Visual and geochemical analyses of obsidian source use at San Felipe Aztatán, Mexico

Journal of Anthropological Archaeology 40 (2015) 266–279

Contents lists available at ScienceDirect

Journal of Anthropological Archaeology journal homepage: www.elsevier.com/locate/jaa

Visual and geochemical analyses of obsidian source use at San Felipe Aztatán, Mexico Daniel E. Pierce University of Missouri, Department of Anthropology, Swallow Hall, Columbia, MO 65211, United States

a r t i c l e

i n f o

Article history: Received 31 August 2013 Revision received 10 September 2015 Available online 29 September 2015 Keywords: Obsidian Trade Stratification Visual sourcing X-ray fluorescence West Mexico Aztatlán

a b s t r a c t Social inequality is a problem at the forefront of anthropological inquiry. The material record can give us valuable information about what resources were used at a given archaeological site. In this study, I have used obsidian and its distribution as a proxy for access to resources. By determining which obsidian sources were used and by whom, we can begin to understand the differential access to resources which is paramount to understanding social inequalities. This research uses a combination of visual sourcing and XRF geochemical sourcing to identify patterns in volcanic sources of obsidian artifacts at Post-classic San Felipe Aztatán in Nayarit, Mexico. Despite excellent quality sources being easily accessible, more sophisticated lithic reduction techniques such as prismatic blade production, seem to have been used only with more distant sources. With no substantial qualitative differences among obsidian sources, purely social factors likely resulted in the temporal and spatial distribution patterns at San Felipe Aztatán. Here, the limited area in which Pachuca obsidian is found may indicate an area of elite residence or elite activity, while its limited temporal distribution may reflect the emergence of trade and influence of the Aztatlán tradition dating to the Amapa phase during the Classic Period (500–750 CE). This research may have greater application for other sites across space and time. If obsidian source can be utilized to identify social stratification, we may be able to understand the spatial and social organization of specific sites as well as the complex dynamic trading relationships among sites. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Access to resources has long been understood as one of the most important indicators of social status. As one’s status increases, the access to more costly and diverse resources may also increase. Given this, we can use the material record and distribution of resources as indicative of social stratification. Commonly, this concept has been utilized through studies of ceramics, architecture, faunal remains, etc. But rarely has obsidian been used despite its abundance. In this study, obsidian source correlated with artifact distribution has been used to indicate the development of more complex socioeconomic structures. Through time, more resources became available as they were more limited in their distribution, indicating purposeful differential usage. This project focuses on the obsidian exchange in the Aztatlán trading system as reflected at San Felipe Aztatán in northern Nayarit, Mexico. The Aztatlán complex was a trade-oriented tradition centered in Western Mexico. Once thought to be a product of Mixteca-Puebla colonization (Smith and Heath-Smith, 1980), researchers are beginning to acknowledge that the underlying http://dx.doi.org/10.1016/j.jaa.2015.09.002 0278-4165/Ó 2015 Elsevier Inc. All rights reserved.

trading and cultural relationships far predate such influence (Garduño Ambriz and Gamez Eternod, 2005; Ohnersorgen et al., 2012). Though initially limited to the coastal plain, the trading network grew through time to eventually stretch from West Mexico to the Northern Arid Zone into the American Southwest (Kelley, 2000; VanPool et al., 2008). Aztatlán traders traded shell, ceramics, copper, and cloth far from their western Mexican homeland. Obsidian was also central to this trading network as reflected by its widespread distribution. Fortunately, distinct geochemical signatures of obsidian sources grant us an excellent opportunity to identify from where the artifacts originated. Scholars have noted the expansive trade networks but the complexities and specific trade relationships of this system remain unclear. This research attempts to clarify the issue by better identifying how specific obsidian sources were used within the Aztatlán settlement of San Felipe Aztatán and how use patterns may have shifted over time. For decades, researchers have attempted to identify obsidian source through visual sourcing, but have largely abandoned macroscopic sourcing with the increase in availability of more sophisticated geochemical archaeometric methods such as X-ray

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

florescence (Andrefsky, 2005). However, as discussed below, visual sourcing can still be a valuable, viable method in cases when there are macroscopically distinct obsidian sources, and when geochemical sourcing cannot be applied due to a lack of access to equipment or the inability to transport lithic material for analysis. In this project, I macroscopically examined approximately 1500 obsidian artifacts to identify obsidian color groups based on color and texture. These groups were used to identify visual source categories of the obsidian artifacts. I then geochemically sourced 13% of the total artifact assemblage to establish correlations between visual categories and specific geochemical sources. Through chisquare tests, these sources were then correlated to artifact morphology. Changes in source use over time were also identified based on temporally diagnostic ceramic types identified for many of the strata. Each source group was treated as an independent assemblage and then compared with other sourced groups to identify diachronic consumptions patterns in source frequencies, as well as differences in reduction techniques. Finally, I considered spatial differences in frequencies of lithic sources within San Felipe Aztatán. Four separate areas have been excavated, likely reflecting household refuse (and minimal moundfill in upper strata of two of the four units). However, the relative frequencies of each lithic source vary from location to location. Given the statistically significant distributions, the differential source frequencies likely reflect social factors such as varied economic status of the household as opposed to random differences in source selection. 1.1. Cultural historical background San Felipe Aztatán was an Aztatlán regional center and potential production center for obsidian artifacts (Garduño Ambriz and Gamez Eternod, 2005). This site is part of a larger physiographic province extending from San Blas on the central Nayarit coast northward approximately 200 km to Mazatlan in southern Sinaloa (Garduño Ambriz, 2007: 37; Scott and Foster, 2000). Historically, the waterways here were an important corridor of transportation, trade, and communication that connected the coastal plains to the highlands of western Jalisco. This area served as a rich refugia containing a plentiful assortment of various floral and faunal resources corresponding with the numerous microenvironments. Several important Aztatlán regional centers arose during the Early and Middle Post-classic periods. These settlements, including Amapa (Meighan, 1976), Chacalilla (Ohnersorgen, 2004, 2007), Coamiles (Duverger, 1998; Garduño Ambriz, 2006), and San Felipe Aztatán (Gamez Eternod and Garduño Ambriz, 2003; Garduño Ambriz, 2007; Garduño Ambriz and Gamez Eternod, 2005), feature many mounds, platforms, plazas, and ball courts. Smaller secondary centers have also been identified by Mountjoy (2000) (Fig. 1). The Aztatlán tradition continued to thrive into the Middle Postclassic. But, no diagnostic artifacts beyond this, dating from the Late Post-classic Santiago phase (1350–1530 CE), have been identified at San Felipe Aztatán (Garduño Ambriz and Gamez Eternod, 2005). Though we are not certain what led to the decline of this far reaching mercantile tradition, scholars have speculated that the developing Tarascan empire may have inhibited further growth and ultimately led to the decline of the Aztatlán system (e.g., Pollard, 2000). Obsidian has proven to be an interesting avenue for studying the Aztatlán system. Prismatic blades are razor sharp and efficiently created with minimal waste once the core is prepared. The technology is more complex than generalized lithic reduction and often requires trained specialists. Though obsidian is readily available throughout the region, we know it was traded extensively. Regardless of the reason for it, trade does not appear to be driven by demand for useful blades alone, given the abundance of blades from various local sources as well as distant ones. Rather,

