On the cover: 12 bells unearthed at Lamanai, including complete, flattened and miscast specimens. From Simmons and Shugar 2013: 141
The significance of Copper bells in the Maya Lowlands
-
from their appearance in the Late Terminal Classic period
to the current day
-Arthur Heimann
Master Thesis
S2468077
Prof. Dr. P.A.I.H. Degryse
Archaeology of the Americas
Leiden University, Faculty of Archaeology
(1084TCTY-F-1920ARCH)
T
ABLE OF CONTENTS1. INTRODUCTION ... 5
1.1. Subject of The Thesis ... 6
1.2. Research Question... 7 2.MAYA SOCIETY ... 10 2.1. Maya Geography... 10 2.2. Maya Chronology ... 13 2.2.1. Preclassic ... 13 2.2.2. Classic ... 14
2.2.3. Terminal Classic: From the Classic to the Postclassic. ... 15
2.2.4. Postclassic. ... 25
2.2.5. Postcontact ... 28
3.HISTORIOGRAPHY OF MESOAMERICAN ARCHAEOMETALLURGY ... 30
3.1. Central America Metallurgy ... 30
3.2. Maya Metallurgy ... 31
3.2.1. The lost-wax technique ... 37
3.2.2. Annealing and Open-mould Casting ... 39
3.2.3. Summary of copper artefacts ... 40
3.2.4. Bell’s form and style ... 43
4.BELLS AS INSTRUMENTS ... 46
4.1. Maya music and Performances: ... 46
4.2. The functions of Maya Performances: ... 47
4.2.1. In the Classic Period ... 48
4.2.2. Postclassic Continuity ... 52
4.2.3. Historic Maya dances and performances ... 53
4.3. Maya Bells distribution and attested usages: ... 55
4.3.1. Mesoamerica Homogeneity ... 55
4.3.2. Postconquest and Contemporary usage of bells in dances and plays: ... 57
4.4. Bells as voices of the ancestors ... 60
4.5. Conclusion ... 61
5.ORNAMENTATION OR COINAGE? ... 63
5.1. Acquisition of Raw Materials... 63
5.2. The Production Process: Labour Needs and Arrangements ... 67
5.2.1. Copper Workshops and context... 67
5.2.2. Coppersmith status... 76
5.2.3. Gender attribution ... 77
5.3. Placement on the body ... 78
5.4. Exchangeable Commodities ... 79
5.4.1. Cacao ... 80
5.4.2. Bells ... 81
6.MISCAST AND FLATTENED BELLS: ... 84
7.CONCLUSION AND FUTURE AVENUES... 87
ABSTRACT ... 92
BIBLIOGRAPHY ... 93
5
1. I
NTRODUCTIONMaya copper artefacts appeared in AD 900 and rapidly spread throughout the region. Notes from conquistadors, while narrating their journey, mentioned encounters with Maya carrying crucibles and copper artefacts on their boats (Columbus 1959 in Paris 2008: 47). Multiple excavations in the lowlands have yielded copper-alloyed artefacts. Within those metallic assemblages, bells are the most abundant objects1. Archaeological excavations, along with
conquistador’s reports and ethnolinguistic analyses revealed several functions attributable to those bells. Firstly, a translation from the Cordemex Dictionary of the terms “cheh ok” and “cheh ok mascab” as “bells that are used by dancers” (Vásquez 1980: 87 in Simmons and Shugar 2013: 152) has led researchers to consider that bells may have been most prominently used as instruments in Maya performances (Simmons and Shugar 2013; Hosler 1995 and Graham 2004). Secondly, seventeenth-century conquistador, Diego Lopez de Cogolluda referred to those bells as money, explaining that “the money that they [Maya] used was small bells and jingle bells of copper, that had value, according to their size” (1688 in Simmons and Shugar 2013: 151; Paris 2008: 47), thereby attributing an economic function to those items. Lastly, scholars believe that bells may also have been used and worn as ornaments. They based their argument on the link that copper had with the Sun as its excrement2 (Knowles 1984: 462 in Scott E. Simmons and Aaron N. Shugar 2013: 151; Speal 2014: 97). Despite these acknowledged multiple functions, a full-fledged study on those bells has yet to be probed.
1
For instance, 482 of the 559 metallic objects unearthed at Mayapán are bells (Paris 2018: 6). They also represent 42 per cent of Lamanai’s copper assemblage (Simmons et al., 2009: Table 1).
2 This coincides with Hosler’s research (1994, 1995) on Mexican metallurgy, whose work
demonstrated that some metallic colours and sounds were perceived as divine. This link will be touched upon in the chapter dedicated to Maya’s Ornamentation as it most certainly played a decisive role in the adoption of bells as status markers.
1.1. Subject of The Thesis
This thesis initially wanted to test whether copper was integrated as a commodity complex in the Postclassic Mesoamerican World System. Research examining Mesoamerican commodities and their economic significance was first initiated by Blanton, Fargher, and Espinosa, who applied "a goods-based approach to world-systems." They looked at the economic dimensions, dynamics, and contexts of five resources: obsidian, salt, cacao, cotton cloth, and pottery (Berdan 2016: 132). Such systems study “the complexity of the labor, the requirement to obtain the resources, the time allocation bottlenecks, the relationships between the good and dynamic distribution systems, the good’s impact on secondary industries or markets, and broader impacts of increased production of the commodity” (Berdan 2016: 132). Since my interest lies in the Maya culture, I wanted to take a similar approach and establish the circulation and manufacture of copper in the region. However, the scarcity of archaeometallurgical remains complicates this endeavour. Experimental archaeology is, indeed, fundamental to determine the labour, time and resources invested. Unfortunately, our current understanding is not yet sufficient to realise a full attempt at reconstructing their chaînes opératoires. Consequently, this study will rather focus on the significance that bells held, in a hope to shed light on the impact that copper had on Maya. The first two chapters will be, respectively, dedicated to reviewing Maya History and to summarising our current knowledge on Mesoamerican metallurgy. A focus will be placed on Maya bells production in order to acknowledge the complexity and requirements that need to be met to cast an object. The other chapters will analyse the various functions of bells.
This method will allow us to discuss the likeliness of adopting bells as instruments, money and ornaments in a pragmatic framework. Lastly, the final page will be dedicated to the remaining questions and recommendations for further research.
7 1.2. Research Question
The research question is the fundamental ground used to conduct research. It encompasses various sub-questions that will be discussed in their respective chapters.
The research question is:
Can we assess copper bell’s significance in the Maya lowlands from their appearance in the Terminal Classic to their current uses?
The sub-questions are:
1. Do we have a general understanding of their casting process?
Maya metallurgy is still investigated, and numerous questions remain. The lack of complete melting and dewaxing furnaces in the region complicates any attempt at replicating their technique3. However, at Lamanai, Mayapán and El Coyote, evidence of on-site metallurgy
indicate that Maya produced their artefacts themselves, and several casting techniques have been coined by the teams excavating these cities4.
3 Kiln fragments have been found at El Coyote. Efforts at reconstructing the furnace have been probed,
and a hypothesis on the furnace’s aspect has been formulated. Nevertheless, no copper artefact has been found so far, forbidding scholars to determine whether the tradition that emerged in Honduras is the same as the lowland tradition from Belize and the Yucatán peninsula. As pointed by Paris in her article reflecting on the World System Economy of Mayapán, regions close to ore sources will most likely develop metallurgy first (2008: 45). Based on this remark, Honduras might be the country where Maya metallurgy emerged or the birthplace of an entirely different tradition. This area marks the cultural frontier of Mesoamerica, a porous border with visible South American influences (Urban et al. 2013: 79). This further complicates the identification of this furnace as a product of Maya tradition. The currently available information on this furnace will still be reviewed on the section dedicated to metallurgy, but the reader must be cautious before considering the metallurgical centre of El Coyote Maya.
