The End of Pre-Columbian Pipil Civilization, Ciudad Vieja, El Salvador
With contributions by: Jeb J. Card, Department of Anthropology, Tulane University

The Chipped Stone Industry

Methodological and Theoretical Considerations

The chipped-stone industry of Ciudad Vieja is classified according to technological and behavioral considerations (Clark and Lee 1979; Sheets 1975, 1978; Fowler 1981, 1991a). The emphasis on technological and behavioral considerations allows a reconstruction of the nature and structure of an artifact industry. Another advantage to this method of classification is that the analyst bases the classification on inherent attributes derived from the manufacturing process, resulting in a classification that is more-or-less inherent in the artifacts themselves. "An industry refers to the artifacts, including the wastage in their manufacture, of a common raw material, which were manufactured using a common body of procedures" (Sheets 1978:3). Therefore, we are viewing inherent, not arbitrary or artificially imposed, categories of material culture.

The fundamental criteria for recognizing taxonomic units in behavioral classification are "behavioral discontinuities" (Sheets 1975:375). These are points in the sequence of artifact manufacture at which the artisan changed from one technique of production to another. The classic example of a behavioral discontinuity is the shift in a typical Mesoamerican chipped-stone industry from percussion to pressure in core reduction or in the manufacture of obsidian prismatic blades (Sheets 1975:279; Fowler 1981:278-279). The evidence indicating this shift in behavior or manufacturing technique is physically recorded on the artifacts themselves and reflected in the artifact classes "macroblade" and "small percussion blade" (removed from a polyhedral core by percussion) versus "prismatic blade" (removed from the core by pressure).

These artifact classes, while recognized and designated by the analyst, result from the behavior of the knapper who made the blades. Thus, when the analyst accurately identifies and interprets these behavioral discontinuities and uses them as a basis for ordering behavioral types, the typology becomes a model of the real structure of the industry. Typologies of the more complex Mesoamerican chipped-stone industries represent linear reduction models that reflect manufacturing behavior and production technology. It has been shown that these sequences vary considerably through time and space according to raw material, raw material availability, demand for lithic tools, functions for which the tools were designed, levels of knapper skill and specialization and skill, volume of production, and many other factors too numerous to mention (Fowler 1991a:3).

Deliberate modification of a blank or preform to produce a special-purpose object. Macroblades may be modified by unifacial retouching to produce scrapers or through bifacial retouching to produce projectile points. Another example would be the modification of a prismatic blade segment through unifacial pressure retouch to create a drill or an arrow point. The morphology of retouched or modified artifacts manifests itself in the final stage of artifact production. Therefore, morphological criteria should be used at a low taxonomic level to subdivide technological types.

Form is related to function, and functional analysis through microscopic examination of edge-wear is also an important part of behavioral and technological analysis. Many chipped-stone analyses conduct functional analysis on only a sample of the total number of artifacts. In the present analysis, functional analysis was conducted on all artifacts.

Procedure

The first step in any artifact analysis is the initial sorting based on raw material. At Ciudad Vieja, initial sorting took place in the field rather than in the laboratory, as all artifact classes were bagged or packaged separately according to raw materials. Once in the laboratory, all artifacts were carefully washed, labeled, and inventoried. Once the initial stage of classification was underway, we employed basic sorting categories derived from previous obsidian analyses from Chalchuapa (Sheets 1978) and Cihuatán (Fowler 1981), pre-Columbian Salvadoran archaeological sites with large obsidian artifact assemblages. The basic sorting categories derived from these previous researches were: debitage (waste flakes), utilized flakes, macroblades, prismatic blades, cores, unifacially retouched macroblades, unifacially retouched prismatic blades, and bifacially retouched macroblades. Nevertheless, it became immediately apparent upon first inspection that the Ciudad Vieja obsidian assemblage is far less complex than that of either of these two pre-Columbian sites used as a point of departure and a basis of comparison. Specifically, the overwhelmingly predominant type class in the Ciudad Vieja assemblage is the prismatic blade, while macroblades are absent, and cores are exceedingly rare. Bifaces are also uncommon. The implications of this truncated assemblage will be discussed below.