267

the obsidian trade may simply be a byproduct of the general Aztatlán emphasis on trade. In understanding the provenance of the traded obsidian we can then begin to understand the trading relationships between these regional centers and the people living at San Felipe Aztatán. The site, first discovered by Sauer and Brand (1932), was initially called Loma de la Cruz and is now believed to represent the remains of the ethnohistorically documented capital town of the Aztatlán province (Anguiano, 1992). Similar to other Aztatlán regional centers, little is known of its cultural past. Unfortunately, excavations have been limited to salvage projects by the Instituto Nacional de Antropología e Historia (INAH) in the past decade in response to modern development (Gamez and Garduño Ambriz, 2003; Garduño Ambriz, 2007; Garduño Ambriz and Gamez Eternod, 2005). 1.1.1. Previous obsidian provenance research Obsidian is extremely abundant in West Mexico. Shifting tectonic plates produced by the convergence of the Sierra Madre Occidental and the Trans Mexican Volcanic Belt have created an active volcanic zone across southeastern Nayarit, northwestern Jalisco, and southern Zacatecas (Ohnersorgen et al., 2012). To date, 26 geochemically distinct obsidian sources have been chemically characterized in the region (Glascock et al., 2010), with most research focusing on the source areas of northwestern Jalisco and Durango (Darling, 1993, 1998; Darling and Hayashida, 1995; Esparza Lopez, 2008; Jiminez Betts and Darling, 2000; Trombold et al., 1993; Weigand, 1989; Weigand and Spence, 1982). Comparatively scant research has focused on the relationship between these sources and the distribution of obsidian artifacts on the west coastal regions, which is a focus of this study. Along the west coast of Mexico, obsidian was the most common lithic material and was used for the production of flaked stone artifacts including prismatic blades. At the large Aztatlán site of Chacalilla for example, over 99.5% of the 3849 chipped stone artifacts were produced from obsidian (Ohnersorgen, 2007). Similarly, at San Felipe Aztatán, 1501 of the 1562 (96.1%) lithic artifacts are obsidian. Polyhedral cores have also been recovered from several other Aztatlán sites indicating local production. However, few cores have been recovered relative to the abundant blades. There are multiple possible explanations for this discrepancy. It is possible that San Felipe Aztatán was not a blade production site and rather imported finished blades from certain sources. It is similarly possible that the limited excavations at this site did not sample the production area within the site. Future study may shed light on the cause for the disparity between cores and the abundance of blades. Obsidian is typically sourced using various geochemical methods, but macroscopic analysis can be useful in many cases (Braswell et al., 2000; Stark et al., 1991). Though a great way to identify the provenance of each artifact, often full geochemical sourcing is not feasible for a multitude of reasons. Often, only small samples of material are allowed to leave foreign countries for analysis. Furthermore, expensive equipment such as is required for geochemical analysis is frequently unavailable in many Latin American countries. Finally, when only a sample is geochemically sourced, often the sample selection process will be necessarily biased based on various factors, such as the research question itself and the inherent destructivity of certain bulk geochemical sourcing methods. Therefore, the sample may not represent the full range of variation in obsidian usage at the site. Luckily, geochemical sources are often so visually distinct that one can identify volcanic source purely through visual categorization with sufficient accuracy (Braswell et al., 2000). This method has been tested on numerous occasions. Through blind testing, Braswell et al. (2000) have demonstrated the great efficacy of visual sourcing. In their test, the four authors blindly sourced the

268

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Fig. 1. Location of coastal Aztatlán settlements.

same sample of 36 obsidian artifacts from a region familiar to them all. Of these artifacts, three of the researchers independently agreed fully in all source assignments. The same 35 of the 36 obsidian artifacts were assigned to the correct geochemical source based on visual souring alone by all three researchers. The fourth researcher also correctly identified the geochemical source visually 33 out of 36 times. Without a doubt, the potential efficacy of visual sourcing is demonstrated by this study. Similar positive results had been found by Aoyama (1991), Braswell et al. (1994), Carpio Rezzio (1993), Heller and Stark (1998), McKillop (1995), and Sanchez Polo (1991). Despite the successes of these visual sourcing experiments, Braswell et al. (2000) caution that these methods can be more difficult in areas where more than one source is similar in appearance (such as Central Mexico) and areas where many sources were used such as the northwest frontier of Mesoamerica. Given that West Mexico has no fewer than 26 identified obsidian sources, I chose to evaluate the efficiency of macroscopic sourcing using a combined strategy in which I evaluated the diagnostic color groups using X-ray fluorescence (XRF) analysis of a diverse sample to ensure that the color groups corresponds to specific geochemical sources. Glascock et al. (2010) have commented on the relative quality of the various West Mexican sources. Several sources produce obsidian of poor quality, but 13 of the 26 identified sources are categorized as excellent quality (Glascock et al., 2010). These obsidian sources are widely dispersed and provided West Mexican people with ready access to good quality obsidian. It appears that the trade of obsidian may not have been focused on the differing quality, but instead on creating trade networks for the purposes of trading other materials and establishing cooperative relationships and alliances between sites (Kelley, 2000). This can ultimately result in elite control of access to this network. As has been witnessed ethnographically around the world, often unnecessary trade can develop as a kind of costly signal as a way to show social status and create

powerful trading partners with whom sociopolitical alliances can be created. Though typically these signals are readily visible, more nuanced differences based on attributes other than visual difference cannot be discounted. 2. Methods and analytical background Excavations at San Felipe Aztatán were conducted in November and December of 2002 under the direction of Mauricio Garduño Ambriz and Lorena Gamez Eternod of the Instituto Nacional de Antropología e Historia (INAH). Four distinct areas were excavated: (1) Frente Calle Morelos, (2) Frente Calle Hidalgo, (3) Plataforma Adosado Sur, and (4) Plataforma Oeste (Fig. 2). Plataforma Adosado Sur and Plataforma Oeste are both near mound structures including the great mound Loma de la Cruz, while Frente Calle Morelos and Frente Calle Hidalgo are further from the larger mound features (Garduño Ambriz and Gamez Eternod, 2005). The lack of continuous stratigraphic units prevented the use of consistent excavation levels among the areas. Levels for each area were defined using natural stratigraphic breaks and obvious changes in material culture. In situations where little change could be observed or extensive stratigraphic mixing was assumed, levels were typically reduced to arbitrary 10–20 cm levels. The excavation areas were chosen through non-random stratified sampling based on the likelihood of intact subsurface deposits and the limited impact of modern construction on underlying deposits. Though the excavation locations are not contiguous, the use of diagnostic ceramic artifacts allows for dating of strata for site-wide comparisons. They also reflect the same general chronological patterns in that the most recently deposited levels of all units tend to be at least roughly contemporaneous and are more recent than the lower levels in all units based on superposition. A small amount of mixing has occurred, but the different stratigraphic units do generally reflect changes in artifact composition indicative

269

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Fig. 2. San Felipe Aztatán (redrawn from ‘‘Garduño Ambriz, Mauricio, and Gamez Eternod, Lorena, 2005. Programa Emergente de Rescate Arquelogico en San Felipe Aztatán, Municipio de Tecuala (Nayarit). Informe Tecnico Final/Trabajos de Sondeo Arqueologico. Archivo Tecnico Centro INAH Nayarit, Tepic”).

of generally intact subsurface deposits. Any mixing that may have occurred would only decrease the strength of, as opposed to creating, the differences identified in this research as materials from slightly different times may be lumped together as single analytical units. As a result, stratigraphic mixing does not appear to be a significant factor here. 2.1. Analytical methods The obsidian artifacts, which are permanently curated at the INAH museum in Tepic, Nayarit, were analyzed in San Blas, Nayarit in June and July of 2011. Recorded attributes were: (1) visual group, which reflects distinct geochemical sources; (2) temporal and spatial provenience; and (3) flake morphological traits such as flake weight, length, width, cortex coverage, and morphological type. These variables have been previously identified by Odell (2004) as among the ‘‘minimal set of attributes” typically recorded for lithic assemblages. In all cases, I have organized my data collection to allow for a statistical approach aimed at detecting differences in aggregate among excavation levels and units, as opposed to attempting to focus on the technological or functional significance of individual artifacts. 2.1.1. Macroscopic and geochemical sourcing Archaeologists are rightly suspicious of obsidian provenance studies based on macroscopic analysis alone because of questionable replicability and the likelihood of misidentifying sources (Braswell et al., 1994, 2000; Levine, n.d.; Moholy-Nagy and Nelson, 1990). Some sources have so much variation that pieces from the same source might appear macroscopically distinct while obsidian from different sources may look identical (Moholy-Nagy and Nelson, 1990). However, several studies have demonstrated the potential of visual sourcing given the correct conditions (e.g., Aoyama, 1991; Braswell et al., 1994, 2000; Carpio Rezzio, 1993; McKillop, 1995; Sanchez Polo, 1991; Tykot, 2003). The

potential of using a macroscopic approach is particularly compelling in this case because of the difficulty of securing INAH permission to transport complete lithic assemblages for analysis to the United States (Braswell et al., 2000). Still, the effectiveness of macroscopic sourcing must be demonstrated using independent lines of evidence. While I could not secure permission to export the entire obsidian collection, I was given permission to analyze a substantial sample of the artifacts (13%) at the University of Missouri Research Reactor (MURR) using XRF geochemical analysis to ensure the defined color classes correctly and adequately reflected geochemical source. I began the sourcing analysis by building on the method developed by Clark (1985) and Clark and Bryant (1997) and sorting each artifact into distinct color/texture groups (Table 1). Visual sourcing of the San Felipe Aztatán obsidian was done simply by holding each artifact up to the sunlight. Assemblages from each excavation level were analyzed in aggregate. Looking at obsidian artifacts from the same context facilitated easier visual source groupings by allowing a range of variation in color and texture to be considered. As natural sunlight comes through the transparent volcanic glass, various hues are visible that are indicative of source group.