4 More copper artefacts have been excavated at Lamanai than in any other sites in the southern
lowlands and is only rivalled by Mayapán and Chichén Itzá, located in the northern lowlands. Even though the Sacred Cenote of Chichen Itza has yielded important quantities of copper and several bells,
Bells were produced using the lost-wax technique5. Consequently, this process will be mainly studied. Nevertheless, a summary of the various copper objects will still be presented, along with the casting method for two-dimensional objects.
2. Were bells used as instruments and if so, did they hold a specific function?
Next, the function of bells as instruments will be harder to establish since the author’s knowledge of archaeomusicology is somewhat limited. Moreover, analyses on Maya music usually only define Classic Maya musicality and their instruments6. This poses an issue since bells appeared and spread in the Late Terminal Classic/Early Postclassic (Simmons 2013, 2017). Consequently, this thesis will not be an attempt at analysing bell’s rhythm and timbre. It will instead examine the significance of dancing and the integration of bells in performances. Establishing, this way, whether bells were exclusively used as instruments, or if they held any symbolic importance.
3. Were bells worn as ornaments and if so, why were they introduced?
4. Is the viewing of bells as money an accurate transcription of their use? Or is this the result of European glasses put on by conquistadores observing a culture unknown to them?
These two questions will be discussed in the same section since they overlap each other. Indeed, to answer any of the two, a focus on the context of the finding is mandatory. Bells have been found in various sites in the lowlands. However, little information is available on them since they are usually simply mentioned. Full-fledged metallurgical projects have only
it will not be discussed in this thesis since the metallurgy over there appear to be from a different tradition.
5 This technique has been attested throughout the world; therefore, comparing the techniques
developed in certain regions with the archaeological remains found in the Maya lowlands should reveal the complexity and cost of casting bells as well as the general process.
6
See for instance, the book edited by Zalaquett, Nájera and Sotelo in 2014: Entramados sonoros de
9 been undertaken at Mayapán, Lamanai7 and El Coyote. Consequently, the other sites will be simply listed, and the provenance of their bells noted. Afterwards, the significance of cacao beans and proven usage as a form of currency will be compared with the knowledge gained on bells. This comparative approach should determine the likeliness of bell’s adoption as money. Lastly, the lost-wax technique’s complexity will be decisive in establishing whether bells could have been produced in sufficient quantity to postulate their economic function.
This thesis will try to incorporate studies from a wide variety of discipline ranging from ethnology, fieldwork reports, archaeomusicology and metallurgy. The information used for this thesis originates from publications and literature on the subject pinpointed. I did not partake in those excavations, nor did I study those objects first-hand.
7 Tipu, in Northern Belize, was studied alongside Lamanai and the authors chose to discuss its
metallurgical assemblage as a part of Lamanai own assemblage. Consequently, this town cannot be addressed individually but will be referred to at several occasions.
2.
M
AYA SOCIETYThis section will tackle the Postclassic Maya social complexity in the lowlands. The geographical repartition of the Maya, along with their history, will be looked at. Their history is split into four major periods; Preclassic/ Formative (ca. 2000 B.C.-A.D. 250), Classic (A.D. 250-900), Postclassic (A.D. 900-1450) and Postcontact (Pierce 2016, 61). Those phases have themselves been further divided into sub-periods and vary geographically; the transition that led the Classic to the Postclassic is called “Terminal Classic” (A.D. 750-1000)8 and will be of
interest in the discussion. Indeed, bells started to be cast at this period and became increasingly present in assemblages during the Postclassic. The differences that might have emerged between the Classic and Postclassic are, therefore, critical elements to establishing bell’s adoption as a valued commodity.
Moreover, extra-regional interactions, mandatory to obtain copper9 and well-developed at the
time of the conquest, have likely begun during these two centuries. Afterwards, the social and economic connectedness of the Postclassic era will be discussed under the World System Model, with regards to its impact on the Maya Region. Lastly, Alexander, who discusses the changes brought by the Spanish Conquest, advocates for a subdivision in two periods; starting from 1450 to 1750 (2012: 4) and from 1750 to our days (2012: 7).
2.1. Maya Geography
The Maya area encompasses a territory of 324,000 km2 composed of numerous biomes. It
covers a region from the lower Lempa River in El Salvador to the Isthmus of Tehuantepec in Mexico, englobing the Yucatan Peninsula, the southern Mexican states of Chiapas and Tabasco, the countries of Guatemala, Belize, western Honduras, and El Salvador. This territory was divided into subsections by archaeologists; the Pacific Coastal Plains (a region fertilised by a volcano), the lowlands (the territory situated between sea level and 1000m) and the highlands (the area above one thousand metres with mountains as high as 4,220 m).
8
Maya history is a continuum. There are no specific dates to which the end of a period can be attributed. Therefore, having two periods that overlap each other is not abnormal.
9
11 The northern and southern lowlands both present a tropical climate. Notwithstanding, despite similar weather, their vegetation differs. Two elements fostered these differences; firstly, the southern part has watercourses flowing over, while in the northern lowlands, water is almost exclusively underground, being occasionally accessible through large holes attaining the surface (Reyes-Foster 2014: 4711). Secondly, while the southern region is characterised by a dense vegetation nourished by heavy rain (around 2000-3000 mm each year), the northern lowlands present a flora of low bush and scrub planted in limestone (Reyes-Foster 2014: 4713).
13 2.2. Maya Chronology
2.2.1. Preclassic
This era can itself be separated into four subperiods; the early, middle, late and terminal Preclassic. The Early Preclassic (2000BC-1000BC) saw the emergence of agricultural villages, following the shift from a hunter-gatherer to a sedentary lifestyle. In the same time, Olmecs developed in nearby Tabasco and traded with highlands and Petén Maya.
Sedentary groups employing ceramics started to settle in the lowlands during the middle period (1000-400 BC). Excavation at Seibal reached the first layer of occupation (AD 1000). This survey showed that Seibal, among other sites in the area including Mirador, San Isidro, Ocozocoautla, Chiapa de Corzo10 shared standardized spatial plans; their ceremonial structures all consist of E-Group complexes, and large platforms placed along the north-south axis of the towns (Inotomata et al. 2013: 470). These structures were used as astronomical observatories and kept their significance throughout Maya history (Freidel et al. 2017). Social differentiation can be first attested; various prestige goods such as obsidian mirrors and jade mosaics started to be traded and imported, along with the first carving of regal portraits on stone stelae. Extensive research has yet to be realised, but current discussions debate on the influence that the Olmec culture had on these nascent communities; some view the Olmec has a mother culture who profoundly influenced the Maya, while others argue for limited contact (Doyle 2017).
During the Late Preclassic (400BC-100AD), settlements expanded, stonework and stucco friezes became increasingly present, and two rival cities appeared; Kaminaljuyu in the highlands and El Mirador in the lowlands (Doyle 2017; Inomata et al. 2014).
