Chapter 8

Zaragoza-Oyameles Obsidian Projectile Points

Cantona’s Place in Early Classic Period Long-Distance Gift Exchange and Interaction

Charles L. F. Knight

The adage “no one remembers who came in second place” has a special applicability to archaeology of the central Mexican highlands during the Classic period. By the end of the Early Classic period around AD 550 Teotihuacan was the most prominent and populous urban settlement in the Americas, with an estimated population from 100,000–200,000 and a site size covering 2,000 ha (Millon 1976:212, but see Cowgill [2015:143] for a revised estimate). Its direct and indirect influence throughout Mesoamerica has received intensive archaeological attention for more than half a century (Braswell 2003a; see Braswell 2003c for an overview of the subject), reflecting the wide array and nature of long-distance exchange, interaction, and influence between people of many disparate regions. At the same time, the second-most-populous city in the Classic period central highlands, which for more than twenty-five years has received intensive archaeological investigation, has typically been overlooked in discussions on central Mexican archaeology of the Early Classic period. This city is Cantona, located in the Cuenca Oriental of eastern Puebla, 37 km northwest of the Cofre de Perote at the edge of the central Mexican highlands (figure 8.1). Although these two cities are located less than 145 km apart and are separated by the easily traversable Tlaxcala Corridor land route, the nature and degree of interaction between them are not well understood. Little in the literature has considered the political, economic, or ideological role or impact of Cantona in the Mexican highlands during the Early Classic period. Publications in English and Spanish on the archaeology of the site (Ferriz 1985; García Cook 2003, 2014; García Cook and Merino Carrión 1998; García Cook et al. 2010; Merino Carrión and García Cook 2007; Rojas Chávez 2001) paint a picture of a major center participating in exchange and communication networks both parallel to and separate from those of Teotihuacan.

Like Teotihuacan, the economic basis of Cantona is argued to have been the control over the extraction and initial reduction of obsidian, and the subsequent long-distance exchange of obsidian polyhedral cores and prismatic blades (García Cook and Carrión 1998:210; García Cook 2003, 2014; García Cook and Merino Carrión 2005; García Cook et al. 2010). The source area Cantona is believed to have controlled is Zaragoza-Oyameles, located 13 km to the north. Using recent data on the bifacial technologies recovered from the Zaragoza-Oyameles source area and published data on excavations undertaken at Cantona itself, this chapter attempts to flesh out the nature of interregional interaction between the two most populous urban centers in the Early Classic period highlands. From a Cantona-based perspective, the occupation of the central Mexican highlands during the Early Classic period becomes more nuanced, recognizing political, economic, and ideological exceptions and pockets of resistance to the gravitational pull of Teotihuacan on the occupation of the region.

Background

The city of Cantona was built on the Tepeyahualco lava flow, or malpaís, the uppermost of a series of andesite lava flows that emanated from the Caldera Humeros during a series of eruptions 40,000–60,000 years ago (Ferriz 1985:363–364; figure 8.2). The site has been divided into three zones by archaeologists—a northern, middle, and southern zone—based on the distribution of architecture and the natural extent of the lava flow. Archaeological attention has been concentrated in the southern zone, where the level of preservation is best, and that also contains the civic-ceremonial core of the site (García Cook 2003:317). All architecture at Cantona is constructed from the porous tezontle (a volcanic lava rock) found throughout the area. Site architecture was made without the aid of mortar or plaster. Rather, buildings were made from dry-laid tezontle of different colors to produce visual contrasts: black-gray for roads, stairs and buildings, temples, and so on; and red for some building facades. Site layout utilized the natural topography of the overlapping, 13–75 m thick lava flows. For instance, the highest points of the terraces formed by the terminal edge of the upper flow in the southern zone produced a naturally elevated terrace upon which the civic-ceremonial core was constructed. This also provided sweeping views of the older, basal flows below, where much domestic, nonelite habitation exists.