Table 1 Description of visual sourcing categories.

a

Color code

Description

Associated geochemical source

1 2 3 4 5 6 7 8 9

Dark opaque green Clear grayish green Pale medium gray Black Darker opaque gray Smoky dull greenish gray Indeterminate colora Yellowish green/brown Miscellaneous

Volcan las Navajas La Joya Ixtlan del Rio Not present Ixtlan del Rio La Joya n/a Pachuca n/a

Used for pieces which no hues were observable due to shape or thickness.

270

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Fortunately, the consistently clear skies during the early summer months allowed consistent lighting conditions for analysis. Table 1 also reflects the geochemical sources associated with each color as determined by the XRF analysis discussed below. A selected 13% sample was then analyzed using a Bruker Tracer IIISD X-ray fluorescence spectrometer. The Bruker Tracer III-SD XRF is a portable spectrometer with a sample changer system developed by Dewitt Systems. The instrument has a rhodium-based X-ray tube operated at 40 kV and a thermoelectrically-cooled silicon-drift detector. The obsidian calibration uses a set of 37 very wellcharacterized obsidian sources with data from previous ICP, XRF, and NAA measurements. The samples were counted for 3 min to measure the minor and trace elements present. The elements measured include Rb, Sr, Y, Zr, and Nb. After data collection, element concentration data were tabulated in parts per million using Microsoft Office Excel. 2.1.2. Artifact group/class code Various taxonomic methods have been used to classify Mesoamerican lithic artifacts (e.g., Clark, 1985; Clark and Bryant, 1997; Santley et al., 1986), but these often employ potentially subjective definitions that lead to poor replicability (see Andrefsky (2005) and Railey and Gonzales (2014) for discussions of inter-observer error in flake classifications). For this analysis, I was able to use comparatively broad, generally replicable technological categories of (1) finished prismatic blades; (2) flakes, including byproducts related to polyhedral core preparation, byproducts of prismatic blade or formal tool production, angular shatter and other miscellaneous utilized or unutilized debitage; and (3) nonprismatic formal tools, including items such as projectile points, scrapers, and worked or notched blades. Flakes featuring evidence of utilization as an expedient tool, as opposed to formal retouch, were classified as a flake (differently than formal tools). This classification does not differentiate between types of flakes, but does measure the production of prismatic blades, formal tools such as projectile points, and production debitage/expedient tools, which is central to understand the technology reflected in each raw material. 2.1.3. Amount of cortex The amount of cortex on the dorsal side of a flake may represent the production stage of a tool, with cortical flakes being more typical of early stage reduction (Andrefsky, 2005; Johnson, 1989; Morrow, 1984; Sanders, 1992; Walker and Todd, 1984; Zeir et al., 1988). The amount of cortex can also reflect cobble size, with small cobbles producing more cortical flakes relative to larger cobbles when reduced using the same reduction techniques (Andrefsky, 2005). In this study, the cortex was recorded using the four rank scale utilized by Andrefsky (2005: 106): a score of 3 is given to flakes with cortex covering their entire dorsal surfaces, 2 is given to flakes with cortex covering between 99% and 50% of their dorsal surfaces, 1 is given to flakes with cortex covering between 1% and 49% of their dorsal surfaces, and 0 given to flakes that lack cortex.

within each visual group, as opposed to the ‘‘typical” member of each group. Therefore, my sample likely reflects the maximum variability within each visual group. I chose this sampling strategy instead of a simple random sample to ensure an understanding of the likely errors in sourcing and to provide the most conservative test possible to evaluate the utility of the color groups. Random samples have been demonstrated to sometime underrepresent variation within heterogeneous populations, especially when there are limited numbers of atypical individuals. By intentionally selecting the atypical examples, I can be as sure as possible that all potential errors are understood, even if they are overrepresented in the resulting analysis. Of the 195 sampled, 181 could be confidently associated to a single source and 12 were chemically consistent with several sources. The remaining three were unassigned sources due to compositional differences between the specimens and known source groups. The visual source groups and XRF produced groups were then compared (Table 1). XRF analysis revealed that visual source Groups 3 and 5 are both seemingly from the same chemical source, Ixtlan del Rio. Similarly, Groups 2 and 6 both mostly represent La Joya. For this reason, I have collapsed these two color groups into Groups 3 and 2 respectively for statistical analysis. For the purposes of assessing reliability of visual sourcing, visual Group 7 was absorbed into Group 9 because no specimens were sampled for geochemical sourcing. In regards to assessing validity of visual categorization, the combined Group of 7 and 9 has been dropped given that it is a miscellaneous color category but has been retained for analytical purposes for other statistical analyses. Given that identifying the specific geochemical source associated with these artifacts is not as important as knowing which sources these artifacts do not come from (e.g., Volcan las Navajas, Ixtlan del Rio, La Joya, and Pachuca), this group can offer valuable information despite its heterogeneity. Table 2 reflects these collapsed color– source associations. Visual groups in fact correlate to a single source more than 84% of the time. The correspondence between color group and geochemical source was confirmed by chi-square testing (Table 2). However, a random sample of the assemblage would have likely produced an even greater correspondence between visual source groups and geochemical sources because the less common color variants that I had intentionally selected would have been comparatively rarer. In other words, the data in Table 2 reflects the maximum likely variation. The actual correspondence of the entire assemblage is in fact likely considerably greater. The nonrandom association between color groups and sources indicate the utility of visual sourcing even in an area of great obsidian variability.

Table 2 Comparison of visual and geochemical sourcing (by XRF) of obsidian from San Felipe Aztatán.a Geochemical source

3. Results I conducted geochemical sourcing analysis using XRF of 195 of the 1476 (13.14%) obsidian artifacts that were visually sourced to evaluate the relationships between color groups and volcanic source. I analyzed samples from all visual source groups (notwithstanding the minimally identified Group 7 which has been subsumed by Group 9). Obsidian samples were non-randomly selected for XRF analysis to maximize the variation represented within each visual group. For example, I selected obsidian specimens that reflected the entire range of color and texture variation

a

Visual groups

Total artifacts per source

1

2

3

8

Volcan las Navajas La Joya Pachuca Ixtlan del Rio Osotero San Juan de los Arcos Boquillas Unidentified source

41 4 0 0 0 0 0 2

1 63 11 1 0 0 0 0

0 0 1 42 1 6 2 0

0 3 11 0 0 0 0 0

42 70 23 43 1 6 2 2

Total artifacts per color group

47

76

52

14

189b

Chi square value = 389.84; chi-square critical value = 38.93; df = 21; alpha = .01. Note that the 189 total artifacts do not reflect the 6 sampled artifacts from color group 9 (MISC). b

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

271

Fig. 3. Location of XRF identified utilized obsidian sources at San Felipe Aztatán.

Fig. 4. Total relative distribution by percent and raw count.

Group 1 has a near perfect correlation with the Volcan las Navajas source. Visual group 2 is almost entirely La Joya obsidian (83%),

but does include Pachuca obsidian (14%) and a small amount of Ixtlan del Rio (1%), and Volcan las Navajas (1%). Color group 3 also reflects more than one volcanic source, but is primarily associated with Ixtlan del Rio (81%). It also includes all of the obsidian from San Juan de los Arcos (12%), Osotero (2%), and Boquillas (4%), as well as a small amount of the Pachuca obsidian. Group 8 is primarily Pachuca (78%) but it also includes some La Joya (22%) obsidian. Thus, the most likely confusion is between La Joya and Pachuca obsidians. However, the confusion is relatively limited (Group 2 contains mostly La Joya while Group 8 contains mostly Pachuca) and the presence or absence of a correspondence between Groups 2 and 8 provides insights into the likelihood of errors. Strata containing Group 2 but limited or no Group 8 likely contain little or no Pachuca obsidian. Conversely, strata containing Group 8 but limited amounts of Group 2 likely contain little or no La Joya obsidian. Strata containing both visual source groups likely contain a mix of Pachuca and La Joya obsidian. In summary, the strong correspondence between the visual groups and the specific sources allowed me to use visual sourcing and a statistical framework to determine changes in source preference through time and across space. The locations of the identified sources are indicated in Fig. 3. Fig. 4 indicates the total proportion of artifact classes by source identified through this combined method. Artifact count (i.e., the number of individual artifacts from a given source) and artifact weight (i.e., the combined weight of all artifacts from that source) correspond well with each other, as one would expect except in contexts where some material is being reduced more intensively or is exposed to greater fragmentation because of use or depositional contexts. For the remainder of the analysis, I will use artifact counts to compare frequencies, but will also include artifact weight as a scale of reference for the entire volume of artifacts from each given source.

