A Taste for Green: A global perspective on ancient jade, turquoise and variscite exchange

1

Production marks on turquoise objects and lapidary technology at Chaco Canyon (New Mexico, USA): an experimental archaeology approach

Emiliano Ricardo Melgar Tísoc and Frances Joan Mathien

Thousands of turquoise objects have been excavated from different sites of the American Southwest; and Chaco Canyon, New Mexico, is well known as one of the principal areas where large concentrations were recovered. For example, in Room 33 at Pueblo Bonito two men were buried with the largest number of such artefacts ever found (Judd 1954; Mathien 1984; 1993; 1997a; 1997b; 2001; Neitzel 1995; Pepper 1996; Windes 1993a; 1993b; 2001). Since then, most researchers focused on the symbolic meaning, morphology, provenance, trade and use of turquoise, but very few studied the production marks visible on these artefacts. In this paper, we present a new approach that characterizes and compares production marks on similar types of items made using experimental archaeology techniques with those seen on artefacts. The results from sites in Chaco Canyon are compared with those in a database for artefacts from sites in the Greater Southwest and Mesoamerica.

En el Suroeste de los Estados Unidos los arqueólogos han recuperado decenas de miles de piezas de turquesa, la mayoría de ellas cuentas e incrustaciones que formaban parte de collares o mosaicos. De todos ellos, el Cañón del Chaco en Nuevo México destaca por la gran concentración de objetos de turquesa que presenta; por ejemplo, en el cuarto 33 de Pueblo Bonito fueron enterrados dos hombres con la mayor cantidad de turquesas registradas en el Suroeste en un solo contexto (Judd 1954; Mathien 1984; 1993; 1997a; 1997b; 2001; Neitzel 1995; Pepper 1996; Windes 1993a; 1993b; 2001). Sin embargo, la mayoría de los estudios que se han hecho sobre las turquesas se han centrado en su minería, comercio, uso, simbolismo, morfología y tipología. Por ello, el estudio tecnológico de las huellas de manufactura de las piezas de turquesa en el Cañón del Chaco aporta nueva información sobre su elaboración y distribución, así como permite confirmar o descartar las distintas herramientas de trabajo que se han planteado a partir de los contextos arqueológicos. Los resultados también se compararon con la tecnología identificada en otros sitios del Suroeste de Estados Unidos y Mesoamérica.

Introduction

Very few studies of turquoise artefacts recovered in sites in the American Southwest analyze, in detail, the manufacturing techniques used to produce them. Melgar (2014) developed an experimental archaeological approach that is non-destructive and non-invasive. Using a specialized polymer tape that records production marks at a very high quality, it is now possible to examine patterns recorded on the tapes under a scanning electron microscope and to identify the tools employed for each of the modifications made during the manufacturing process. The morphology and characteristics of the experimental production marks is part of a database on lapidary objects that Melgar (2014) has coordinated since 2004 at the Great Temple of the Aztecs Museum. That database includes more than 500 experiments with different tools and raw materials, allowing us to identify the tools and techniques employed on lapidary pieces; the results can be compared with studies of artefacts from different sites in Mexico, the American Southwest, and Central America. In this study, we applied the technique to 434 turquoise artefacts from 16 sites in Chaco Canyon to confirm, discard, identify and distinguish the tools that have been proposed for use in the production of turquoise objects from the canyon.

Overview of turquoise technology

Since the first excavations at different sites of the American Southwest, especially in Chaco Canyon, archaeologists have found several thousand turquoise inlays assembled into mosaics and beads or pendants included in necklaces and shorter strings. Recently, new instrumental methods, e.g., stable-isotope analysis that can identify the geological sources of archaeological specimens (Hull et al. 2008; Hull 2012; Thibodeau 2012; Thibodeau et al. 2012) have identified a number of turquoise sources that provided the raw material. Where and how finished ornaments were produced, and with what tools, is still debated.

Only a few workshops had been identified, excavated and analyzed, such as those at Alta Vista in Zacatecas (Kelley 1980, 54–55; Weigand 1993, 252–256, 293; Medina and García 2010, 73–167; Melgar 2011; 2014), Casas Grandes in Chihuahua (Di Peso 1974, 383; Di Peso et al. 1974, 8, 187), and Pueblo Bonito and the Spadefoot Toad Site in Chaco Canyon (Judd 1954, 326–328; Mathien 1993, 289; Windes 1993a, 14, 57, 173, 244 and 382–384). It was recognized that the importance of studying these production areas would be enhanced by the characterization of the tools and techniques employed by the lapidary artisans and the identification of the sequence of production of the turquoise objects.