Unlike the astronomically aligned grid pattern of Teotihuacan’s layout, the internal layout of Cantona is asymmetrical, defined by a series of walled compounds connected by walled streets that follow the natural topography of the lava flows. There are an estimated 7,500 of these elite and nonelite walled compounds (or patios) throughout Cantona; more than 1,500 streets connect the compounds, and seventeen broader causeways exit the city and lead to other destinations (García Cook 2003:319–325; García Cook and Zamora Rivera 2010:34).

The contrast in the physicality of the contemporaneous cities of Cantona and Teotihuacan is considerable and may represent more than just pragmatic responses to environmental constraints. For instance, in his discussion on the Early Classic period construction at Cholula, McCafferty (2007:454) has suggested that the lack of similarities with the contemporaneous Teotihuacan architectural canon may represent an “intentional public rejection of its ideological ‘empire’ and an expression of separate identity.” While the natural topography of Cantona would have created sizable physiographic limitations on how the city could have been constructed, a conscious rejection of Teotihuacan’s identity also may underlie the city’s layout and architectural style.

From 1992-2016, Dr. Ángel García Cook directed a program of survey and excavation at Cantona, with excavations focused on the civic-ceremonial site core and in the adjacent “suburbs” (García Cook 2003; García Cook and Merino Carrión 1998; García Cook and Vackimes Serret 2014:220–222). More than 100 radiocarbon dates have been produced from excavated contexts within the civic-ceremonial core of the city (García Cook and Zamora Rivera 2010:33), dating the original settlement to the Middle Formative period (ca. 600 BC), when it was a small hamlet. By 300 BC the site occupation grew to a population of 10,000 and covered 200 ha. In the late Early Classic period around AD 400, the population is estimated at 50,000 covering 670 ha, and then reaches its apogee of 90,000 inhabitants, covering 1430 ha, in the Late Classic period around AD 750 (García Cook 2003:339). In comparison, the Epiclassic occupation of Xochicalco during the Gobernador phase (AD 650–900) was a time when Xochicalco grew and floresced “into a major administrative center in Central Mexico” covering 400 ha, with an estimated population of 9,000–15,000 people (Hirth 2000:68–69). Cholula, one of the longest-occupied urban centers in Mesoamerica, had a population that covered approximately 4 km² during its Stage 2 construction phase (AD 200–600) during the Early Classic period (McCafferty and Peuramaki-Brown 2007:107). Although population estimates have not been made for this Early Classic period occupation, the size covered by its population at this time is similar to that of Xochicalco. Cholula’s Early Classic occupation does not appear to have been influenced, at least in terms of style or site orientation, by Teotihuacan (McCafferty 2007:454). Both Xochicalco and Cholula grew substantially in the later Epiclassic period after the fall of Teotihuacan (Hirth 2000:68; McCafferty and Peuramaki-Brown 2007:107), but neither of these sites witnessed the degree of growth that occurred at Cantona during the same period, when its population and size more than doubled from the earlier Early Classic period. In short, there is nothing in the central Mexican highlands outside of Teotihuacan itself during the Early Classic period that compares to Cantona in estimated size and population.

One result of twenty years of excavation at Cantona has been the identification of over 350 obsidian reduction activity areas in a 19 ha zone adjacent to the civic-ceremonial core of the city (García Cook 2014:107; García Cook et al. 2010:219). The proximity of these workshops to the civic-ceremonial core and their physical location directly in, yet below the line-of-sight from, the core has resulted in their interpretation as “state workshops” (García Cook 2003:337, 2014:107–110). The excavations of four of these state workshops indicate that prismatic cores and blades were primarily produced for local elite consumption and export beyond Cantona (García Cook et al. 2010:219; García Cook 2014). Broad site survey also suggests that widespread domestic production of utilitarian obsidian implements occurred throughout the site (Rojas Chávez 2001). Green Pachuca obsidian, closely associated with the Teotihuacan polity during the Early Classic period, is not entirely absent from Cantona, but is quite rare. García Cook (2014:139) states that a “mínima expresión” of the material was recovered from site excavations in Cantona’s core and it was found typically in blade form. Of 51,677 obsidian artifacts analyzed from the site survey, only four were made from green obsidian, the same amount as made from chert (Rojas Chávez 2001:322, table 26). In addition, a single green projectile point in the Shumla A style (either Stemmed-A or Stemmed-B, following the Teotihuacan classification [Spence 1996], see below) was recovered from the broader site survey (Rojas Chávez 2001:223).