272

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

However, looking at this assemblage in terms of count rather than weight appears to not alter the relative distribution in any significant way. The one major discrepancy in the two counting strategies is between La Joya and Volcan las Navajas as they switch places ordinally. Even this can be expected however, due the nature of the assemblage in terms of morphology. As La Joya lithics are primarily thin prismatic blades and Volcan las Navajas is primarily flakes of thicker production debris, it is no surprise that Volcan las Navajas lithics tend to be heavier. This appears to be the major contributing factor in affecting the relative distribution if viewed in terms of mass alone.

Nacionales (Scott and Foster, 2000). As San Felipe Aztatán grew in prominence, it may have established trade networks with other surrounding communities that had better direct access to other obsidian sources. For this reason, we see Volcan las Navajas lithics slowly decline proportionately beginning during the Amapa phase and continuing during the Early and Middle Post-classic. But given this new pattern of obsidian usage during the Classic period with a greater emphasis on traded obsidians in particular, it seems as though an important shift took place in production and trade at this time period. 3.2. Comparison of artifact morphology by source

3.1. Changes in source selection through time In many cases, strata could be identified through diagnostic ceramic typologies. However, not all strata could be temporally identified. Little stratigraphic mixing is apparent although it is minimal (Garduño Ambriz and Gamez Eternod, 2005). Importantly, the upper levels of the Plataforma Oeste and Plataforma Adosada Sur (both associated with the largest mound Loma de la Cruz) have been impacted with stratigraphic mixing as a product of mound construction. However, the artifacts found within this fill appear to be nearly entirely of Cerritos phase (900–1100 CE) in origin indicating a Post-Classic construction of the mound, and limiting the impact on the study of the abundance of obsidian from the various sources through time. Below the mixed strata, we find multiple occupational floors dated to the classic period Amapa phase (500–750 CE). Given these findings, distinguishing between Classic and Post-Classic occupations can be easily established based on typological cross-dating, allowing me to identify when important shifts in obsidian use took place at San Felipe Aztatán. Within the strata identified as belonging to the Chinesco (0–200 CE) and Gavilan (200–500 CE) phases, Volcan las Navajas obsidian is most common, with minimal obsidian from Ixtlan del Rio and La Joya. During the Amapa phase (500–750 CE), there is a proportionate increase in La Joya and Ixtlan del Rio artifacts compared to Volcan las Navajas. What is especially intriguing about this time period is that nearly all stratigraphic units containing the distant Pachuca obsidian have also been identified to the Amapa phase (9 of 13 temporally identifiable strata containing Pachuca). Of the total amount of Pachuca obsidian (156.4 g; 128 pieces), 84 pieces (102.2 g) date to the Amapa phase, while only 29 pieces (34.6 g) date to the Cerritos phase. The remaining pieces (n = 15; 19.6 g) are from strata not temporally identifiable through crossdating alone, although stratigraphically originate from levels below Amapa identified strata. Given this, it appears as though the extensive trade networks which typify the Aztatlán tradition may have actually developed prior to the Post-classic during the Amapa phase. By the Early Post-Classic, Pachuca obsidian was actually on the decline indicating a decrease in these exotic obsidians despite great exchange networks. In general, we see a rise in abundance of both La Joya and Ixtlan del Rio artifacts but a retention of Volcan las Navajas obsidian for local consumption. The Aztatlán tradition changed power and trade dynamics in Western Mexico, particularly in the coastal regions and Marismas

Table 3 presents the frequencies of three artifact types within the main source groups. Prismatic blades are the most common artifact type, but manufacturing debitage is also plentiful. Bifacially flaked formal tools are the least common. It appears that frequencies of tool type are also largely affected by source. The Volcan las Navajas group is dominated by flaking debris, whereas the other sources are comprised primarily of prismatic blades. A chi-square test at an alpha of .05 determined that these differences in abundance are indeed statistically significant. Importantly, the Volcan las Navajas group has significantly more debitage, and far fewer prismatic blades than expected as revealed by the adjusted residual values. The general preponderance of the debitage from Volcan las Navajas suggests local reduction and the production of nonprismatic blade flaked stone artifacts. Conversely, the general absence of debitage from the other sources may suggest that this material arrived at the site as prismatic blades, presumably through trade. Given that these artifacts become increasingly common during the Classic period Amapa phase, it appears lithic tool use may have shifted from the local reduction of Volcan las Navajas obsidian to the importation of finished prismatic blades made from more distant raw material at this time. Importantly, non-cortical flakes are heavily represented while cortical flakes constitute only a small minority of the assemblage (Table 4). Of the cortical flakes, however, the majority are from Volcan las Navajas. A chi-square analysis at the alpha level of .05 reflects that these abundances are as well significant. Although not statistically significant when calculating adjusted residuals, there are more Volcan las Navajas flakes in cortical categories 1 and 2 (i.e., cortex covering between 1% and 99% of the dorsal surface) than any other source. When combined with the differences in flake morphology/artifact type, the presence of the Volcan las Navajas flakes and the absence of similar cortical flakes from other sources further indicate the local reduction of Volcan las Navajas obsidian, and the importation of the prismatic blades from other locales. Still, cortical Volcan las Navajas flakes are far less common than non-cortical flakes, indicating that cobbles were of a sufficient size to allow the removal of many interior flakes. Further, the average weight of the Volcan las Navajas flakes is comparable to, and even larger than most of the artifacts from other sources, suggesting that they came from comparably sized cores (Table 5). Therefore, it seems that the difference is caused

Table 3 Frequency of artifact types by visual obsidian source.a Lithic class

Volcan las Navajas

La Joya

Ixtlan del Rio

Pachuca

Misc

Total artifacts per class

Prismatic blades Flakes Formal tools

5 412 3

480 55 1

282 63 10

124 2 1

34 6 1

925 (62.54%) 538 (36.38%) 16 (1.08%)

Total artifacts per source

420

536

355

127

41

1479 (100%)

The italicized percentages are the percentage of the total assemblage which that particular group comprises. a Chi-square value = 996.2; chi-square critical value = 15.51; df = 8; alpha = .05.

273

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279 Table 4 Distribution of cortical artifacts per visual obsidian source. Cortex amount

Volcan las Navajas

La Joya

Ixtlan del Rio

Pachuca

Misc.

Total per cortical type

0% <50% 50–99% 100%

406 9 7 2

532 2 0 1

347 3 2 0

122 0 0 0

39 1 1 0

1446 15 10 3

Total artifacts per source

424

535

352

122

41

1474

Chi-square value Chi-square crit. df

24.70 21.03 12

Table 5 Summary statistics for weight per source group. Color group

Count

Average weight of artifact

Standard deviation

Volcan las Navajas La Joya Ixtlan del Rio Pachuca Misc.