Studies pertaining to the analysis of the manufacturing techniques are scarce (Haury 1931; Judd 1954; Mathien 1997a; Windes 1993a; 1993b), and none of these studies employed microscopes to characterize production marks on artefacts by comparing them with ones resulting from experimental archaeological studies. Therefore, the identification of which tools were employed in the production of the turquoise objects needed a more detailed analysis to distinguish among items made at different turquoise workshop locations and to determine whether finished items were manufactured in one area or were long-distance trade items.

To solve this problem, it was necessary to study the materials from turquoise workshops. To identify which turquoise objects were manufactured in each of the jewellery workshops at the American Southwest, which could be from Northern Mexico, and which could be Mesoamerican productions with imported raw materials, it is necessary to know which tools and techniques were employed in their manufacture. Abrasives of chert powder and reed sticks were mentioned for Aztec stoneworkers by Friar Bernardino de Sahagún (2006, 547 and 675). Raw materials and tools recovered in the archaeological contexts, especially in the lapidary workshops, include sandstone lapidary abraders at Pueblo Bonito and the Spadefoot Toad Site in Chaco Canyon (Judd 1954, 326–328; Windes 1993a, 153–159; 1993b, 213–229), petrified wood from the Spadefoot Toad Site (Cameron 1993), chert polishers from Alta Vista (Melgar 2011), and felsite metates from Paquimé (Di Peso 1974, 579–580). Other researchers proposed production of these pieces by polishing with selenite powder (Windes 1993a, 215, 232, 236, 244 and 257; 1993b, 227 and 229), or drilling with wet canes or wet cacti dipped with sand (McNeil 1986, 114), with cactus spines of Echinocactus wislizini (Mathien 1997a, 1162 and 1204; 1997b, 1227), or with porcupine quills of Erethizon dorsatum (Gillespie 1993, 367; Windes 1993a, 159; Mathien 1997b, 1227).

To know which tools were employed on turquoise objects made in Chaco Canyon, the analysis of their manufacturing technology was conducted as part of a larger project titled Style and technology of the lapidary objects in Ancient Mesoamerica by adding assemblages from the American Southwest. Initial studies distinguished among the different technological traditions from several areas in Mexico and identified which pieces came already manufactured from these workshops (Melgar 2010; 2013; 2014; Velázquez et al. 2010; 2012).

Objectives of this study

The main purpose of this research on Chaco Canyon turquoise is to characterize manufacturing patterns, test proposals of other researchers about the technological standardization and homogeneity of the objects and the sharing tools and techniques despite the provenance of the raw material.

Also, with this research, we investigate whether the tools and techniques employed were the same or if they changed through time.

Finally, the technological analysis should confirm or exclude tools proposed by other researchers such as abrading with sandstone, polishing with selenite powder, and drilling with petrified wood, with cactus spines, or with porcupine spines.

The characterization of production marks is carried out using stereoscopic microscopy and scanning electron microscopy. The latter is the technique that has yielded the best results to identify, with considerable precision, the materials and tools employed in turquoise jewellery manufacturing, and to distinguish between different technologies among lapidary assemblages.

Materials analyzed

To evaluate the tools used to produce Chaco turquoise ornaments, a total of 434 turquoise pieces were analyzed (Fig. 1.1). They were selected based on two previous studies. The first group of twelve was selected because we wanted to compare the patterns of manufacture with results obtained by Sharon Hull and Joan Mathien (Hull et al. 2008) regarding the provenance and procurement of turquoise objects (Table 1.1). This might suggest whether lapidary artists from different jewellery workshops used turquoise from different sources.

The criterion for selection of the second group is based on the context of deposits that relate to the production or consumption of turquoise pieces, including materials in different steps of production from diverse sites, chronologies and typological morphologies. The selection of 422 pieces from 13 sites and contexts was done following Mathien (1997) (Tables 1.2 and 1.3). Results might suggest whether there were changes in methods of production at different sites in separate localities in the canyon and/or over time.