The ceramic record at Cantona (García Cook and Merino Carrión 1998; Gómez Santiago 2010) reveals little interaction with Teotihuacan, either directly through ceramic trade wares, or indirectly via local ware emulation. Thin orange wares, figurines, or vessels supports in the Teotihuacan tradition are almost nonexistent at Cantona. Rather, the city’s ceramic tradition demonstrates a strong local flavor, with a standardized and hyperconservative range of vessel shapes, pastes, and decoration appearing throughout the site’s long occupation. At the same time, some form of connectivity with far-off Mesoamerican polities is evident through the recovery of vessels in styles representative of southern Veracruz, Oaxaca, the Mixteca Poblana, the Bajío, and Campeche (García Cook and Merino Carrión 1998). In addition, a variety of whole and partial shells and shell ornaments were recovered from the civic-ceremonial core, the majority of which date to the late Cantona I to Cantona II period occupations, or 350 BC–AD 600 (García Cook and Vackimes Serret 2014:227). These shell artifacts reflect the involvement of the Cantona elite in trade networks with the Gulf lowlands, the Caribbean, and the Pacific coast (García Cook and Vackimes Serret 2014:239–240). Thus the ceramic and shell data suggest Cantona’s independent involvement in pan-Mesoamerican exchange networks. On the other hand, the obsidian industry, which played such an important part in the economies of both cities, may reveal connectivity not otherwise seen.

Zaragoza-Oyameles Obsidian Source Area Survey

In the winters of 2012–2014, I directed an intensive surface survey and surface collection program of the Zaragoza-Oyameles source area (Knight 2012, 2013, 2015; Knight et al. 2017; figure 8.3). The survey consisted of seven archaeologists walking along parallel transects spaced 5 m apart in plowed and unplowed fields. The locations of all surface tools—such as projectile points, cores, bifaces, and scrapers, as well as all ground stone and ceramics—were marked with a hand-held GPS and then collected. Once identified, sites were either 100 percent surface collected or with a proportional random sample of 5 × 5 m surface units, depending on site size. In total, 1,534 ha (3,790 acres) were surveyed in this manner, resulting in the identification of 48 primary reduction sites, three habitation sites, 50 nonquarry reduction sites, 117 obsidian exposures, 77 surface extraction pits, and one extraction trench.

The surface survey also resulted in the collection of eighty-five complete and partial obsidian projectile points. Following García Cook’s (1967) typology, several types of projectile points, likely arrow points, were identified. These include Hidalgo (Early Classic), Pedernales (Late Classic), Santa Clara (Early-Late Classic), Tecolote (Early-Late Classic), and Texcoco A (Late Classic-Postclassic) projectile points. Additionally, fifteen are similar in outline to the Early Classic, central Mexican points identified at Teotihuacan as Stemmed-A and Stemmed-B by Spence (1996:fig. 2) or, using Tolstoy’s (1971:fig. 2) typology, as Shumla A and Gary (figure 8.4). Tolstoy identified the Stemmed-B (Shumla A) style as most common in the Teotihuacan II period. In García Cook’s (1967:138, plate 11, tables 10, 29–33) comparative analysis of projectile points, he found that the Stemmed-B (Shumla A) point style was most common in the Basin of Mexico throughout the Classic period, AD 350–1100. Finally, in the point typology developed by Sarabia (1996, referenced in Gazzola [2014:227]) for Teotihuacan projectile points, these two point styles are categorized as within the Family C style (Gazzola 2014:fig. 9).

For the discussion here, I use Spence’s original, typologically neutral terms of Stemmed-A and B.