424 (28.77%) 535 (36.3%) 352 (23.91%) 122 (8.28%) 41 (2.78%)

2.16 1.24 1.43 1.22 2.36

2.72 1.82 3.42 0.95 3.53

The italicized percentages are the percentage of the total assemblage which that particular group comprises.

by different reduction strategies as opposed to substantial differences in initial cobble size. This is in agreement with Glascock et al. (2010), who found that the Volcan las Navajas source is not only of excellent quality but of large cobble size as well. 3.3. Correlation of source and excavation area Finally, I examined the data to see if the sources occurred in different proportions in the various excavation areas. To test this, I conducted a chi-square analysis comparing the expected frequencies of the four source groups in each excavation unit at an alpha level of .05. Adjusted residuals indicate that 6 of the 20 combinations are statistically significant, demonstrating that there are proportionally more artifacts from certain sources at specific locations and significantly less from other sources in other areas. Notably, 126 of the 128 Pachuca artifacts are found at the Calle Hidalgo area (Table 6). Garduño Ambriz and Gamez Eternod (2005) have hypothesized that all four of these excavation units represent household refuse (notwithstanding the minimal Post-classic moundfill on the platform units). Likely, if this is household debris, it should then reflect differences in households and the availability of various obsidian resources to specific households. It appears that all households at San Felipe Aztatán had access to traded obsidian but that certain areas/individuals at the site (i.e., the inhabitants at Calle Hidalgo) had greater access to obsidian from the more distant Pachuca source. Importantly, it is clear that all four excavation areas were occupied during both Amapa and Cerritos Phases. This likely reflects the presence of social differentiation and the use of obsidian to reinforce the increased social inequality associated with the Aztatlán system. 4. Discussion This research has focused on three related issues: changes in source usage through time, differences in artifact form based on raw material source, and differences in the raw materials utilized at the four excavation areas. The correspondence between visual group and geochemical source is not perfect, but is statistically significant and is reliable enough to allow strong patterns to be identified. In fact, any difficulty in sourcing would tend to decrease rather than artificially increase the strength of the differences such as the shift from Volcan las Navajas to the other presumably more

distant sources during later times and obscure the true positive correlation. Results from the analysis presented above indicate that reduction strategies shifted from the local production of tools using non-prismatic reduction strategies to the general importation of prismatic blades made elsewhere. The implications of this are profound. I believe that the earlier Volcan las Navajas materials dating to the Chinesco and Gavilan phases reflect a period prior to the development of the extensive trade networks that typify the Aztatlán tradition. In medial strata identified as originating in the Amapa phase, Pachuca obsidian occurs in great relative abundance. This is curious given that it predates the Post-classic Aztatlán fluorescence and the expansive trade networks thought necessary for such a far reaching trade. By the Post-classic, however, Pachuca obsidian became far less common. Concurrently, other more local sources, such as La Joya and Ixtlan del Rio, were continually accessed. This is likely indicative of the retention of more localized trade networks while Volcan las Navajas was continuously used for domestic production and used for less complex expedient tools. Artifact form has a significant relationship with the raw material source as well, and has equally profound effects. By identifying a positive relationship between production debitage and Volcan las Navajas, I have concluded that this source was consistently accessed for local production of everyday lithic items, despite the fact that the core size and obsidian quality are both suitable for the production of prismatic blades. Conversely, the other more distant sources appear to be accessed solely through trade in the form of finer prismatic blades given the small amount of production debitage associated with generalized reduction. Artifact form may have been the primary means of differentiating between locally available and traded obsidian, especially given that the macroscopically visible differences among the sources would have only been evident upon very close inspection. The different artifact morphology would have made other sources more valuable and indicative of not only an extensive trade system but a burgeoning social and economic system based on access to these exotic resources. Finally, I have indicated differences between excavation areas based on source frequency. More affluent households would have had access to more distant sources. Therefore, we should expect to find the Pachuca lithics primarily in a limited number of elite households. This has been confirmed in this study by the concentration of nearly all Pachuca obsidian in only one excavation unit

274

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Table 6 Two-way table of artifacts from each source per each excavation area. Volcan las Navajas

La Joya

Ixtlan del Rio

Pachuca

Calle Hidalgo

144 (330 g)

185 (198.3 g)

92 (190.5 g)

126 (154 g)

13 (32.1 g)

560 (904.9 g)

Calle Morelos

261 (533.9 g)

329 (447.7 g)

861 (1281.1 g)

12 (25.4 g) 10 (31 g)

2 (2.4 g) 0

28 (63.7 g)

Plataforma Adosada Sur Plataforma Oeste

17 (17.4 g)

241 (297.1 g) 10 (11.6 g) 16 (15.2 g)

0

1 (3.3 g) 0

37 (50.3 g) 43 (63.6 g)

Total artifacts per source

427 (920.3 g)

545 (673.4 g)

359 (514.4 g)

128 (156.4 g)

42 (99.1 g)

1501 (2299.9 g)

14 (10 g)

Chi-square value Chi-square crit. df

Misc.

Total artifacts per area

563.97 21.03 12

Underlined values are significantly different than expected based on adjusted residuals.

(Calle Hidalgo). The limited scope of the excavation precludes a full understanding of the distribution of Pachuca obsidian, but it does indicate differential access to imported obsidian and, presumably, social inequality. Given that there is little difference in quality of the obsidian between the various sources identified here, the shift from the Volcan las Navajas obsidian to the more distant sources reflects cultural factors that transcend a simple desire for good quality obsidian. Through time, certain individuals would have utilized their status by obtaining obsidian from more distant sources (such as Pachuca obsidian from The Basin of Mexico), specifically as finished products. Access to more distant sources could have been used as class markers. In this way, it was not an issue of which source produced better quality obsidian blades, but rather which source had the most perceived value based on other factors, such as rarity, origin, and cost. Although the general ubiquity of the non-local obsidian appears to indicate that most people at San Felipe Aztatán had access to more distant sources, the uneven distribution of Pachuca obsidian suggests both that it was valued differently (more) than other sources, and that only a limited subset had access to it. Despite the rising popularity of exotic sources, Volcan las Navajas, was still accessed by many. This source was continually used for expedient flake technology and household use. As the trade network seemingly developed during the Classic period as evidenced by an influx of more costly and distant sourced obsidian, the site was likely gaining influence facilitated by economically powerful elites and the developing Aztatlán tradition. Prior to these trade networks, Volcan las Navajas obsidian dominates the record demonstrating a common use throughout the social structure. As new obsidian sources were traded in, the higher class would have had access to them, while lower classes would have been more limited and resorted to continual use of the proximal Volcan las Navajas source. Subsequent study can further clarify the nature and degree of social differentiation. If this could be confirmed, we could greatly improve our understanding of the social economy of the Aztatlán tradition. 5. Conclusion This project can lend itself to a greater understanding of the Aztatlán tradition and the intricacies of its socioeconomic systems. Further, given the prevalence of obsidian at archaeological sites across time and space, the ideas presented here may be applied to a broad range of archaeological sites. Often, we can expect a correlation between the rise of more extensive trade networks and the increased social complexity as certain individuals gain access to increasingly rarer and more costly resources. I have illustrated the introduction of these trade networks at San Felipe Aztatán through site wide diachronic consumption patterns. I have also attempted to identify social differentiation through tool type and

source frequencies associated with excavation areas. If we accept that selective sampling during excavations has not simply missed production areas within San Felipe Aztatán, it appears that as trade increased, finished prismatic blades were imported rather than domestically produced in greater frequency. This relates to increased social complexity as indicated by differential access to specific valued sources by various household areas. Likely, the elites had access to ‘‘higher quality” obsidian. This ‘‘higher quality” was certainly not an issue of qualitative or functional differences. Rather, it was of perception only reflected by a preference for distant prismatic blades over locally produced percussed tools. By excluding the lower classes from certain trade networks, elites can effectively make the traded obsidian more valuable due to its limited dispersal. Here, I have taken the first step in identifying the social differentiation as expressed through obsidian consumption patterns. Future studies could also benefit from addressing these issues at other sites. If we can establish that specific sources are associated with exclusive social groups, we can then identify these groups strictly based on obsidian source frequencies, saving valuable time and analytical resources by deeming source frequency in and of itself an important reflection of social status for sites such as this. Social inequality has been a topic at the forefront of anthropological inquiry for centuries. Though there are many aspects of the material record which are amenable to such study, obsidian is often ignored when trying to understand prehistoric inequality. Obsidian, however, is often the most common artifact found at sites in many areas of the world. The inability to use obsidian for studies concerning inequality leaves this abundant and informative artifact type underutilized for answering broader anthropological questions. I have demonstrated that obsidian in and of itself can be used to inform us about social inequality and differential access to specific resources. This opens a wealth of opportunity to learn more about not just the material record itself but of the people who created it. With the abundance of obsidian found at countless archaeological sites, the ability to easily and cheaply source these artifacts becomes especially important. Furthermore, I have here shown that accurate visual sourcing can be conducted even in areas of high volcanic diversity. With the ability to accurately determine provenance, we can then better understand who had access to what resources and how this relates to internal social structures. In this paper I have demonstrated how obsidian usage patterns may indicate differential access by individuals at San Felipe Aztatán. Though significant for this site and the greater Aztatlán tradition, the lessons garnered from this study can be applied much more broadly in countless contexts to better understand resource availability and consumption patterns. Finally, the careful and thorough analysis of obsidian assemblages can contribute to a better understanding of social hierarchies and their relationship to the material record.