The next stage of the analysis was to record all metric data and non-metric attributes for every obsidian artifact. Proximal prismatic blade specimens were measured for length, width, thickness, and weight. The type and intensity of platform preparation and platform overhand removal were noted. All measurements were made with a stainless steel sliding caliper. Weights were taken with an Ohaus beam-balance scale. Edge-wear and retouch were also recorded. Microscopic edge-wear analysis was conducted on all artifacts using a Bausch and Lomb jeweler's loupe with 5X-7X-12X magnification, mostly used at 12X, and a Zeiss Stemi DV4 stereomicroscope with continuous variable magnification from 8X to 32X, mostly used at 12X to 16X.  The combination of the loupe and the stereomicroscope proved to be very useful in detecting manufacturing evidence and wear patterns that were invisible to the naked eye.

The Collection

Included in the total analyzed collection are 609 obsidian artifacts from controlled surface collections and excavations of the structures described above. Based on visual inspection virtually all of these artifacts were manufactured from Ixtepeque obsidian. Ixtepeque obsidian tends to be translucent, medium to light gray in color, often with a golden hue, and sometimes with some banding or cloudiness caused by the inclusion of impurities. Located in southeastern Guatemala, Ixtepeque would have been the closest source to the site. Three prismatic blade fragments are of green obsidian, and are presumably of material from the Pachuca, Hidalgo, source (cf. Sheets 1978:13). Several artifacts are a very dark, opaque, gray obsidian and may be from a Mexican source.

Obsidian artifacts are ubiquitous at Ciudad Vieja. They are abundant on the surface of the site, and there is no excavated context in which they do not occur. There is no doubt about the contemporaneity of Ciudad Vieja obsidian artifacts with the main occupation of the site in the second quarter of the sixteenth century.

Typology

Flakes and Utilized Flakes
Frequency. 47 (7%)

A flake by definition is any lithic artifact removed from a core whose length is less than twice its width. Conversely, a blade is at least twice as long as it is wide. These flakes for the most part appear to have been removed from small, cylindrical cores (probably exhausted polyhedral cores) by percussive force. Almost all of them show evidence of use for light cutting or scraping. Table 4 summarizes the metric values.

Click on image to enlarge.

Prismatic Blades (Figure 23, shown above)
Frequency. 557 (85%)

Prismatic blades are long and thin with parallel lateral edges. They were removed, probably by impulsive pressure from carefully prepared polyhedral cores. Several contrasting features distinguish prismatic blades from small percussion blades. Prismatic blade platforms tend to be smaller in area than those of percussion blades. The former do not exhibit platform crushing which does often occur on percussion blades. The bulbs of force tend to be smaller on prismatic blades, and they do not show ripple marks or compression rings which often occur on percussion blades. These characteristics indicate a shift from percussion to pressure flaking as the means for removing the blades.

The distal ends of complete specimens taper to a thin, sharp termination, but there are no complete specimens in the Ciudad Vieja collection. Prismatic blades tend to be trapezoidal in transverse section (with two dorsal ridges), and, less often, some are triangular in transverse section (with one dorsal ridge). They are concavo-convex in longitudinal section (the ventral surface is slightly concave).

Obsidian prismatic blades were used for a multitude of tasks by ancient Mesoamerican cultures. Cutting, scraping, whittling, and shaving were probably among the most common uses. In addition, as mentioned previously, obsidian prismatic blades could be modified into special-purpose implements.

The following tables present metric values for prismatic blades and the size of platforms measured on proximal segments. Since all blades were broken or snapped, the measurement of length is incomplete.