The Terminal Preclassic (100-250AD) yields the first evidence of later meaningful rituals such as bloodlettings to inaugurate new complexes and sacrifices. The end of this period is marked by the abandonment of major urban settlements such as Kaminaljuyu and El Mirador (Doyle 2017).
10
Excepting Seibal, the sites mentioned in this page are situated in the Tabasco department, and are not visible on the map due to their western location.
2.2.2. Classic
Recent studies show that the Classic lowlands gave birth to densely populated towns depended on labour forces from urban centres and the countryside. The ruling elites, the kings and their relative, were considered divine and acted as intermediaries between gods and humans (Pierce 2016: 26). Among their responsibilities, nobles were tasked with allowing ‘rebirth’ and communication with the dead. Moreover, they also invoked and personified deities, and represented the central axis of the cosmos (Freidel 2012:192 in Pierce 2016: 26). Based on these roles, Freidel et al. (1993) discern a shamanistic nature in the cult of divine kingship, where forebears were key actors. Consequently, validating their lineage was central to anyone trying to legitimate their rule. Furthermore, alongside their family tree, a plethora of elements and objects were used to justify their position as elites; such resources held cosmological and religious significance that were shared throughout Mesoamerica by commoners and elites alike. Demarest (1992: 147) asserted that “religion itself was a principal source, perhaps the source, of the power of Maya rulers”. Rulers, at sites such as Tikal and Seibal, were represented as both the centralised authority and as the supreme religious leader, and at times merged ‘their own priestly offices with the functions of gods themselves” (Reilly 2012: 772 in Pierce 2016: 26). Ancestor worship became “central to all aspects of Maya ideology in all periods and at all levels of society”, and offerings were either associated with ancestors or god veneration (Demarest 2013: 375). Rulers and nobles gained significant power from their function as a direct line to the ancestors (Demarest 2013: 375). It, thus, became compulsory to celebrate one own lineage, and integrate ancestors into current life (Rice 2013:704). To do so and perpetuate the past, Maya frequently employed textual inscriptions —on stelae, lintels, polychrome pottery, among others— (Pierce 2016: 27).
These texts were often found in the vicinity of monumental pyramidal temples, which were structures used as funerary palaces. Wendy Ashmore (1989: 272 in Pierce 2016: 27) has explained that Classic Maya rulers were “using principles of site planning based in cosmology as a means to profess and reinforce their membership in the political elite, affiliating symbolically with leaders of the Classic Maya establishment by replicating their surroundings”. Monumental complexes represented a ruler’s accession and conquests, as visible by the text and iconography found on many of these structures (Andres 2005: 34; Demarest 1992; Freidel and Schele 1988). As Pierce states (2016: 27) ‘this venue of temples,
15 plazas, and palaces served as a setting for elite political ritual, bolstering political power and confirming elite status’.
This period is seen as a golden age that was disrupted by the collapse that occurred in the ninth and tenth centuries.
2.2.3. Terminal Classic: From the Classic to the Postclassic.
A General Overview
By the ninth century, profound changes drastically modified Maya landscapes and routine. Maya no longer built colossal constructions in most cities, and almost entirely ceased to inscribe texts on stone monuments. Moreover, the manufacture and distribution of many status items and ritual goods disappeared (Sharer and Traxler 2006: 499). Although these changes were felt throughout Mesoamerica, some regions were more affected; most notably the central and southern lowlands. These two areas saw the high point of power struggles. Wars had already broken out by the beginning of the eighth century and, within a century, capitals were deserted. Signs of commoners having occupied noble habitations before departing are visible in the archaeological record, and by the ninth century, the southern lowlands were devoid of its inhabitants11. Although scholars have searched for a prime factor
that would have inexorability led to the decline of civilisation and a fall of authority, this immense change is still investigated, and its causes hotly debated. Evidence of decreased in agricultural yields caused by deforestation and frequent droughts that severely eroded the soil, have been found and could have caused the collapse (Sharer and Traxler 2006: 499-513). Political processes such as migration, war, changes in trade are also considered factors in the decline (Demarest et al. 2004: 546–548). Aimers compiled in a table all the possible causes of the collapse in her article (2007: 33) and showed the intricacies of the debate12. The model applied to explain this crisis can be classified under the umbrella of push-pull models. In push models, scholars look for the negative factors leading to the decision of migrating, while pull models look for the positives that attract someone and lead them to join a new territory. In this framework, archaeologists tend to overemphasise the role played by ecological disasters and
11
Sharer and Traxler have estimated than 90 per cent of the inhabitant of central and southern lowlands either died or fled the region (2006: 525).
12
economic crisis, and consequently, tend to simplify the socio-ideological and political reasons. Besides, even though the decision to migrate is taken by the authorities, the actual migration is dependent on a household level.
Furthermore, as touched upon above, the Maya region was the theatre of a plethora of cultures, and not all the areas suffered the same fate. For instance, in Belize, the northern lowlands, and the Mopan Valley, social and cultural life changed gradually, with some cities even developing. Thus, to have a better understanding of this period, presenting each region that surrounds the lowlands is necessary.
The North-westernedge
This area encompasses the Gulf Coast of Tabasco, Campeche and the territory from the Usumacinta River to the Pasion region13. Mexican traits have been identified on ceramics and other artefacts (e.g., Andrews IV 1943; Berlin 1956; Drucker 1943a, b; Hellmuth 1967; Proskouriakoff 1951 in Aimers 2007: 334). This attested trade led scholars to assert that Teotihuacan’s collapse may have stirred up conflicts and competition for resources between groups who had exchanged with Teotihuacán in the past (Sabloff and Rathje 1975 in Aimers 2007: 334). These conflicts led to the decline of Piedras Negras and Yaxchilán in the ninth century. Fine-paste pottery, originating from the Veracruz–Tabasco territory (Allen 2011: 131), had been produced on-site since the Preclassic and eventually replaced the Late Classic polychromes of Campeche and Petén across the lowlands (Forsyth 2005). This represents ‘a critical shift in spheres of interaction away from Central Petén and toward the western lowlands and periphery’ (Foias and Bishop 2005: 37).
The Petén region
Sharer and Traxler define the Petén Region has a gradual transition between the northern and southern lowlands (Reyes-Foster 2014: 4712). By A.D. 830–850, the decline in population had already begun in urban sectors such as Tikal (Valdés and Fahsen 2004). Moreover, newly built or recently resettled towns did not develop enough and collapsed rapidly.
13 Palenque, Piedras Negras, Yaxchilán, and Bonampak are some of the best-known sites there. These
17 For instance, ‘El Mirador’ was first founded in the 6th century BCE and inhabited until the first century A.D. This town was, then, resettled in the ninth century14 but his occupation did not last as it was either deserted at the end of the same century or by A.D. 1000.
Although severely touched, the entire region did not collapse at once. Laporte (2004), who studied 26 polities in the Mopán Valley, has indeed demonstrated that each site reacted differently to the collapse, with some flourishing. Besides, Puuc Maya immigrated in the region and brought their culture (Andrews and Robles 1985: 66). The valley developed, and northern influences were then gradually absorbed and are visible in cities that survived the collapse. The Mopán valley eventually started to decline in the twelfth century, likely due to Itza’s centralisation and pressure (Aimers 2007: 337). Lastly, as mentioned above, an increasing proportion of Mexican fine wares is attested in the Petén. Adams (1973:140 in Allen 2011: 132), who excavated at Altar de Sacrificios, only found a handful of fine paste wares in layers dated to the latter half of the ninth century but saw an increased abundance of this type during the Jimba Period (A.D. 909-948).