At Teotihuacan, finely made points of the Stemmed-A and Stemmed-B variety are closely associated with the military and elite (Carballo 2011:133–145, 159–163; Parry 2014:292; Sugiyama 1989), as several caches containing many examples of these point styles have been recovered from the Moon and Feathered Serpent Pyramids. In addition, points in these styles made from central Mexican obsidian have been recovered from burials and caches at several Mayan centers, such as Tikal (Moholy-Nagy et al. 1984), Altun Ha (Pendergast 2003:238, fig. 9.2, 1990:fig. 121), Caracol (Chase and Chase 2011:10), Uxmal (Braswell 2013:164), and Calakmul (Braswell and Glascock 2011:129, note 1), as well as at Balberta on the Pacific coast of Guatemala (Bove and Medrano Busto 2003:50), and Mirador Mound 20 in Chiapas, Mexico (Agrinier 1970:39, 67, figs. 52, 86). At the same time, less finely made versions of these point styles have also been found in domestic contexts at Teotihuacan, such as gray specimens recovered in the Oztoyahualco compound (Hernández 1993:409, fig. 292).

Obsidian evidence for interaction between Teotihuacan and Cantona is not overwhelming. At Cantona, a single Stemmed-A point made from green obsidian was recovered during site excavations, but no description of those excavations is provided (Rojas Chávez 2001:223), while obsidian from the Zaragoza-Oyameles source was found in the Moon Pyramid dart point workshop materials analyzed by David Carballo (Carballo et al. 2007:40). Thus, there appears to have been some degree of obsidian exchange occurring between Teotihuacan and Cantona, but how this apparent interaction relates to the production of the central Mexican style dart points is unclear.

Chemical Characterization and Flaking Patterns

In a small number of cases in which chemical characterization of gray obsidian projectile points (Stemmed-A and/or Stemmed-B, and undescribed styles) found along with green Pachuca points in Early Classic ritual/elite contexts in eastern Mesoamerica have been carried out, several were found to have been made from Zaragoza-Oyameles obsidian (Bove and Medrano Busto 2003:53; Moholy-Nagy et al. 1984). All surface points collected during the Zaragoza-Oyameles Regional Obsidian Survey were chemically characterized using portable-XRF by the Missouri University Research Reactor (Knight et al. 2017). The results show that all were made from Zaragoza-Oyameles obsidian, thus were locally produced. Surface artifacts systematically collected at the Zaragoza-Oyameles source area included all stages of bifacial reduction, from large blades and flakes to preforms and finished products, as well as the debitage resulting from their production (Knight 2012, 2013, 2015). Evidence from the Cantona survey and excavation indicates that both Stemmed-A and Stemmed-B varieties in gray obsidian were consumed there (Rojas Chávez 2001:223). As a result, the data suggest several scenarios for the introduction of gray Stemmed-A or Stemmed-B points into elite and ritual contexts in eastern Mesoamerica. The first is that Stemmed-A and Stemmed-B points could have been produced at the source area, under the auspices of the Cantona elite or independently by local producers. These could have been exchanged with Teotihuacanos who, in turn, gifted them to the Maya. However, ceramic data from Cantona indicates that the Cantona elite actively participated in long-distance exchanges with eastern Mesoamerica and, therefore, could have exchanged such points with them directly and independently of Teotihuacan.

Another scenario envisions that Zaragoza-Oyameles material made its way into Teotihuacano workshops, where the points would have been produced by Teotihuacano artisans. As mentioned above, at least two examples of Zaragoza-Oyameles obsidian were recovered from the Teotihuacan dart point workshop (Carballo et al. 2007:40). On the other hand, data on obsidian exchange networks during the Early Classic period from areas immediately adjacent to the Basin of Mexico suggest that little, if any, Zaragoza-Oyameles obsidian was brought into the Basin of Mexico. At several sites located within the Tlaxcala corridor, Carballo and colleagues (2007) found that Zaragoza-Oyameles obsidian was common in Formative period occupations, but beginning in the Classic period was replaced by obsidians associated with Teotihuacan, such as Otumba and Pachuca. Outside of the Basin of Mexico and adjacent areas, the proportion of Zaragoza-Oyameles obsidian increases substantially in Early Classic assemblages, such as in the Tehuacán Valley (Drennan et al. 1990:188–189), in the Valley of Oaxaca (Elam 1993; Pires-Ferreira 1975), in the Lower Rio Verde Valley of Oaxaca (Joyce et al. 1995), and in the Isthmus of Tehuantepec (Zeitlin 1982). In the southern Gulf lowlands, Zaragoza-Oyameles obsidian completely dominates the chipped stone assemblages of consumption sites at this time (Knight and Glascock 2009; Stark et al. 1992, Santley et al. 2001).