275

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Acknowledgments I would like to thank the following people for assisting me in my research. Dr. Jeff Ferguson, Dr. Chris VanPool, and especially Dr. Todd VanPool for helping me with all aspects of this project, and continually providing suggestions as to how to make this research stronger. I would also like to thank Dr. Michael Glascock at the Missouri University Research Reactor; Dr. Michael Ohnersorgen;

the department of Anthropology at the University of Missouri and the W. Raymond Wood Opportunities for Excellence in Archaeology Fund for providing me with partial funding for this analysis. Thank you to several anonymous reviewers and Geoffrey Braswell for helping me to greatly improve this manuscript. Finally I would like to give special thanks to my good friend and colleague, Mauricio Garduño Ambriz of the Instituto Nacional de Antropología y Historia in Tepic, Nayarit for providing artifacts, data, and assistance.

Appendix A

Sample I.D.

Source

Visual sourcing code

Rb

Sr

Y

Zr

Nb

dep001 dep002 dep003 dep004 dep005 dep006 dep007 dep008 dep009 dep010 dep011 dep012 dep013 dep014 dep015 dep016 dep017 dep018 dep019 dep020 dep021 dep022 dep023 dep024 dep025 dep026 dep027 dep028 dep029 dep030 dep031 dep032 dep033 dep034 dep035 dep036 dep037 dep038 dep039 dep040 dep041 dep042 dep044 dep045 dep046 dep047 dep048

Volcan las Navajas Volcan las Navajas San Juan de los Arcos Ixtlan del Rio Pachuca La Joya La Joya Volcan las Navajas La Joya La Joya Ixtlan del Rio Ixtlan del Rio La Joya La Joya La Joya La Joya La Joya Ixtlan del Rio Ixtlan del Rio Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas La Joya Macrogroup 2 Pachuca Ixtlan del Rio La Joya La Joya Pachuca La Joya Volcan las Navajas Volcan las Navajas Osotero Ixtlan del Rio La Joya Ixtlan del Rio Volcan las Navajas Macrogroup 2 Macrogroup 2 Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas

1 1 3 5 8 6 2 1 2 8 3 3 2 2 2 2 2 5 5 1 1 1 1 1 1 1 8 3 2 2 9 9 1 1 9 5 2 2 1 2 2 1 1 1 1 1 1

215 180 120 87 209 169 179 207 174 196 107 117 157 169 177 173 167 104 120 191 177 183 167 163 130 126 226 109 172 196 225 152 199 191 115 97 178 103 165 155 118 181 176 162 182 169 211

12 11 55 80 6 4 5 18 9 4 101 94 3 3 6 6 5 99 109 15 14 17 13 15 9 4 6 97 5 6 7 3 14 14 86 90 7 83 13 5 3 12 14 14 10 20 17

132 119 19 13 111 74 80 142 76 90 21 22 63 70 80 68 80 19 23 133 122 141 118 110 82 64 117 18 75 84 116 84 140 139 19 21 80 26 120 61 49 126 121 113 126 129 152

1182 1056 162 135 895 762 784 1217 795 756 162 163 721 758 786 744 781 166 171 1100 1104 1157 1010 964 755 550 945 172 790 798 947 683 1174 1098 183 161 806 172 1019 636 518 1089 1047 952 1050 1097 1236

122 118 20 19 92 66 71 133 70 67 25 19 61 64 68 68 69 20 21 122 114 134 118 109 80 40 101 20 65 56 98 60 123 117 14 18 66 20 126 53 45 118 120 103 116 128 125

(continued on next page)

276

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Appendix A (continued)

Sample I.D.

Source

Visual sourcing code

Rb

Sr

Y

Zr

Nb

dep049 dep050 dep051 dep052 dep053 dep054 dep055 dep056 dep057 dep058 dep059 dep060 dep061 dep062 dep063 dep064 dep065 dep066 dep068 dep069 dep070 dep071 dep072 dep073 dep074 dep075 dep076 dep077 dep078 dep079 dep080 dep081 dep082 dep083 dep084 dep085 dep086 dep087 dep088 dep089 dep090 dep091 dep092 dep093 dep094 dep095 dep096 dep097 dep098 dep099 dep100 dep101 dep102 dep103 dep105 dep106 dep107 dep108 dep109 dep110 dep112

La Joya Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio San Juan de los Arcos Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio La Joya La Joya Macrogroup 2 La Joya La Joya La Joya Pachuca Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas Pachuca Pachuca Pachuca La Joya Pachuca La Joya Pachuca Pachuca Pachuca Pachuca Ixtlan del Rio Ixtlan del Rio La Joya Pachuca La Joya Pachuca Pachuca Pachuca Pachuca La Joya Pachuca La Joya La Joya La Joya La Joya Volcan las Navajas La Joya Macrogroup 2 Volcan las Navajas Volcan las Navajas Ixtlan del Rio Macrogroup 2 Ixtlan del Rio Volcan las Navajas La Joya Volcan las Navajas San Juan de los Arcos

9 3 3 3 3 3 3 3 5 5 5 5 5 2 2 2 2 2 2 2 1 1 1 1 8 8 8 8 8 8 8 8 8 8 3 3 2 2 2 2 2 2 2 6 6 2 2 2 2 1 1 1 1 1 5 9 3 1 2 1 3

160 108 115 122 100 126 123 111 116 72 104 84 81 156 172 123 143 180 210 210 181 186 143 145 205 192 202 148 204 151 186 178 191 202 124 92 170 202 154 195 210 198 199 162 215 186 164 173 186 224 128 99 155 205 99 141 104 188 153 176 116

4 89 104 99 93 110 51 101 103 60 96 76 76 6 8 9 5 6 4 7 13 16 11 11 8 7 6 7 8 8 7 5 7 7 24 81 5 8 7 6 6 6 6 5 5 7 3 4 7 11 12 7 30 19 93 1 87 14 7 15 48

74 17 21 17 20 19 19 20 22 18 22 15 20 75 78 48 68 75 103 103 133 129 95 111 117 107 105 83 114 84 106 96 96 110 20 19 76 102 86 107 102 107 105 88 111 81 72 80 84 156 71 70 98 143 21 61 21 146 70 122 16

762 149 175 177 162 160 169 176 169 120 170 139 145 730 783 522 688 788 841 873 1055 1108 902 899 924 856 871 654 907 693 845 832 830 871 121 147 741 883 670 873 844 876 892 703 896 843 745 797 807 1273 716 556 905 1217 167 611 164 1167 736 1097 153

63 12 20 18 17 16 20 18 21 10 21 16 14 59 65 45 53 71 82 93 122 129 96 98 95 91 88 66 94 77 85 87 77 91 17 15 60 100 65 90 89 85 86 67 83 70 60 61 72 144 67 56 110 142 18 45 19 133 62 120 18

277

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279 Appendix A (continued)

Sample I.D.

Source

Visual sourcing code

Rb

Sr

Y

Zr

Nb

dep113 dep114 dep115 dep116 dep117 dep118 dep119 dep120 dep121 dep122 dep123 dep124 dep125 dep126 dep127 dep128 dep129 dep131 dep132 dep133 dep134 dep135a dep135b dep136 dep137 dep138 dep139 dep140 dep141 dep142 dep143 dep144 dep145 dep146 dep147 dep148 dep149 dep150 dep151 dep152 dep153 dep154 dep155 dep156 dep157 dep158 dep159 dep160 dep161 dep162 dep163 dep164 dep165 dep166 dep167 dep168 dep169 dep170 dep171 dep172 dep173

San Juan de los Arcos San Juan de los Arcos unassigned Pachuca San Juan de los Arcos Pachuca Pachuca Pachuca Pachuca Volcan las Navajas Pachuca Volcan las Navajas Volcan las Navajas Ixtlan del Rio Ixtlan del Rio La Joya La Joya Ixtlan del Rio Volcan las Navajas Volcan las Navajas Volcan las Navajas Ixtlan del Rio Ixtlan del Rio La Joya La Joya La Joya La Joya La Joya Macrogroup 2 Ixtlan del Rio Volcan las Navajas Volcan las Navajas Volcan las Navajas Volcan las Navajas La Joya La Joya La Joya Osotero La Joya unassigned Ixtlan del Rio Boquillas Volcan las Navajas Volcan las Navajas La Joya La Joya La Joya La Joya Macrogroup 2 La Joya La Joya La Joya Macrogroup 2 La Joya La Joya Macrogroup 2 Ixtlan del Rio Volcan las Navajas Ixtlan del Rio La Joya La Joya