With regard to platform treatment and overhang-removal treatment, all of the 96 proximal segments had their platforms ground before being removed from the core. All were lightly to moderately ground. Grinding or abrading the platform surface increases the friction between the platform surface and the flaking tool, reducing tool slippage and flaking errors. Abrading the platform surface also breaks the surface tension of the platform, reducing the amount of force required for the removal of the blade (Sheets 1978a:12). Likewise, all 96 specimens had their platform overhang removed by light to moderate grinding or scraping around the edge of the core. Platform surface treatment and overhang removal are traits that change through time in ancient Mesoamerican industries, with light to moderate grinding becoming increasingly frequent in the Postclassic (Sheets 1978a:11-12, 73-74).

The cutting edge-to-mass (CE/M) ratio of Ciudad Vieja prismatic blades, calculated by dividing the total acute cutting edge (total length X 2) in centimeters by the total weight in grams (Sheets and Muto 1972), is 3.74.  This number was calculated on the entire collection of 557 prismatic blades. The CE/M ratio is an indicator of efficiency in blade production as well as scarcity or distance from the source of raw material. In general, knappers tended to maximize cutting edge when obsidian is scarcer (Fowler 1991a:3; Sheets 1978a:11, 1978b; Sidrys 1976:155-167, 1978:150-152). The CE/M ratio of 3.74 for Ciudad Vieja is relatively high but slightly lower than that of some Postclassic sites. For comparison, the CE/M ratio of prismatic blades from Cihuatán, an Early Postclassic Pipil site located very near Ciudad Vieja, is 4.38 (Fowler et al. 1987:154). If the difference between 3.74 for Ciudad Vieja and 4.38 for Cihuatán is significant, which it may not be, this could mean that the indigenous inhabitants of the villa of San Salvador had better access to obsidian, or at least to prismatic blades, than the Early Postclassic Pipils who inhabited Cihuatán roughly 400-500 years before the Conquest. It is quite possible that the Conquest changed or upset many of the socioeconomic factors that limited the exchange or acquisition of blades in pre-Hispanic times (Fowler et al. 1987:154-155).

Click on image to enlarge.

Small Percussion Blades (Figure 24, shown above)
Frequency. 25 (4%)

Small percussion blades are generally slightly larger and thicker than prismatic blades, although smaller examples may fall within the lower end of the range for prismatic blades. They were removed from a conical, polyhedral core by a percussion blow, as evidenced by platform crushing and larger bulbs of force at the proximal end. Otherwise, they closely resemble prismatic blades and may have been used for the same purposes. Small percussion blades could have been used for many of the same tasks to which prismatic blades were put, and they could also have served for heavier work as well.

The following tables present metric values for small percussion blades and the size of platforms measured on proximal segments. Since all blades were broken or snapped, the measurement of length is incomplete.

Polyhedral Cores
Frequency. 3 (>1%)

Polyhedral prismatic blade cores were shaped by percussion from macrocores and designed for the production of prismatic blades. They begin their lifespans with conical or cylindrical forms. As more and more blades are removed from a core, it assumes the shape of a pointed ellipsoid or a pointed parabaloid. At this point, when no more blades can be removed, the core is referred to as exhausted. All three of the polyhedral cores known from Ciudad Vieja are exhausted. In addition, all of them were smashed or broken by bipolar percussion. One of them shows signs of having been used as a reamer, with its parallel ridges worn smooth in a rotary motion.

The following table summaries metric values for these three specimens.

Laterally Retouched Implements
Triangular Points on Prismatic Blade Blanks (Figure 25, shown above)
Frequency. 7 (1%)

These are small, laterally retouched projectile points made from snapped prismatic blades. They may be made from a proximal or medial segment. The prismatic blade platform may form the proximal or distal end of the point. All are triangular in outline, plano-triangular in transverse section, and concavo-convex in longitudinal section. The edges and points are formed by careful pressure unifacial retouch along both edges of both faces. Retouch is usually a bit heavier from the ventral to dorsal surface. The bases are straight, un-thinned, and slightly ground. The following table summarizes metric values for these points.