William Parry (2014:292) has observed that many of the points cached at Teotihuacan were more finely made than similar point styles used in domestic contexts. Attributes that indicate such fine finishing include biface symmetry, thinness, and parallel pressure flaking on both faces. Hirth and colleagues (2003:147) have noted that the vast majority of central Mexican fine points exhibit diagonal pressure flaking from the lower left to upper right on each face (figure 8.5). This pressure flaking pattern is the result of the way in which the point is held during flake removal and the angle of the pressure tool, assuming the knapper was right-handed (Hirth et al. 2003:148–150). The lower-left to upper-right patterning seems to be typical of the Early Classic period projectile points recovered from the Moon Pyramid (Carballo 2011:figs. 5.20, 7.1) and those from the Feathered Serpent Pyramid burials (Sugiyama 1989). In addition, the illustrated Stemmed-A green obsidian points recovered from elite caches at the Classic period Mayan site of Caracol all have the lower-left to upper-right patterning (Chase and Chase 2011:10, fig. 5).

In the examples recovered from Zaragoza-Oyameles, all points in the central Mexican Stemmed-A and Stemmed-B styles exhibit the attributes of fine flaking, but they differ in the direction of pressure flake removal. The points recovered from the Zaragoza-Oyameles source area are marked by an upper-left to lower-right diagonal flaking pattern. The predominance of the upper-left to lower-right diagonal flaking pattern in the Zaragoza-Oyameles examples suggests local variation in the method of pressure flake removal, assuming the knappers were right-handed. Thus, even if local knappers were emulating a central Mexican projectile point style, the local tradition of pressure flake removal differed for this point style. Of the other seventy projectile points recovered, most did not exhibit fine pressure flaking. Where it occurs, however, a variety of pressure flaking styles existed, often on the same point. These included the upper-left to lower-right diagonal flaking pattern, lower-left to upper-right diagonal, and perpendicular to the edge.

Discussion

One avenue to investigate whether the gray obsidian points found in contexts outside of central Mexico were made by Zaragoza-Oyameles artisans would be to determine whether they exhibit the upper-left to lower-right diagonal flaking pattern. Thus far, the published data are not conclusive. For instance, the base of a gray obsidian projectile point in the Stemmed-A style was recovered at Tres Zapotes from Classic period contexts (Hester et al. 1971:pl. 1-a; Weiant 1943:121, pl. 78–72). Since the vast majority of Classic period obsidian consumed at Tres Zapotes and elsewhere in the Gulf lowlands was from Zaragoza-Oyameles (Knight and Glascock 2009; Stark et al. 1992; Santley et al. 2001), it may be assumed that this point also is from that source. The flaking pattern shows removals perpendicular to the edge along the left margin and indeterminate on the right margin; perhaps the opposite face would show greater detail, but there is no published image of it that I am aware of.

Farther afield, the only published image of a chemically characterized Zaragoza-Oyameles point is from Tikal (Moholy-Nagy et al. 1984:fig. 3c, or see Moholy-Nagy 2003:fig. 65s for a slightly cleaner version), which exhibits a flaking pattern unlike either attributed to central Mexican points. While several of the pressure flake scars along the right margin of the illustrated example are oriented upper-left to lower-right, they are insufficient to characterize the entire point as exhibiting this flaking pattern. Of the four illustrated gray obsidian points in Tikal problematic deposit PNT-21 (Iglesias Ponce de León 2003:fig. 6.5a–d), two (a, b) illustrate the lower-left to upper-right flaking pattern of central Mexico, as do three of the four green points (e–g). None exhibit the upper-left to lower-right diagonal flaking pattern.