Belize
Both permanence, such as in Lamanai and Kakabish (Haines 2014) and collapses, such as in Altun Ha and La Milpa (Pendergast 1982, 1992; Hammond and Tourtellot: 2004) are attested in northern Belize. The prosperity of some sites might be due to their strategic location (Aimers 2007: 339). They had new trade openings with the Yucatán peninsula (Masson and Mock 2004), while others (e.g., La Milpa) rapidly declined in the Terminal Classic (Hammond and Tourtellot 2004). Masson’s studies of Laguna de On and Caye Coco indicates that the greater regional interaction led to easier access to luxury resources and items to both commoners and elites alike (Masson 2000).
Since Lamanai, is among the three towns where on-site casting is attested in the lowlands, decrypting the changes that occurred in the Late Terminal might hold valuable information.
14The presence of fine paste ceramic, introduced in the region in A.D 830 and found throughout at El
LAMANAI
The site of Lamanai, situated on the bank of the New River Lagoon, never collapsed during the transition and even flourished (Pierce 2016: 2). This polity is a testimony of the complexity of the Maya collapse. Indeed, although located less than 50 kilometres away of Colham, Lamanai did not decline and was much less impacted by the process occurring throughout the region.
In 1985, Pendergast explained that the transition from the Classic to the Postclassic is hardly noticeable at Lamanai. Going further, he states that the visual modifications are a result of innovation and traditions (1985: 91-92). Notwithstanding, monumental constructions still appear to have slackened in the Late Classic period (Chase and Rice 1985:86), and the economic centre has shifted to the southern part15 (Chase and Rice 1985: 87). Instead of
constructing new edifices, several pre-existing temple-pyramids were modified16. Inscriptions disappeared, and carved monuments were sometimes destroyed or reused in Postclassic
15
Where the material assemblage shows ties with Mayapán (Chase and Rice 1985: 88)
16Maintaining and renovating Classic temple-pyramids led scholars to see it as sign of religious
perpetuity.
19 construction. The lagoon was urbanised, suggesting the increasing importance of this site as a trade centre. The transformation of the Ottawa Palace Group required the most effort; it likely began in the Late Classic and continued into the Terminal Classic Period. This regal edifice was renovated and modified. Its changes include the destruction of a stucco frieze, representing rulership, and the backfilling of an associated patio (Pierce 2016: 248). To accomplish this, roughly 21,000 metric tons of boulders were required (Pendergast 1985: 232). At the top of the palace, the initial plaza had been enlarged by levelling the preceding structures, and edifices made of perishable materials were erected on the newly expanded plaza. The use of such materials is remarkable and must have been in stark contrast with the other buildings located in the city centre, which typically utilised stone (Pierce 2016: 314). This extensive modification reveals that the socio-political situation in Late/Terminal Classic Lamanai was still stable enough to provide resources and labour for major reconstructions (Pierce 2016: 314).
Nonetheless, modifying the Ottawa Palace is certainly a proof of ideological change, with some areas previously reserved for ruler ritual removed or blocked (Pierce 2016: 314). As Pierce expresses it; “The public aspects of divine performance requirements of the ruler have diminished or disappeared” (2016: 315). Lastly, the ceramic assemblage remained somewhat identical, but the number of external resources augmented. According to Aimers, this suggests stronger ties with the Mesoamerican world system (2007: 349). Pierce agrees and states that this transitional period views a growth in interregional interaction that led to “increased trade, heightened warfare, migrations, and the spread of what is seen as a new religion in the cult of Quetzalcoatl” (2016: 7). Those interactions17 might have provoked a fundamental ideological
change at Lamanai, triggering their inhabitants to reject the Divine Kingship.
17
Determining exactly how cultural exchange influenced Lamanai is rather challenging. The architectural style shows influence from Petén; however, a handful of elements support a northern influence such as the vertical slab risers at the Ottawa Complex, a collonated building east of Plaza N10[2] and a circular structure (Pierce 2016: 313). Lastly, Late Classic architectural similarities to Uxmal where noticed by Graham (2004: 236), suggesting that the site’s long-distance contacts began before the collapse.
The Yucatán peninsula
Many sites in the northern lowlands were flourishing while the south was declining (Ball 1994; Carmean et al. 2004; Freidel 1981; Smyth et al. 1998: 254; Suhler et al. 2004 in Aimers 2007: 338). Puuc centres such as Uxmal and Sayil developed and adopted a joint government, a multepal, similar to Mayapán and, possibly, Lamanai’s systems. (Schele and Mathews 1998: 240; Wren and Schmidt 1991 in Aimers 2007, Paris 2013, Pierce 2016). Nevertheless, the fortification of various towns in the ninth and tenth century points to an increase in competition and warfare (Dahlin 2000; Webster 1978 in Aimers 2007). Puuc settlements eventually started to decline in the following century (Andrews and Robles 1985: 66) and migrated to the Petén zone.
The main occupation of Chichén Itzá spans from the Terminal Classic to its abandonment in either A.D. 1100 (Cobos 2004) or A.D. 1250 (Suhler et al. 2004). This second hypothesis presupposes longer domination on the peninsula, and therefore, a greater influence on the region18. This city was multicultural, with depictions of Maya and Mexican elites19. Scholars argue that this city partook in a deepened Mesoamerican interaction sphere (Aimers 2007; Paris 2008) and that the intensified connectedness between the Yucatan peninsula and the rest of Mesoamerica is partly tributary to Chichén Itzá’s rose to prominence.
North-western Honduras
Urban and her team have excavated the south of El Coyote. The discovery of a fragmented kiln along with crushed slags led the excavators to consider this area as the metallurgic centre of the town. This site was founded in the middle Preclassic (1000-400) and emerged as a regional economic and politic centre during the Late Classic. El Coyote remained relevant throughout the Terminal Classic and maintained its prominence until the Early Postclassic (AD. 1000-1300). El Coyote seems to have changed its social structure during the Terminal Classic as visible by the abandonment of the late Classic monumental centre where several edifices were ‘stripped of their ashlar facing stones’ (Wells 2003b). Those blocks were directly reused to build a new complex. This new structure differs from its predecessor since the Late Classic complex, enclosed and not easily accessible, was replaced by an open plaza
18Itza’s culture stretched as far as Lamanai (Pendergast 1990)
19
21 with several edifices. Such structures include residential buildings, a ball court and other unidentified monuments. McFarlane explains that ‘this new setting favoured public displays and gathering and reflect a movement to a more corporate style of governance with an incipient mercantile economy’ (2005: 193–199). Fine paste wares are also introduced there, including local versions of the fine pastes seen elsewhere.
Conclusion:
The presence of fine paste wares across the lowlands is undoubtedly due to elite interaction (A. Chase and D. Chase 2004, 2005). In this framework, the reduction in polychrome vessels in favour of Mexican wares is proof of a Mexicanization of the elites. According to Aimers, this ‘reinforces the idea that the collapse was primarily an elite phenomenon and that the Late Classic and the Terminal Classic were together an era of elite interdependency and peer-polity interaction around the lowlands’ (2007: 335). The collapse, itself, was caused by a plethora of phenomena occurring at diverse locations and at different rates; this event was not homogeneous, and each region reacted differently.