Table 1.1. First group of turquoise pieces analyzed by Hull to determine provenance information.

Fig. 1.1. Examples of turquoise pieces analyzed from Chaco Canyon: raw material from 29SJ395 (a), recycled pendant used for inlay from Spadefoot Toad Site (b), unfinished pendant from Spadefoot Toad Site (c), bead blanks from Kiva G at 29SJ627 (d), inlays from Pueblo Alto (e), and beads from 29SJ423 (f).

Technological analysis of turquoise objects

Given the scarcity of studies examining production marks to identify the specific tools and techniques used to make turquoise objects, an experimental archaeology approach offers rich possibilities for researchers in resolving the problem of how to study this material. According to this approach, all activities in human societies are regulated; therefore, artefacts are used and produced by following a set procedure that gives them specific characteristics (Ascher 1961, 807; Gándara 1990, 51). This implies that the production or use of similar objects following the same pattern should yield identical results (Ascher 1961, 793; Coles 1979, 171; Velázquez 1999, 2). As a result, the major assumption of experimental archaeology is that the utilization of particular work tools, made of a given material, employed in a specific way under certain conditions will leave characteristic features or marks that can be differentiated (Binford 1977, 7; Tringham 1978, 180; Velázquez 1999, 2; 2007, 23). In this manner, more than 500 experiments with different types of tools and raw materials have been carried out, replicating the modifications detected on the lapidary pieces from Mesoamerica, Northern Mexico, and the American Southwest (Fig. 1.2 and Table 1.4). It is possible to identify the tools and techniques employed in different localities and workshops.

Table 1.2. Second group of turquoise pieces analyzed for technological patterns.

Table 1.3. Chronology of sites selected for the second group of turquoise pieces analyzed.

Fig. 1.2. Experimental archaeology using lapidary items: abrading with basalt (left) and cutting with obsidian blades (right).

Analysis of the production marks is carried out using three levels: macroscopic (the naked eye), low amplification optical microscopy (OM) (10× and 30×), and scanning electron microscopy (SEM) (100×, 300×, 600× and 1000×). The OM used in this study is located in the Hibben Center at the University of New Mexico in Albuquerque, and the SEM is in the SEM Lab at the Subdirección de Laboratorios y Apoyo Académico of the Instituto Nacional de Antropología e Historia in Mexico City.

Table 1.4. Materials employed at the Experimental Archaeology Workshop.

To evaluate the results of conjectural proposals and raise them to the level of confirmed procedures and tools employed in the lapidary items, the archaeological production marks are characterized and compared with the experimental ones to identify the tools in each modification.

The employment of the macroscopic level and low amplification microscopy to obtain photos at 10× and 30× allows us to determine that the use of lithic tools to abrade surfaces or edges, cuts, perforations and incisions, produces clearly marked scratch patterns that could even be identified without magnification. These marks differ considerably from the marks left by the use of abrasives, where fine lines are imperceptible on a macroscopic level and barely distinguishable at a magnification at 10× or 30×. However, it is also evident that with this means of observation, it is not possible to differentiate between the marks left by similar tools made of different materials. This is the case for obsidian and flint instruments which indistinctly produced similar patterns of straight, parallel lines on cuts on flat surfaces, or else concentric circular patterns on perforations.

To resolve this limitation, we employed a technique that would allow for greater precision in analysis: scanning electron microscopy (SEM). SEM is an ideal technique for the analysis of the surface characteristics of materials: topology, texture, porosity and other features.

Because the moving of materials to SEM labs from the Museums or store rooms at National Parks require special permits, we employed a technique to produce replicas in polymers softened with a drop of acetone for seconds; the film is pressed immediately onto the zones of the objects to be analyzed, obtaining a perfect mold on the polymer tape in one minute or less. Later, these molds were covered with gold ions for their observation at the SEM lab in a high vacuum. The replication method, typical of metallography, avoids having to move the archaeological materials, because the polymer samples are produced in the repositories where the original material is kept. Finally, it is possible to examine pieces that will not fit into the microscope’s sample chamber (10 cm × 10 cm × 5 cm) and produce high-quality images of modifications that do not conform to a flat plane, such as perforations.

Results

With the aid of experimental archaeology and the analysis of production marks on experimental and archaeological pieces, the following results were obtained.