3 3 3 3 3 2 2 2 2 1 8 1 1 3 3 2 2 3 6 1 1 3 5 2 2 2 2 2 2 3 1 1 1 1 6 2 2 3 2 5 3 3 1 1 6 6 6 6 6 6 2 2 2 2 2 2 3 1 3 2 6

103 106 155 195 135 231 224 216 212 205 190 184 175 111 111 169 192 130 200 169 161 97 111 182 184 162 176 171 113 107 168 162 192 171 179 160 179 113 173 198 107 73 183 143 166 177 172 165 134 188 178 156 110 162 158 103 111 159 97 168 169

44 45 16 7 60 10 9 6 7 12 8 12 11 102 109 4 7 118 9 15 12 88 92 5 5 6 3 5 4 101 15 14 17 15 5 4 4 104 3 10 98 54 15 15 7 6 7 4 6 5 5 3 4 7 4 5 86 15 89 5 5

21 21 31 110 23 122 120 111 119 152 103 123 115 23 22 72 79 18 141 124 129 21 18 77 78 65 68 72 45 22 122 127 132 109 75 67 75 21 78 33 21 13 116 112 64 81 78 70 56 82 84 64 48 76 73 51 17 109 24 73 72

149 157 149 902 185 965 963 916 937 1294 843 1020 973 180 180 751 795 180 1198 1041 1067 159 174 812 808 773 766 751 494 164 1019 990 1141 967 738 751 784 187 739 139 179 136 1043 927 762 801 781 747 569 805 774 709 502 750 740 522 166 954 170 747 773

16 17 14 89 21 97 93 96 95 132 88 97 97 20 19 65 63 19 126 113 129 19 21 72 70 66 72 66 44 17 113 108 125 105 65 69 71 13 64 18 16 11 120 98 62 64 68 65 45 72 59 61 38 59 63 41 18 106 18 64 65

(continued on next page)

278

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279

Appendix A (continued)

Sample I.D.

Source

Visual sourcing code

Rb

Sr

Y

Zr

Nb

dep174 dep175 dep176 dep177 dep178 dep179 dep180 dep181 dep182 dep183 dep184 dep185 dep186 dep187 dep188 dep189 dep190 dep191 dep192 dep194 dep195 dep196 dep197 dep198 dep199 dep200 dep201 dep202

Macrogroup 2 Boquillas Ixtlan del Rio Ixtlan del Rio La Joya La Joya La Joya La Joya Macrogroup 2 Volcan las Navajas Volcan las Navajas La Joya La Joya La Joya Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Volcan las Navajas Pachuca Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio Ixtlan del Rio La Joya La Joya Osotero

9 3 3 3 2 2 2 2 6 1 1 2 2 2 3 3 3 1 8 5 5 3 3 3 3 2 2 9

129 78 103 106 154 181 187 157 121 182 176 170 158 176 108 121 108 152 207 122 103 104 97 114 106 142 179 112

3 58 100 90 3 4 7 5 2 12 18 6 5 8 87 99 84 13 10 97 103 97 92 104 96 6 8 85

68 16 23 19 73 73 84 73 56 129 129 78 68 75 21 22 22 105 110 21 21 19 24 18 24 65 81 18

648 139 178 162 707 809 819 732 575 1078 1108 814 741 795 166 182 172 931 903 171 174 175 174 186 174 687 803 186

49 12 20 20 58 72 71 63 53 129 124 63 69 65 17 21 15 103 90 22 18 20 18 20 21 60 63 14

References Andrefsky Jr., William, 2005. Lithics: Macroscopic Approaches to Analysis, second ed. Cambridge Manuals in Archaeology, Cambridge University Press. Anguiano, M., 1992. Nayarit: Costa y Antiplanicie en el Momento del Contacto. Universidad Nacional Autonoma de Mexico, Mexico City. Aoyama, K., 1991. Litica. In: Nakamura, S., Aoyama, K., Uratsuji, E. (Eds.), Investigaciones arqueologicas en la region de La Entrada, vol. 2. Servicio de Voluntarios Japoneses para la Cooperacion con el Extranjero y el Instituto Hondureno de Antropología e Historia, San Pedro Sula, pp. 39–203. Braswell, G.E., Andrews, E.W.V., Glascock, M.D., 1994. The Obsidian artifacts of Quelepa, E1 Salvador. Ancient Mesoamerica 5, 173–192. Braswell, G.E., Clark, J.E., Aoyama, K., McKillop, H.I., Glascock, M.D., 2000. Determining the geological provenance of obsidian artifacts form the Maya region: a test of the efficacy of visual sorting. Lat. Am. Antiq. 11, 269–282. Carpio Rezzio, E.H., 1993. Obsidian at Balberta. In: Bove, F.J., Medrano, S., Lou, B., Arroyo, B. (Eds.), The Balberta Project: The Terminal Formative – Early Classic Transition on the Pacific Coast of Guatemala. Memoirs in Latin American Archaeology, No. 6. University of Pittsburgh, Pittsburgh, pp. 83–106. Clark, J.E., 1985. Platforms, bits, punches, and vises: a potpourri of Mesoamerican blade technology. Lithic Technol. 14, 1–15. Clark, J.E., Bryant, D.D., 1997. A technological typology of prismatic blades and debitage from Oje de Agua, Chiapas, Mexico. Ancient Mesoamerica 8, 111–136. Darling, J.A., 1993. Notes on obsidian sources of the southern Sierra Madres occidental. Ancient Mesoamerica 4, 245–253. Darling, J.A., 1998. Obsidian Distribution and Exchange in the North-Central Frontier of Mesoamerica. Unpublished Ph.D. Dissertation, Department of Anthropology, University of Michigan, Ann Arbor. Darling, J.A., Hayashida, F.M., 1995. Compositional analysis of the Huitzila and La Lobera obsidian sources in the southern Sierra Madre Occidental, Mexico. J. Radioanal. Nucl. Chem. 196, 245–254. Duverger, C., 1998. Coamiles, Nayarit: hacia una periodizacion. En Antropología e Historia del Occidente de Mexico. Memorias de la XXIV Mesa Redonda de las Sociedad Mexicana de Antropología, vol. 1. Universidad Nacional Autonoma de Mexico y Sociedad Mexicana de Antropología, Mexico City, pp. 609–628. Esparza Lopez, Rodrigo, 2008. Los Yacimientos de Obsidiana de el Pedernal-La Mora: Una Explotacion Constante durante el Desarrollo de la Tradicion Teuchitlan. In: Weigand, P.C., Beekkman, C., Esparza, R. (Eds.), Los Yacimientos de Obsidiana de el Pedernal-La Mora: Una Explotacion Constante durante el Desarrollo de la Tradicion Teuchitlan. In Tradicion Teuchitlan. El