At Balberta, on Guatemala’s Pacific coast, Bove and Medrano Busto (2003:53) mention that three gray Zaragoza-Oyameles points were recovered in association with other gray points and a green point/effigy cacao cache in the site core, and separately in other elite contexts dating to the Early Classic period. The green points were made in the quintessential central Mexican style associated with Teotihuacan (Bove and Medrano Busto 2003:50), but no information is given on the general style or flaking pattern of the gray obsidian points. They (Bove and Medrano Bust 2003:52) add that fine paste wares also were recovered from the same contexts, which, according to the NAA characterization conducted on six samples, likely were produced in the Gulf lowlands.

While the green Pachuca points recovered at Balberta reflect some type of relationship with Teotihuacan or Teotihuacanos, the fine paste wares and Zaragoza-Oyameles obsidian points can be interpreted as reflecting connections with the Gulf lowlands and, perhaps, indirectly with Cantona. Another perspective is offered by Bove and Meddrano Busto (2003:52) who interpret the presence of the fine paste wares as evidence for possible indirect ties to Teotihuacan, via the Gulf lowlands centers related to Teotihuacan, such as Matacapan. However, our interpretations of the strength of Teotihuacan’s influence in the Gulf lowlands, and beyond, via Matacapan have undergone considerable revisions since first proposed by Santley (1983, 1989). For instance, data from the last twenty years in the Sierra de los Tuxtlas, and beyond, indicate that the distribution of Teotihuacan-related obsidian and ceramics was strongest within Matacapan’s regional hinterland (Braswell 2003b:111; Pool and Stoner 2004:94–97; Santley and Arnold 1996, 2005:190; Stoner 2012; cf. Philip J. Arnold III and Lourdes Budar, chapter 7 in this volume). While Santley and his colleagues (2001) have argued that Matacapan was a major node in the regional distribution of highland obsidians, there is little evidence from consumer sites beyond Matacapan’s hinterland of that role (Pool and Stoner 2004:82–86; Stark et al. 1992; Stoner 2012). Zeitlin (1982:268–269) pointed out a massive increase in the use of Zaragoza-Oyameles and, neighboring, Altotonga obsidians beginning in the Early Classic period in data from the southern Isthmus of Tehuantepec. This increase in Zaragoza-Oyameles and Altotonga obsidians corresponded to a virtual disappearance of El Chayal and Guadalupe Victoria obsidians that were so prevalent in the region previously. He posited that these two sources could have been controlled by the central Veracruz center of El Tajín and thus represented a parallel and contemporaneous distribution network to that of Teotihuacan. At the time, Zeitlin (1982:269) suggested that Teotihuacan may have had some indirect control over the distribution of these obsidians into the Isthmus region via El Tajín, thus creating a situation of “dual-administration.” We now recognize that it was Cantona, independent of Teotihuacan, and not El Tajín that was the real powerhouse behind the distribution of Zaragoza-Oyameles obsidian throughout Mesoamerica from at least the Early Classic through early Post-Classic periods (Braswell 2003d; García Cook 2003, 2014; García Cook et al. 2010).

At Kaminaljuyú, numerous gray obsidian points were recovered from a context that included fine central Mexican style points made of green obsidian (Kidder et al. 1946:137–138). The green points all exhibited the lower-left to upper-right flaking pattern, while the gray points exhibited either the upper-left to lower-right diagonal flaking pattern or a pressure flake removal pattern different from either of these. The only problem is that the gray points illustrated are not good examples of either Stemmed-A or Stemmed-B point styles, looking much cruder than those found at the Zaragoza-Oyameles source area. The gray Kaminaljuyú points may not represent any connection with central Mexico at all, but rather may have been made of gray El Chayal obsidian, the source closest to Kaminaljuyú. Geoffrey Braswell (2003b:130) suggests something similar, adding that he believes the gray Kaminaljuyú points to be poorly crafted homologies of Teotihuacan-style points made from local obsidian. The same can be said for the illustrated, unsourced gray points from Tikal (Moholy-Nagy 2003:figs. 64–67). Moholy-Nagy and her colleagues (1984:111) note that these gray points also could represent local emulation of central Mexican styles, which might explain the variation in patterns of pressure flake removals. This issue could easily be clarified with chemical characterization using a nondestructive, portable-XRF machine.