At Zacpetén, obsidians, a high-status good, became increasingly accessible, and its procurement was facilitated to all social classes (Schwarz 2013: 253). This suggests the emergence of a heterarchical system. Indeed, following the collapse, elites could no longer monopolise or regulate tool acquisition, and commoners gained access to previously restricted sphere of exchange.
Nevertheless, this easier procurement seems to diminish in Late Postclassic/Early Historic Zacpetén, and political control and centralisation re-emerged (Schwarz 2013: 253). As explained by Schwarz, greater post-collapse accessibility of valued items has analogies in other societies. Conlee, who studied the consequences of the Wari collapse (2006: 111), remarked that peasant communities gained access to the highly desired Spondylus shells outside the political hierarchy; they obtained and exploited this resource on their own after the fall of centralised authority.
Copper, therefore, emerge in a transitional phase where the Divine Kingship ideology was diminishing, along with the increased accessibility to previously reserved goods to both the lower and upper classes. Simultaneity, various markers of elite culture such as carved stelae and hieroglyphic polychromes reduced, and forms of joint government appeared in certain polities. Lastly, as stated above, the import of exotic resources increased during this period.
CAUSES OF THE DEMISE, IN AIMERS 2007
ENVIRONMENTAL
Soil erosion/loss of fertility/change to untillable savanna
(Atran 1993; Beach et al. 2006; Cooke 1931; Emery et al. 2000; Harrison 1977; Jacob 1992, 1996; Morley and Brainerd 1956; Paine and Freter 1996; Pohl 1990; D. Rice 1978, 1996; Wingard 1992; Wiseman 1985; Wright and White 1996)
Climate change/drought
(Broecker 1995; Chepstowlusty 1996; Dahlin 1983, 1987, 2000, 2002; Dunning 1992; Folan 1981; Folan et al. 1983; Gill 1994, 2000; Gunn and Adams 1981; Gunn et al. 1995, 2002; Haug et al. 2003; Hodell et al. 1995, 2001, 2005; Hunt and Elliot 2005; Kerr 2001; Leyden 2002; Leyden et al. 1998; Lucero 2002; Messenger 1990; Pierrebourg 1996; Robichaux 2002; Shimkin 1973; Wahl 2005; J. Webster 2000; Yaeger and Hoddell 2002)
Overpopulation/ subsistence stress
(Culbert 1974, 1977, 1988; Dunning and Beach 1994; Haviland 1967; Johnston 1994)
Earthquakes
(Bevan and Sharer 1983; Mackie 1961)
Hurricanes
(Sabloff 1973b)
Ground slope change
(Moseley 1983)
Volcanic activity
23 Deforestation
(Abrams and Rue 1988; Brenner 1983a, 1983b; Brenner et al. 2002; Curtis et al. 1998; Deevey 1978; Deevey et al. 1979; Dunning et al. 1997, 1998; Islebe et al. 1996; D. Rice and P. Rice 1984; Schreiner 2002; Shaw 2003; Wiseman 1985)
DISEASE/PLAGUE
(Acuna-Soto et al. 2005; Pozo Ledezma 1985; Saul 1973; Spinden 1928; Wilkinson 1995)
Insect infestation/plant blight
(Brewbaker 1979; Turner 1974)
SOCIOPOLITICAL
Peasant revolt or class conflict
(Altschuler 1958; A. Chase and D. Chase 2004, 2005; Erasmus 1965; Hamblin and Pitcher 1980; Kaplan 1963; Kidder 1950; Lowe 1982; Mason 1943; Morley and Brainerd 1956; Palka 1995, 1997; Satterthwaite 1936, 1937; Thompson 1931, 1954, 1970)
Warfare
(Demarest 1978, 1993, 1996, 1997, 2004, 2006; Demarest and Valde´s 1995; Demarest et al. 1997; Emery 1997; Foias 1996, 2004; Foias and Bishop 1997; Freidel and Rutledge 2001; Inomata 1995, 1997, 2003, 2006; Palka 2001; Van Tuerenhout 1996; Webster 1977, 1978, 1993, 2000a)
EXTERNAL EXPLANATIONS Change in trade routes
(Rathje 1973)
Competition from Central Mexico
(Freidel 1986a, b; Webb 1973, 1978)
Invasion with resettlement
Invasion without resettlement
(Adams 1973; Ball 1977; D. Chase and A. Chase 1982; Graham 1973; Hester 1985; Sabloff 1973a; Smith 1981)
POLITICAL-IDEOLOGICAL PATHOLOGY Fatalism
(Dornan 2004; Dunham 1990; Houston et al. 2001; O’Mansky and Dunning 2004; Puleston 1979; Pyburn 1996)
25
2.2.4. Postclassic.
Postclassic World System
To explain Mesoamerican interconnectedness, scholars have applied Wallerstein’s World System theory. This model was initially intended to be a Marxist interpretation of the rise of a Capitalist European World Economy (Wallerstein 1974:67). Such systems consider ‘units such as cultures, tribes, nations, states, classes, or ethnic groups’ in a single economy based on macro scaled trade (Paris 2008: 43). An essential element to the world system theory is the
unequal repartition of labour between core, semi-periphery and periphery zones (Chase-Dunn and Hall 1997:35-36), resulting in the emergence of dominant states in central economic
zones and weak states in peripheral areas. As explained by Paris (2008), scholars studying Mesoamerica often refine Wallerstein’s original definitions since this region may be better viewed as several subsystems formed by both regional and interregional exchange circuits of goods and information (Smith and Berdan 2003: 30).
Smith and Berdan (2003: 24) have, thus, changed the traditional divisions of world-system theory from—cores, peripheries, and semi-peripheries— to core zones, affluent production zones, international trade centres, resource-extraction zones, unspecialised peripheral zones, and contact peripheries (Smith and Berdan 2003: 24; Table 3). Core zones are similar to Wallerstein’s definition of cores, but the idea that they must dominate peripheries was rejected20 (Smith and Berdan 2003:24). Affluent production areas are still areas of primary economic production, but with a lower degree of urbanisation and political centralisation. Lastly, significant resources are obtained in extraction zones and are exchanged in international trade centres (Smith and Berdan 2003: 31). Under this framework, the polities, that either maintained their status throughout the Late Classic or emerged during the Postclassic, are interconnected and could import or export resources depending on the wealth of the inhabitants and local availability.
The absence of copper ore sources in the lowlands signifies that it had to be imported and, therefore, as explained by Paris, a world system approach is likely the better suited to analyse Maya metallurgy (2008).
20 According to their model, core zones are better defined by their high populations, their influence, and urbanization, and include key centres such as Chichen Itza, Mayapán, Tula, Tajin, Cholula, and Tenochtitlan.
Fig. 4: Map of Mesoamerica, showing core zones, affluent production zones, and international trade centres involved in the production, consumption,
and exchange of metal artefacts.
27
Mayapán
Mayapán, situated in the Yucatan Peninsula, was the most influential centre in the Northern Lowlands from A.D. 1100 to 1450, with its earliest establishment dated to the eleventh century. This city was a buffer of luxury and utilitarian goods, producing and importing them (Paris 2008: 45). They sold salt, cotton, slaves, cacao, dyes, Fine Orange pottery, and obsidian to different regions of Mesoamerica (Piña Chan 1978: 42), with international trade routes to central Mexico, Oaxaca, the Mexican Gulf Coast and to the Ulua Valley in Honduras (Paris 2008: 45). Craft production activities were spread all over the city (Masson and Peraza Lope 2007: 2) and usually held in small commoner houses. Russell’s excavation advances a population of 12,000–13,000 in the city peak period, comprising the agglomerations outside the city wall (Masson et al. 2006 in Paris 2008: 45).