Small, triangular points made on prismatic blade segments are widespread in Mesoamerica during the Postclassic period (Fowler 1989:372-377). Some have straight bases, while others have thinned, concave bases. The similarity in morphology and technology of these points suggests that they were all made in the same place, possibly by Mexican auxiliary troops.

Bifacially Retouched Implements
Stemmed Projectile Points (Figure 26, shown above).
Frequency. 4 (>1%)

The four bifacially retouched obsidian artifacts recorded here are projectile points or projectile point fragments made on small macroblade blanks. The three complete examples are plano-convex in transverse section and concavo-convex in longitudinal section. They have ovate blades with expanded, convex stems, and they are side notched. It is tempting to speculate that these points were weapons used by Mexican auxiliary troops.

Miscellaneous
Frequency. 14 (2%)

Fourteen artifacts or debitage fragments were classified as miscellaneous. These include three scraper-like implements made on small percussion blade segments and a drill-like tool on a small percussion blade segment. The remaining miscellaneous artifacts are unclassified.

Discussion

The existence of an indigenous obsidian industry at an early Spanish conquest town in northern Central America is not surprising. Obsidian industries persisted for decades after the Conquest in many regions of Mesoamerica. The persistence of obsidian prismatic blade manufacturing in the Valley of México after the Conquest is historically well documented (Clark 1982, 1989). Therefore, this kind of continuity in material culture at the villa of San Salvador is to be expected. But lithic specialists familiar with traditional Mesoamerican core-blade industries will immediately recognize an important change that occurred. The Ciudad Vieja obsidian assemblage does not exhibit the full array of products of a typical core-blade reduction sequence as known for pre-Hispanic sites in southern Mesoamerica such as, for example, Chalchuapa (1978) and Cihuatán (Fowler 1981) (for other examples, see Fowler [1991:3]).

The Cihuatán obsidian assemblage will be summarized here because it is of Early Postclassic date, and it is probably very typical of a pre-Hispanic Pipil obsidian assemblage. The Cihuatán obsidian industry shows the full Mesoamerican reduction sequence in all its diversity: macroflakes, macroblades, and small percussion blades detached from macrocores; scrapers, bifaces, small percussion blades, and prismatic blade cores made from the reduction of large polyhedral cores; prismatic blades and retouched, modified prismatic blades; exhausted polyhedral cores and a wide variety of all forms of debitage from core-blade production and biface and uniface manufacture. Raw material was acquired from several Guatemalan sources (especially Ixtepeque and El Chayal) for use by Pipil obsidian craftsmen at Cihuatán (Fowler et al. 1987).

In contrast, the Ciudad Vieja obsidian industry is remarkable for its relative simplicity and paucity of typological composition, indicating that it was not the product of a local reduction sequence. There is little evidence that prismatic blades, or any other artifact types for that matter, were produced on site, with the exception of the bipolar bashing of exhausted prismatic blade cores to produce small percussion blades. There is no core-blade debitage and there are no blade production mistakes present in the assemblage such as hinge fractures or outrepassé (plunging) blades as one would expect if blades were being produced at the site (cf. Sheets 1978b). The pattern of evidence presented here indicates that prismatic blades were probably imported ready-made for sale or barter at the town. Like prismatic blades, exhausted blade cores and medium-size flakes also may have been traded or brought into the site occasionally. Exhausted polyhedral cores make very handy secondary tools such as drills and reamers, and they were probably inexpensive. The predominant source seems to have been Ixtepeque in southeastern Guatemala.

Thus, while we see strong continuity in the use of obsidian tools, especially the omnipresent prismatic blade, at the villa of San Salvador by resident Pipils and Mexican auxiliaries, a drastic change occurred in the patterns of procurement and manufacture of these implements. In less than a generation after the Conquest, a highly specialized skill was lost, and ancient trade and exchange patterns were greatly altered forever.

Previous Page  |  Table of Contents  |  Next Page

Return to top of page