Nonetheless, in eastern Mesoamerica projectile points made from Zaragoza-Oyameles obsidian were being included in ritually and politically significant clustering of exotic materials, often associated with Teotihuacan. However, the idea that it was the Teotihuacan elite that were incorporating projectile points made from Zaragoza-Oyameles obsidian seems unlikely for several reasons. First, all evidence indicates that Cantona was independent from Teotihuacan and, as a result, would not have fallen under its political and ideological influence and/or control. Considering the significant symbolic importance of the gifted items associated with Teotihuacan (Spence 1996), and Teotihuacan’s control over the nearby black-gray, Otumba obsidian, there is no reason for black-gray Zaragoza-Oyameles obsidian points to have been included in the gifts that represented Teotihuacan in long-distance gift exchange. In fact, if Teotihuacan was the only highland Mexican polity involved in long-distance gifting, then we would expect Teotihuacanos to actively discourage other independent highland Mexican polities from doing the same. As a result, we would not expect to find points made from Zaragoza-Oyameles obsidian in Early Classic elite and ritual contexts in eastern Mesoamerica at all.

Second, any artisan responsible for creating the official projectile points representing Teotihuacan in long-distance gift exchange would have been able to distinguish the local Teotihuacan-sanctioned black-gray Otumba obsidian from nonlocal, foreign-controlled Zaragoza-Oyameles black-gray obsidian. Therefore, the presence of Zaragoza-Oyameles debitage in the Moon Pyramid workshop was not an accident or happenstance resulting from the artisans acquiring whatever obsidian was available. It was there on purpose. While the nature of the Cantona elite’s control over the extraction and early-stage production of tools at the Zaragoza-Oyameles source area is still being investigated, it is safe to say that Cantona’s elite would have been aware of material from the source making its way to Teotihuacan and would have controlled such exchange. Therefore, its presence in the Moon Pyramid workshop likely represents some form of limited gifting between Cantona and Teotihuacan and was meant for specific, possibly ritual or elite use at Teotihuacan, not as part of Teotihuacan’s program of long-distance gift exchange. That only a single Stemmed-A point in green obsidian has thus far been recovered at Cantona suggests that whatever the nature of interaction between Cantona and Teotihuacan was, at least in regards to obsidian, it was either fleeting, strained, or considerably limited.

Finally, the evidence of ceramics from across Mesoamerica at Cantona underlines the fact that the Cantona elite were well established in long-distance exchange networks, especially via the Gulf lowlands (Braswell 2003d), and thus powerful enough to independently cultivate their own relationships with foreign polities outside of the central Mexican highlands. How Maya, and other eastern Mesoamerican elites chose to arrange the gifts they received—such as in burials or caches, and so forth—is another issue. But it is very possible that they combined gifts from several exotic, central Mexican polities into one tableau of greatest significance to themselves. This concept is emphasized by Demarest and Foias (1993:170–171) in discussions of the nature of central Mexican and Maya interaction during the Early Classic, from an “internalist” perspective (Stuart 2000).

After the demise of Teotihuacan around AD 550 (Cowgill 2015:233), the central Mexican Stemmed-A and Stemmed-B point styles ceased to be dominant in the Basin of Mexico and elsewhere. Rather, sites from the Basin demonstrate the use of the Ramec point style (the San Marcos point style in the Texas typology and in García Cook’s [1967] typology), throughout the Epi-Classic period. At the Metepec biface workshop in Teotihuacan (Nelson 2009), the Ramec point was the principal point style produced. This point style also has been recovered at Xochicalco in Epi-Classic contexts (Andrews 2002:fig. 7). In the illustrated examples of Ramec points from Teotihuacan, the lower-left to upper-right flaking pattern is present (Nelson 2009:fig. 5). However, at Xochicalco, the lower-left to upper-right pattern as well as the upper-left to lower-right flaking pattern were evident on Ramec points (Andrews 2002:fig. 7). While this point style appears to have become a chronological marker in certain areas within central Mexico, not a single point in this style was recovered during the survey of the Zaragoza-Oyameles obsidian source area (Knight 2012, 2013, 2015), or at Cantona during its zenith (Rojas Chávez 2001).