Surfaces

With an optical microscope, all of the objects from sites in Chaco Canyon presented a smooth, lustrous surface, with very fine and fairly hazy straight lines (e.g. Fig. 1.3a). The comparison of the production marks on these pieces with those made experimentally in the laboratory showed that all of the pieces were smoothed without abrasives, and only the finished objects were burnished (Fig. 1.3b).

Fig. 1.3. Illustration of abrading tools: archaeological surface of inlay at 10× (a) and 100× (c) compared with the experimental traces produced by sandstone lapidary abrader at 10× (b) and 100× (d).

Fig. 1.4. Illustration of abrading tools absent in Chaco´s turquoise collection: basalt (a), rhyolite (b), granite (c), and limestone (d), all at 100×.

To be able to identify the material employed during the abrading and surface finishing, the archaeological pieces were analyzed with a scanning electron microscope. The objects showed fine 3-µm lines agglomerates that formed bands measuring approximately 10 µm in thickness (Fig. 1.3c). This resembles the production marks made with sandstone slabs or grinding stones and burnishing with animal skin (Fig. 1.3d). These marks differ from those produced experimentally using other rocks, like basalt, rhyolite, granite or limestone (Fig. 1.4).

Edges

The edges of the artefacts showed well-defined straight lines under the optical microscope (Fig. 1.5a). By comparing these with experimental production marks, it was possible to determine that all of pieces were cut with lithic tools (Fig. 1.5b), but it was difficult to distinguish which stone was used for that purpose at this level of analysis.

Fig. 1.5. Illustration of cutting tools: artefact edge at 10× (a) and 100× (c) compared with experimental traces produced by sandstone lapidary abrader at 10× (b) and 100× (d).

To identify which material was used for cutting the turquoise, the pieces were observed with a scanning electron microscope. The production marks were characterized by edges crossed by fine 3-µm lines agglomerated to form bands measuring approximately 10 µm in thickness (Fig. 1.5c). This resembles production marks made with sandstone slabs or grinding stones to reduce the irregular edges (Fig. 1.5d) detected on turquoise blanks and unfinished pieces. These marks differ from those produced experimentally using obsidian (Fig. 1.6a), chert (Fig. 1.6b), and petrified wood flakes (Melgar 2014). Thus, the edges of the turquoise pieces were ground and, surprisingly, the lapidary artists did not cut them with lithic flakes.

Fig. 1.6. Illustration of tools absent in Chaco´s turquoise collection: cutting with obsidian (a) and chert flakes (b) at 600×, and drilling with obsidian (c) and chert burins (d) at 1000×.

Drilling holes

Only beads and pendants have perforations. All of the drilled pieces have well-defined concentric circular scratches (Fig. 1.7a), similar to those obtained with lithic burins (Fig. 1.7b); and they differ from the fine, hazy scratches produced with abrasives.

To identify the specific materials employed in drilling the perforations, they were examined under a scanning electron microscope. The following pattern was seen: the well-defined concentric circles display bands 1.5 µm in thickness (Fig. 1.7c), which coincides with perforations made with petrified wood burins (Fig. 1.7d), and differs from patterns produced with obsidian (Fig. 1.6c) or chert burins (Fig. 1.6d) (Melgar 2014).

Conclusions

Technological studies of lapidary objects, with the support of experimental archaeology and the characterization of production marks based on stereo microscopy and Scanning Electron Microscopy, enable us to precisely identify most of the materials and tools employed for each one of the pieces examined in this study. Thus, it is possible to confirm or dismiss propositions regarding which materials and lapidary tools recovered from identified workshops were employed in the production of turquoise pieces.

Fig. 1.7. Illustration of drilling tools: artefact perforation at 10× (a) and 100× (c) compared with the experimental traces produced by petrified wood burin at 10× (b) and 100× (d).

The analysis of production marks on the turquoise pieces from the different contexts at different sites through time in Chaco Canyon confirmed the employment of sandstone lapidary abraders and petrified wood burins for all artefacts examined.

Regarding drilling of perforations, no abrasive was detected for polishing, so we discarded the proposal that selenite powder was used during this procedure. Among the artefacts examined, there was no evidence for drilling with cactus spines or porcupine spines. This, however, does not rule out their use as