objects is larger, with a stronger presence of medial and proximal segments. In the second case, we have more carving refuse of smaller dimensions, and consequently, most of the objects are complete. Furthermore, the already elaborated artifacts such as blades and projectile points are but a few and were analyzed separately.

Comments

This is the first time that a systematic and formal research is conducted with the huge amount of refuses originating in one or several production workshops of obsidian objects in Guatemala. The fact that the site of El Baúl is located in a south coastal area far away from the obsidian sources, makes the presence of production workshops even more interesting, as this may indicate, on one side, that there was a strong demand of obsidian products at the south coast in the Late Classic period, and on the other, that the final production was being carried out right there at the site, in places destined to such effect, duly supplied with previously processed materials, the result of exchange with the entities that controlled the major obsidian sources.

With the data at hand, it may be suggested that at El Baúl there was at least one large scale production workshop of two major types of artifacts, namely, the prismatic blades, of moderate dimensions, and the projectile points. Considering the amount of refuse, it is probable that the production met the local demand as well as the demand of the surrounding areas under the control of this large regional center.

The results of our investigation will be highly valuable to conduct comparative studies with other lithic workshops existing in Mesoamérica. In turn, they will be of help to gain knowledge on the economic behavior of ancient societies regarding the exploitation of natural resources in the pursuit to serve the needs of urban cores and their surrounding areas.

Previous Page  |  Table of Contents

Return to top of page