Tozzer theorised that this city might have been the core of religious teachings for the Yucatán Peninsula (1941: 25–26) as numerous religious building and artefacts were discovered there. Moreover, Masson (2000) found temples and effigy incenses reminiscent of the “Mayapán-style” in Northern Belize, and the Petén Lakes revealed similar features (Pugh 2003); as expressed by Paris, this ‘attest to a regional sphere of religious influence’ (2008). She adds that ‘Mayapán also had administrative functions, represented by colonnaded halls and elite palaces and residences in its centre’. Masson and Peraza Lope found pieces of evidence of a potential marketplace at the north of Mayapán’s centre. This place may have served for both local and long-distance merchants (2007).
Colonialist notes describe a tripartite system composed of nobles, commoners and slaves (Tozzer 1941: 26). Notwithstanding, Chase believes that they may have oversimplified the social complexity of the region and dismissed persons of intermediate status (1992: 121). Paris (2013) adds that within these broad categories, various specialities existed with distinct levels of income. She states that Mayapán’s elites surely worked in civic and religious offices, ‘such as neighbourhood administrators, speakers, and proclaimers; secondary governorships, scribes, or notaries; district deputies, judges, or tribute collectors; military captains; logistical officials; K’atun lords; ball-game counsellors; territorial administrators; and a variety of specialized priests, singers, and sacrificers’ (Carmack 1981, 15–17; Chase 1992, 120; Masson and Peraza Lope 2007, 4; Masson, Peraza Lope, and Hare n.d.; Roys 1943, 33 in Paris 2013: 172).
Meanwhile, commoners likely accomplished several functions as ‘agricultural producers, various types of craft specialists, warriors, local merchants, and mercenaries’ (Masson and Peraza Lope 2004; Masson, Peraza Lope, and Hare n.d.; Tozzer 1941 in Paris 2013: 172).
2.2.5. Postcontact
European arrival in the New World disrupted the sphere of interaction that defined the Postclassic. Although Maya have survived, their symbols and social markers may have changed. Consequently, summarising Maya Postcontact history is necessary. On a side note, modern use of bells is attested in Maya performances; thus, discussing the changes in Maya music and dances is of the utmost importance. Notwithstanding, this section is not suited for this. Therefore, the review of historic Maya dances will be realised in the chapter ‘bells as instruments.
1450 to 1750
This sub-period is characterised by a period of demographic decline due to war, the spread of diseases and the plunder of their territory. As the region became part of Spain's empire, indigenous people experienced a drastic and prolonged decline (Alexander 2012: 5). Various resources were asked of native communities as tributes and taxes, including their bells. Numerous settlements were forcibly moved as part of the congregation's policy, and their inhabitants forcefully baptised (Farriss 1984; MacLeod 1973 in Alexander 2012). Historians have estimated a demographic decline of more than 90 per cent of the population between 1500 and 1550 (Farriss 1984). The few remaining were segregated and suffered from droughts, famines and yellow fever outbreaks. This led to a continuous decline until the 18th century, where they reach their lowest point. Under the colonial policy of congregación, villages were merged to found new towns and local centres. Their decision led to the formation of a two-tiered hierarchy of cabeceras (municipal capitals) and subjetos (secondary towns) (Farriss 1984: 158). Moreover, exiled communities were obliged to settle in street grid cities centred around a church and a central plaza (Alexander 2012: 5).
Nevertheless, archaeological surveys have demonstrated that this process was not systematic and homogenous. Indeed, two excavations on the north coast of Yucatán21 hint at regional
disparities in the demographic decrease (Andrews and Robles Castellanos 2009; Kepecs 1997,
21
29 1999 in Alexander 2012). Besides, eastern Yucatan, Guatemala, Laguna de Tesminos and the Petén were not subjugated22. Lastly, the Crown targeted resources and labour force to produce goods and sell them on an international market. Yet, land control, agricultural production and the management of essential resources such as cocoa beans still belonged to locals (Alexander 2012: 6).
1750 to 1910
In the frontier areas of Petén and the lowlands of Chiapas, Métis traders exchanged with locals who, then, incorporated glass earthenware, metal tools and pharmaceuticals in their community (Palka 2005). Other communities chose to move to secluded areas (Palka 2009). On the Yucatán's north coast, Africans settled in haciendas, alongside natives, and indebted Asian workers, leading to the emergence of new cultural traditions. Alexander concludes his review of Postcontact Maya history by explaining that ‘Many investigators recognize that Mayan speakers in Guatemala and Mexico share comparable experiences and a common heritage, but no pan-Maya ethnic identity exists in the region. Today, Yucatán's inhabitants generally reject indigenous identities, even though many native communities retained control of land (...) instead, current attitudes and practices were reconfigured through a long-term process of ethnogenesis’(Alexander 2012: 9).
22
Adding to that, the area under Spanish authority does not form an uninterrupted territory, but an archipelago.
3.
H
ISTORIOGRAPHY OFM
ESOAMERICANA
RCHAEOMETALLURGY3.1. Central America Metallurgy
This section discusses research realised in metallurgy. As explained by Simmons and Shugar, ancient technology study has been vastly investigated in Mesoamerica (2012: 1). The functions held by lithic and bone objects have been uncovered and helped to understand Mesoamerica social environment. Yet, as they inform us, information regarding copper is on the contrary recent and still shallow (2012: 1). Pendergast (1962) and Hosler (1994) are precursors in Mesoamerican archaeometallurgy, and their contributions are still held in high regards. Indeed, Hosler first noted that two distinct traditions had existed in Mesoamerica; the "West Mexican" and the "South-eastern Mexican." The former is the older, with the first metallic artefacts unearthed in layers dated to AD 600 in the states of Michoacán, Guerrero, Jalisco, Colima, and Nayarit. It was introduced via maritime trade with Ecuadorian populations (Hosler 2009: 188-189; Simmons and Shugar 2013: 3; Paris 2008: 47). The technique and tools employed, as well as the objects cast, are reminiscent of the tradition developed in Costa-Rica and the Isthmus (Bruhns and Hammond 1982: 178) and were uniformly distributed (Simmons and Shugar, 2012: 3).
31 The significance held by copper among Western Mexican population has been heavily studied by Hosler, who suggested that both copper's natural colour and sound led to its rapid spread throughout the region. As it never superseded utilitarian object made of stone, she argued that it essentially played a symbolic role (Hosler 1994: 208). Moreover, she found instances where the quantity of thin largely exceeded the standard ratio to cast bronze. Indeed, some objects were composed of 23% of thin, which is highly unusual as a typical bronze is generally made of 3 to 5% of thin and 95% of copper. With such a high quantity of thin, a copper alloyed becomes much more breakable and is no longer suited for practical usage. She thus explained that thin might have been added to gain a silvery colour and resemble the moon (Hosler 1994: 210).
Pendergast’s classification of metallic artefact as either utilitarian objects, objects of personal adornment, or ceremonial objects (1962: 521) is still used. Simmons, for instance, categorised the artefact found at Lamanai and Tipu according to this model (Cockrell and Simmons 2017: 162).