In the Epiclassic period, Zaragoza-Oyameles obsidian becomes one of the earliest central Mexican obsidians to appear in the northern Maya lowlands in large quantities (Braswell 2003d:140). The quick introduction of relatively large quantities of Zaragoza-Oyameles obsidian into the northern lowlands occurred during Cantona’s post-Teotihuacan florescence, when it achieved its maximum size and population and when it could capitalize on the interaction and exchange relations it had cultivated in the previous Early Classic period. I interpret both of these post-Teotihuacan patterns as reflecting a continuation of Cantona’s independence from polities in the Basin of Mexico and its focus on long-distance interactions with the Gulf lowlands, isthmus, Pacific coast, and all points further east.

Closing Statements

While the current data cannot answer the question of whether the projectile points made of Zaragoza-Oyameles obsidian received by foreign elites came from Cantona or Teotihuacan, the concept that long-distance gift exchange during the Classic period between the central Mexican highlands and eastern Mesoamerica may reflect something other than just Teotihuacan-based gift exchange is an important point (see Demarest and Foias 1993:171; Marcus 2003:355). Much has been made of the green Pachuca points found in ritual contexts outside of central Mexico. Perhaps because of Pachuca’s distinctive green color and the close association with gray Otumba obsidian and Teotihuacan, archaeologists outside central Mexico have not been as rigorous in identifying the variety of gray obsidians that they have encountered in similar ritual contexts as the Pachuca materials. Central Mexican–style projectile points were being produced at the Zaragoza-Oyameles source area and consumed at Cantona. They were produced through a local tradition that resulted in a pressure flaking pattern different from that found in the Basin of Mexico, and one that is easily recognized and that may turn out to be a diagnostic attribute. It appears that they also were part of the long-distance gift exchange that the Cantona elite were actively participating in with the rest of Mesoamerica. As the web of Early Classic interaction and exchange in Mesoamerica becomes better understood (Braswell 2003c:14–19), we may find that the Maya, and others, were utilizing a much broader suite of exotic materials than previously considered in their political, economic, and ideological constructions. In the case of interactions with Cantona, this may have meant projectile points and prismatic blades made from Zaragoza-Oyameles obsidian. But it is not just the presence of Zaragoza-Oyameles obsidian in these far-off locales that needs to be addressed. A broader questions is what mechanism(s) resulted in obsidian from west Mexico, such as Ucareo and Zacualtipan (Braswell 2013:164; Moholy-Nagy et al. 1984:table 2; Moholy-Nagy 2013:table 6), for example, entering eastern Mesoamerica during the Early Classic? Should we reasonably expect that Teotihuacan was responsible for the movement of every type of highland Mexican obsidian into eastern Mesoamerica during its florescence? We know that numerous obsidian exchange networks were in place before Teotihuacan existed (Boksenbaum et al. 1987; Cobean et al. 1971; Pires-Ferreira 1975, 1976; Pool et al. 2014), as well as after (Braswell 2003d). Therefore, as several authors have observed (Demarest and Foias 1993:162–164; Marcus 2003:355), numerous exchange networks independent of Teotihuacan were likely at play during the Early Classic period provisioning eastern Mesoamerica with obsidian.

My aim in this chapter has been to present data from the Zaragoza-Oyameles obsidian source area, which suggests the involvement of Cantona in the “web of interaction” with contemporaneous polities in the Maya lowlands, independently of Teotihuacan. I am confident that with more rigorous analysis and recording of highly portable artifacts, such as obsidian projectile points in all contexts, that Early Classic interaction between eastern and western Mesoamerica will be shown to have involved numerous independent polities beyond the umbra of Teotihuacan.

References