Colegio de Michoacan y el Secretaria de Cultural del Estado de Jalisco, Zamora and Guadalajara, pp. 143–166. Gamez Eternod, Lorena, Garduño Ambriz, Mauricio, 2003. Rescate Arqueologico CECYTEN, Plantel 06 San Felipe Aztatán, Municipio de Tecuala (Nayarit). Reporte tecnico de los trabajos de sondeo/Analisis de materiales arqueologicos. Archivo Tecnico del Centro INAH Nayarit, Tepic. Garduño Ambriz, Mauricio, 2006. Investigaciones arqueologicas en Cerrro de Coamiles, Nayarit. Reporte tecnico final de la primera temporada de campo (2005). Archivo Tecnico del Centro INAH Nayarit, Tepic. Garduño Ambriz, Mauricio, 2007. Arqueologia de rescate en la cuenca inferior del rio Acaponeta. Diario de Campo (Boletin Interno de los Investigadores del Area de Antropologia, CONACULTA-INAH) 92, 36–52. Garduño Ambriz, Mauricio, Gamez Eternod, Lorena, 2005. Programa Emergente de Rescate Arquelogico en San Felipe Aztatán, Municipio de Tecuala (Nayarit). Informe Tecnico Final/Trabajos de Sondeo Arqueologico. Archivo Tecnico Centro INAH Nayarit, Tepic. Glascock, M.D., Weiland, P.C., Esparza Lopez, R., Ohnersorgen, M.A., Garduño Ambriz, M., Mountjoy, J.B., Darling, J.A., 2010. Geochemical Characterization of obsidian sources in western Mexico: the sources of Jalisco, Nayarit, and Zacatecas. In: Kuzmin, Y.V., Glascock, M.D. (Eds.), Crossing the Straights: Prehistoric Obsidian Source Exploitation long the Pacific Rim. Archaeopress, BAR International Series S2152, Oxford, U.K., pp. 201–217. Heller, L., Stark, B.L., 1998. Classic and postclassic obsidian tool production and consumption: a regional perspective from Mixtequilla, Veracruz. Mexicon 20 (6), 119–128. Jiminez Betts, P.F., Darling, J.A., 2000. Archaeology of Southern Zacatecas: the Malpaso, Juchipila, and Valparais-Bolanos Valleys. In: Foster, M.S., Gorenstein, S. (Eds.), Greater Mesoamerica: The Archaeology of West and Northwest Mexico. The University of Utah Press, Salt Lake City, pp. 155–180. Johnson, Jay K., 1989. The utility of production trajectory modeling as a framework for regional analysis. In: Henry, Donald O., Odell, George H. (Eds.), Alternative Approaches to Lithic Analysis. Archaeological Papers of the American Anthropological Association, pp. 119–138. Kelley, J.C., 2000. The Aztatlán mercantile system: mobile traders and the northwestern expansion of Mesoamerican civilization. In: Foster, Michael, Gorenstein, Shirley (Eds.), Greater Mesoamerica: The Archaeology of West and Northwest Mexico. University of Utah Press, Salt Lake City, pp. 137–154. Levine, Marc, n.d. Evaluating a New Technique for Obsidian Provenance Sourcing: Combining ‘‘Visual grouping” with Neutron Activation Analysis. Unpublished manuscript. University of Colorado at Boulder. McKillop, H.I., 1995. The role of northern Ambergris Caye in Maya obsidian trade: evidence from visual sourcing and blade technology. In: Guderjan, T.H., Garber,

D.E. Pierce / Journal of Anthropological Archaeology 40 (2015) 266–279 J.F. (Eds.), 1988, Maya Maritime Trade, Settlement, and Populations on Ambergris Caye, Belize. Maya Research Program and Labyrinthos Lancaster, Pennsylvania, pp. 163–174. Meighan, C.W., 1976. The Archaeology of Amapa. In: Nayarit (Ed.), Monumenta Archaeologica, vol. 2. The Institute of Archaeology, University of California, Los Angeles. Moholy-Nagy, H., Nelson, F.W., 1990. New data on sources of obsidian artifacts from Tikal, Guatemala. Ancient Mesoamerica 1, 71–80. Morrow, Carol A., 1984. A biface production model for gravel-based chipped stone industries. Lithic Technol. 13, 20–28. Mountjoy, J.B., 2000. Prehispanic cultural development along the southern coast of West Mexico. In: Foster, M.S., Gorenstein, S. (Eds.), Greater Mesoamerica: The Archaeology of West and Northwest Mexico. The University of Utah Press, Salt Lake City, pp. 81–106. Odell, George, 2004. Lithic Analysis. Plenum Publishers, New York. Ohnersorgen, Michael A., 2004. Informe Tecnico Final de Proyecto ‘‘Investigaciones Arqueologica Preliminar de Chacalilla, Nayarit, Mexico, Temporada 2003”. Manuscript on file, Instituto Nacional de Antropología y Historia (INAH). Centro INAH Nayarit, Tepic. Ohnersorgen, Michael A., 2007. La organizacion socio-economica y la interaccion regional de un centro Aztatlán: Investigaciones arqueologicos en Chacalilla, Nayarit. Informe Tecnico Parcial, Temporada de 2005. Manuscript on file. Instituto Nacional de Antropología y Historia (INAH), Centro INAH Nayarit, Tepic. Ohnersorgen, M.A., Glascock, M.D., Mountjoy, J.B., Garduño, M.A., 2012. Chemical Analysis of Obsidian Artifacts from the West Coast of Mexico: Implications for Regional and Interregional Economic Organization. Manuscript on file. Department of Anthropology, University of Missouri, Saint Louis. Pollard, Helen Perlstein, 2000. Tarascan external relationships. In: Foster, M.S., Gorenstein, S. (Eds.), Greater Mesoamerica: The Archaeology of West and Northwest Mexico. The University of Utah Press, Salt Lake City, UT, pp. 71–80. Railey, Jim A., Gonzales, Eric J., 2014. The problem with flake types and the case for attribute analysis of debitage. In: Shott, Michael J. (Ed.), Works in Stone: Contemporary Perspectives on Lithic Analysis. The University of Utah Press, pp. 11–33. Sanchez Polo, R., 1991. Las navajas de obsidiana de Kaminlajuyu’/San Jorge: un estudio tecnologico funcional. Unpublished licenciatura thesis. Escuela de Historia, Area de Arqueologia, Universidad de San Carlos Borromeo de Guatemala, Guatemala. Sanders, Paul H., 1992. Archaeological Investigations along the Pend Oreille River: Site 45PO149. Center for Northwest Anthropology, Project Report 18. Washington State University, Pullman.

279

Santley, R.S., Kerley, J.M., Kneebone, R., 1986. Obsidian working, long-distance exchange, and the politico-economic organization of early states in Central Mexico. In: Isaac, B.L. (Ed.), Research in Economic Anthropology, Suppl. 2. JAI Press, Greenwich, pp. 101–132. Sauer, C.O., Brand, D.D., 1932. Aztatlán: Prehistoric Mexican Frontier on the Pacific Coast. Ibero-Americana 1. University of California Press, Berkeley. Scott, S.D., Foster, M.S., 2000. The prehistory of Mexico’s Northwest Coast: a view from the Marismas Nacionales of Sinaloa and Nayarit. In: Foster, M.S., Gorenstein, S. (Eds.), Greater Mesoamerica: The Archaeology of West and Northwest Mexico. The University of Utah Press, Salt Lake City, UT, pp. 107–135. Smith, M.E., Heath-Smith, C.M., 1980. Waves of influence in post-classic Mesoamerica? A critique of the Mixteca-Puebla concept. Anthropology 4, 15– 50. Stark, B.L., Heller, L., Glascock, M.D., Elam, J.M., Neff, H., 1991. Obsidian-artifact source analysis for the Mixtequilla region, south-central Veracruz, Mexico. Lat. Am. Antiq. 3, 221–239. Trombold, C.D., Luhr, J., Hasenaka, T., Glascock, M., 1993. Chemical characteristics of obsidian from archaeological sites in Western Mexico and the Tequila source area: implications for regional and pan-regional interaction within the northern Mesoamerican periphery. Ancient Mesoamerica 4, 255–270. Tykot, Robert H., 2003. Determining the source of lithic artifacts and reconstructing trade in the ancient world. In: Nick Kardulias, P., Yerkes, Richard W. (Eds.), Written in Stone. Lexington Books, Oxford, UK, pp. 59–86. VanPool, T.L., VanPool, C.S., Rakita, G.F.M., 2008. Birds, bells, and shells: the long reach of the Aztlatlán trading tradition. In: Nielsen, Glenna (Ed.), Touching the Past: Traditions of Casas Grandes. BYU University Press, Provo, Utah, pp. 5–14. Walker, Danny N., Todd, Lawrence C., 1984. Anthropological Salvage at 48FR1398: Castle Gardens Access Road Site, Fremont County, Wyoming. Occasional Papers on Wyoming Archaeology No. 2, Laramie. Weigand, P.C., 1989. The Obsidian mining complex at La Joya, Jalisco. In: Gaxiola, G. M., Clark, J.E. (Eds.), La obsidian en Mesoamerica, Coleccion Cientifica, Serie Arqueologica, No. 176. INAH, Mexico City, pp. 205–211. Weigand, P.C., Spence, M.W., 1982. The Obsidian mining complex at La Joya, Jalisco. In: Weigand, P.C., Gwynne, G. (Eds.), Mining and Mining techniques in Ancient Mesoamerica. Anthropology, vol. 6. State University of New York, Stony Brook, pp. 175–188 (special issue). Zeir, Christain J., Kalasz, Stephen M., Peebles, Anne H., Van Ness, Margaret A., Anderson, Elaine, 1988. Archaeological Excavation of the Avery Ranch Site (5PE56) on the Fort Carson Military Reservation, Pueblo County, Colorado. Submitted to USDI National Park Service, Rocky Mountain regional. Fort Collins. Centennial Archaeology Inc., Colorado.