3.2. Maya Metallurgy
The latter tradition, less known, appeared by the end of the Late Classic Period with ore sources found in Chiapas, southern Guatemala and western Honduras23 (Simmons and Shugar
2012: 4). The earliest sign of Maya metallurgy seems to date to 1000 AD at El Coyote in Western Honduras24 (Urban et al. 2013: 77). An increasing interest on this tradition led to the
launching of The MAP (Maya Archaeometallurgist Project), and various sites that yielded copper artefacts begun to be excavated. This project, started in 1999, aimed at examining ‘the organisational structure of ancient Maya metallurgy and the roles metal objects played in Postclassic and Spanish Colonial Maya society’, as well as, defining the chaînes operatoire and the techniques used (Simmons and Shugar 2013: 106).
To answer these numerous interrogations, Simmons and Shugar extensively excavated Lamanai. There, they found signs of on-site casting and unearthed both utilitarian metallic objects, such as fish hooks and needles, and symbolic or ornamental artefacts such as button
23
This implies that no copper ore sources were available for the lowlands Maya who had to import it.
24
Older copper alloyed artefacts have been found inside the Grand Cenote in Chichen Itza. However, stylistic and chemical analysis has determined a Mexican provenance.
ornaments and bells (2013: 14). The earliest copper artefacts can be found in layers dated to the end of the Classic and beginning of the Postclassic periods. Simmons and Shugar, therefore, argue that the inhabitants started to cast copper themselves approximately around 950 AD - 1200 AD, and continued to do so under Spanish control until 1700 AD (2013: 107). They noticed a lack of utilitarian objects in Early and Middle Postclassic time (950-1350) and only found metallic objects in burials reserved to the elites. The artefacts unearthed include rings, bells and clothing ornament alongside ceramic vessels, shark teeth and pyrite mirrors (Simmons and Shugar 2013: 108-109).
Chemical analyses demonstrated that copper-alloyed artefacts were initially imported from west Mexican and south-eastern Mesoamerican sources and brought to Lamanai to be melted (Simmons and Shugar 2013: 108). By the Late Postclassic (AD 1400), recycling habits emerged, and importation diminished. This recycling habit is noticeable by the composition of the alloyed used. Every object’s composition varies a lot from one another and cannot be traced to a unique ore source. The table below presents the chemical analysis of a pouring reservoir found at Lamanai and the presence of both thin, arsenic and iron with inclusions of various metal confirms their theory.
33 Simultaneity, the manufacture of utilitarian objects started and became accessible to commoners as well25 (Simmons and Shugar 2013: 112).
Tab.3: Chronological distribution of copper artefacts and debris found at Lamanai
25
An analysis of copper availability to the population is attempted on the section dedicated to ornamentation.
Mayapán, in the northern lowlands, has yielded a considerable quantity of copper26 as well and does show signs of on-site casting. Cruz Alvarado, the ceramist of the Mayapán Project, has determined so after having unearthed objects with the “presence of green residues with dark grey paste and sign of exposure to high heat on them” (J.L. Meanwell et al., 2013: 4308). The likelihood of on-site casting is further supported by the discovery of a bell mould and an elliptical tripod vessel with vitrified areas27, respectively found in a commoner dwelling and
in a temple (J.L. Meanwell et al. 2013: 4309). Analyses have been realised on the vessel and have offered valuable information. Indeed, traces of Al, Si, Cl, K, Ca, Ti, Mn, Fe, Ni, Cu, As, Sr, Ag, Sn, Au, Y and Zr have been noted. Al, Si, Cl, K, Ca, Ti, Mn and Fe are part of the composition of the clay itself.
Notwithstanding, the presence of Cu, Ag, As, Sn and Au are leftovers of metal casting and were either due to multiple successive castings or a single melt produced from a variety of recycled objects. As seen above, similar mixed compositions have been found at Lamanai, where recycling habits have been attested. Therefore, the same explanation was offered for Mayapán (J.L. Meanwell et al. 2013: 4309).
Fig. 6: Mayapán’s bell mould and the tripod vessel
The bell mould has also been studied, and XRF analysis shows no sign of copper ever poured in28. Finally, at both Lamanai and Mayapán (Meanwell et al. 2013: 4310) prills, casting sprues
and pouring reservoir have been excavated.
26
599 artefacts have been found in the city and 86 per cent of them, 482, were bells (Paris 2018).
27
Ceramic vitrification begins at a temperature neighbouring a thousand degrees (Rice, 1987: 91), which is the heat required to melt their alloyed copper. Indeed, pure copper’s fusion point is closer to 1100°C. However, as explained above, copper from the late postclassic period onward was generally recycled and rarely not alloyed with either thin or arsenic.
35 Those prills are “drops of liquefied metal spilt out of the mould during the casting process”, and a casting sprue is the “channel through which the molten metal is poured into the mould” (Meanwell et al. 2013: 4310). All of this indicates the use of the lost-was techniques to cast these objects.
El Coyote’s metallurgical centre will now be touched upon. The southern part of the city has yielded evidence of on-site metallurgy. This zone was excavated in almost its entirety but lacked any copper artefact. However, the functions of several major structures have been identified and revealed information on the processing of copper before its melting.
Fig. 7: Map of the copper-working area of El Coyote
Firstly, the structure 407 is formed by four large stones and measures roughly five square meters. One of these boulders presents a depression29 in its top. This kind of hole is often
symptomatic of a block used as stationary anvils. Urban et al. thus offer a similar explanation for those four boulders and suggest that copper ore was either beneficiated or broken up there (2013: 84). The presence of a stone hammer in the vicinity of crushed ores is further proof of 29
this block used as an anvil. Several adobe furnace fragments, a vitrified adobe, and slags were also situated in the soils close to the boulders. The refined minerals were then brought to structure 372, 14 metres south of the crushing zone. This structure consists of a one-meter-high polished stone platform on the northside that covers 1 m2 (Fig. 8). This platform was
certainly used as a support on which smelting furnaces were posed, as visible by the smelting furnace still in situ. This kiln is 0.3 m high and measures around 0.15 × 0.20 m, with walls of poor refractory adobe blocks30. These blocks are vitrified on the inside but unbaked and eroded on the exterior (Urban et al. 2013: 84). The pattern visible in the vitrified clay attests that the mass flew toward the bottom of the furnace (Fig. 15). The south side of the kiln was found broken out, undoubtedly, to let the molten copper and slag fall in a shallow pit below the furnace. Based on the numerous used furnace fragments31 and the clay low refractory quality, Urban et al. posited that those kilns could only be used once before being discarded (2013: 84). Compton tried to recompose the aspect and the functioning of the furnace and determine that ‘their walls were made in L shapes, with four combined to make a hollow quadrilateral container’ (Urban et al. 2013: 96). Once cooled, copper prills still imprisoned in slags were brought to a pebble stone pavement to be pulverized. The course of the nearby Quebrada Seca may have been partly redirected, when necessary, to flood the pavement (Fig. 9) to remove the crushed slag without washing away the heavier copper prills. The pavement, or water table, has a slight slope west to east, likely signalling the flow of water (Urban et al. 2013: 86-87). The last step, the melting, has left no trace.
Fig. 8: West Face of El Coyote structure 372
30
Several furnace fragments have been recorded and the size of kilns appears consistent, with their exterior faces systematically falling between 0.2–0.3 m.
37 Fig. 9: Picture of the Water Table Fig.10: Picture of a wall of the furnace
Several layers of occupation and numerous furnace fragments attest of the long occupation of Coyote’s metallurgical centre. Nevertheless, establishing its first use is an arduous task. Urban et al. found two pieces of evidence pointing to the Terminal Classic / Early Postclassic. Firstly, the structure 99, that overlooks the copper-processing centre had three rocks glued by slag alongside Terminal Classic / Early Postclassic pottery of the Plumbate and Las Vegas types. Next, on the roof of some Terminal Classic / Early Postclassic terraces near Structure 99, small pieces of copper ore have been found (2013: 96).
3.2.1. The lost-wax technique
The exact chaînes operatoire of this technique has yet to be studied for the Maya. Detailing the process is, therefore, not yet possible. Still, by summarising its principle via the work realised by Verly32 and by contrasting it with our current information on
Maya and Aztec metallurgy, the general process should be unravelled.
Fig. 11: A bell found with its down sprue and pouring reservoir in Lamanai
32 I have worked under him for the past three years, and he is currently publishing his thesis.
Verly has replicated the technique developed in Ancient Egypt and notably, their lost-wax technology. I do not state that the process, I am about to explain, is perfectly applicable to the Maya tradition. It will, however, provide a basic understanding of the technique.
Egyptian Tradition
Firstly, the artisan must shape in wax the object that he wants to cast in bronze. Three layers of different thicknesses and compositions (made of donkey dungs, clay, and sand in various proportions) are then attached to it. Once the first two layers applied, a pouring reservoir made of donkey dungs and clay is created. This reservoir is placed between the second and third layers in other to stick everything together and create an entrance to pour the molten copper. The moulds are then left to dry and eventually placed in a dewaxing furnace.
Egyptian dewaxing furnaces appear similar to pottery kilns. Both types of furnaces have heath tunnels underground, where the fire is started33 and, both have their vault split into two parts. The bottom part (underground) is the zone where the heat tunnels converge. The flames enter through this part but cannot reach the upper part where the moulds/potteries had been placed. These two parts are separated by a grid on which the moulds are posed. During excavation, Verly noted that none of the moulds34 showed signs of fire exposure and that they had a porous surface. To obtain this porosity, three temperature plateaus must be respected. This process takes around 9 hours and reaches 650 degrees. The furnace is then no longer alimented and cools down for a few hours. Moulds, emptied of wax residue, are retrieved. Bronze can finally be poured in.
Aztec tradition
The Florentine Codex offers a relatively detailed explanation of the lost-wax technique developed by the Aztecs, notably the realisation of bell moulds35.
33 at the exterior of the furnace
34 He studied several Egyptian moulds already emptied of their wax but not yet filled with bronze 35
Once more, I do not state that the Aztec technique is perfectly applicable to the Maya. However, greater similarities are to be expected.
39 The core is composed of a mix of charcoal and clay and is covered by a layer of rolled beeswax and copal resin. The model is then covered with pounded charcoal and coarse clay. Finally, a channel and a pouring reservoir are added to it. The mould is then left to dry for a couple of days before being placed in a brazier to evacuate the wax.
Among the currently noticeable variances between these two techniques and the tradition developed by the Maya, I noted that the tempers differ. Indeed, investigations support the potential use of copper dust to heighten their mould’s pyro resistance (Ernst 2019: private conversation). Next, no trace of donkey dung and sand have been revealed during analyses, but the use of chamotte is attested (J.L. Meanwell et al., 2013: 4310). Unfortunately, the somewhat limited knowledge we have on their layers does not allow us to verify their composition, nor their exact number.
Lastly, signs of fire exposure are to be expected if a ceramic comes in direct contact with flames. Therefore, to verify whether Maya used open brazier instead of a dewaxing kiln, a closed-up inspection of the mould’s surface might reveal information.
3.2.2. Annealing and Open-mould Casting
The vast majority of metallic objects are cast using the lost-wax technique. However, a few were realised via a different process. Indeed, the above-mentioned procedure is rather costly since both the model in wax and the mould in clay are lost in order to retrieve the object. Casting in an open mould is much easier since the form of the object, one wishes to cast, must simply be drawn in an open mould made of either sand or mud36. Consequently, most copper-alloyed artefacts considered utilitarian, such as fish hooks, pins and needles are manufactured using the open-mould technique.
So far, twenty-three axes have been found at Lamanai (Shugar and Simmons 2012: 145), scholars initially thought that they were cast in an open-mould and then hammered and sharpened to obtain the final product. Such a process fragilizes the axe. Consequently, in
36 And is, therefore, easier to replicate. Ronde-bosse figurines or any 3D modelled objects, are much
harder (if not impossible) to cast using an open-mould and require almost systematically the use of wax as a medium.
order to increase the ductility of the material, continuous annealing was required (Rothenberg et al. 1978 in Simmons and Shugar 2008: 130). This method is visible through microanalyses that should reveal small grains of similar size with annealing twins. If on the contrary, an object was only hammered, distorted twins and working lines should appear (Shugar and Simmons 2012: 145). None of the two above-mentioned microstructures were seen through microanalysis. Simmons, therefore, theorises that Maya either produced them in a bivalve mould or via the lost-wax technique (Shugar and Simmons 2012: 147).
Fig. 12: Microanalysis of copper axes found at Lamanai
3.2.3. Summary of copper artefacts
Rings are the most abundant copper artefacts found at Tipu, with 26 complete objects and three fragments. Although some present a false filigree design, most were plain without any ornament. Similar designs can be found in West Mexico and Ecuador (Hosler 1994: Figures. 3.7, 4.4), as well as at Chichen Itza and Mayapán (Lothrop 1952:83–84 in Simmons and Cockrell 2018:167 ; Paris 2008:52).
41 Ornaments of copper have been found at Tipu and Lamanai and were likely used as pieces of adornment. Simmons found seven oval ornaments in one burial and described them as ‘made of interconnected S-scroll wires attached to an elliptical frame’. They measured slightly less than 3 centimetres in length and 2 cm in width. Alongside their ornamental value, these items may have also functioned as garment fasteners. This form has not been reported from any other sites in the Maya area (Bray 1977:385 in Simmons and Cockrell 2018: 168).
Fig.14: Clothing ornament found at Lamanai
Four tweezers have been recovered at Lamanai but none at Tipu. They are either disc-shaped or had elongated edges. Once more, alongside its ornamental value37, these objects were employed as hair removal (Simmons and Cockrell 2018: 169).
Fig. 15: Tweezers
At Lamanai, 23 axes, axe fragments, and chisels have been recovered. Mexican axes are usually thin and small, too fragile to be used as a woodcutting instrument. Scholars have determined that those axes may have been used as a form of currency or at least ornaments. However, axes found in the lowlands are usually larger and more resistant.
This could suggest that axes held a utilitarian function in the Maya region instead of the symbolic value of their Mexican counterparts, but further research need to be done to determine it with certainty.
Lace tags are thin, hollow rods. Tipu contained nine late tags in its metallic assemblage, but none of them has been recovered at Lamanai. They all come from burials, excepting one situated in the west nave of a Spanish church (Simmons and Cockrell 2018: 170).
Fig. 16: Lace tags found at Lamanai
Needles have been unearthed at both sites; 17 at Tipu and ten at Lamanai. The needles are thin, solid rods ending in a loop (Simmons and Cockrell 2018: 170).
Fig. 17: Picture of a needle found at Lamanai
A total of five fish hooks have been found at Lamanai and Tipu. They were situated in commoner settings. These towns proximity to watercourses easily explain the presence of fishhooks (Simmons and Cockrell 2018: 172).
