Skip to main content

Expanding the Taquara/Itararé Model: Feasting and Social Complexity in the Southern Brazilian Highlands

by Philip Riris[1], Department of Archaeology, University of Exeter



What avenues to power were taken by emergent elites in early complex societies? Archaeological patterns that concern ceremonial mortuary architecture, feasting activity and hot-rock cooking facilities are considered relevant in moving towards an answer to this question and will be described in a discursive outline. Following new fieldwork in July-August 2008, a case study from the twelfth and thirteenth centuries in Misiones Province, Argentina, site PM01, is presented in an abbreviated site report. The new PM01 data is considered in the light of known literature on mortuary architecture from the southern Brazilian highlands, feasting practices and ceremonial architecture. New data from PM01 confirms earlier findings by Iriarte et al. (2008) and appears to correlate with an emergently complex political structure based on lineal descent in the Taquara/Itararé tradition, closely related to commensal feasting practices and ancestor veneration.

Keywords: Feasting, emergent complexity, Argentina, Taquara/Itararé, burial mounds



Questions on socio-political complexity in pre-Hispanic food-producing groups have frequently been of interest to archaeologists working with prehistoric cultural remains in the Americas (e.g. Siegel, 1999; Carr, 2006; Dillehay, 2007). Deeper comprehension of the historical processes that led to the emergence of complex societies permits archaeologists and anthropologists to understand the means by which elites gained power, reproduced and maintained it, and whether that power was ascribed or achieved (Hegmon, 2005). Approaches adopted towards these problems have commonly incorporated multiple lines of research, such as palaeoenvironmental data (Bitencourt and Krauspenhar, 2006; Iriarte and Behling, 2007) or the reconstruction of trade networks, thought to represent patterns of production and consumption across social domains (Lazzari, 2005). Particular attention has been paid, however, to architectural and artistic expressions of material culture as an index of the presence and level of hierarchy in a given society (Kolb, 1994).

Archaeological explorations of the broader topic of emergent complexity in South America, have, as a result, tended to focus on Andean cultures. Arguably some of the earliest and most visually conspicuous indigenous remains are to be found in this region; the erection of public architecture has been a feature of Andean societies since the Late Preceramic, beginning roughly around 3000 BC (Moseley, 2001; Silverman, 2004). In comparison, lowland regions, chiefly the Amazon and La Plata river basins, have until recently been deemed incapable of supporting politically centralised populations with associated settlements and agricultural systems. Debates on the veracity of this position have continued to clash along the lines of ecological determinism and cultural variability (see Meggers, 1954; Meggers, 2001; Neves, 1998; Heckenberger et al., 2001). Readers of this article need to be aware of this salient issue concerning archaeology outside the Andean Cordillera.

Figure 1: Map of approximate distributions of pre-Hispanic cultural groups, including the Taquara/Itararé tradition

Figure 1: Map of approximate distributions of pre-Hispanic cultural groups, including the Taquara/Itararé tradition. Marked point 1 is the location of site PM01. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)
[Click on image for full-size version]

The question remains of how complex, in the socio-political sense, were the peoples of lowland South America, in what ways were they complex and by what processes did this occur? To make progress towards an answer, a ceremonial mortuary monument from the Taquara/Itararé tradition (Iriarte et al., 2008: 947) of the southern Brazilian highlands (Figure 1) is presented here as a case study. Following the research of Iriarte et al. (2008), the focus of which was the significance of memorial feasting to social complexity, I will contextualise and critically assess new evidence from the fieldwork season of July-August 2008 at the site of PM01 in Misiones, Argentina and provide interpretations of the activities that were performed which are visible in the archaeological record. The new data will be compared to and incorporated with other evidence. For the purposes of this article, the aforementioned questions concerning societal complexity can be rephrased as: what archaeologically detectable social practices were acted out in the past and contributed to the process of emergent Taquara/Itararé socio-political complexity?


The Taquara/Itararé Tradition of Southern Brazil

The Taquara/Itararé tradition (Beber 2004) of the southern highlands of Brazil is characterised by the construction and maintenance of ceremonial mound/enclosure complexes such as PM01. The term merges the two names given to the pre-Hispanic peoples of Rio Grande do Sul and Paraná provinces, respectively, chiefly as a matter of convenience (Iriarte and Behling, 2007: 116, 120). Groups related to the Taquara/Itararé, of linguistic stock, spread southwards to the study region from central Brazil after 3000 BP, and have been implicated in the introduction of agriculture and ceramics with their southern expansion (Marrero et al., 2007: 302; Iriarte et al.,2008: 948; Bitencourt and Krauspenhar 2006: 110). With some discontinuity following European contact in the sixteenth and seventeenth centuries, their direct descendants consist of the post-contact Kaingang and Xokleng groups described in ethnographies by, amongst others, Métraux (1947), Alphonse Mabilde (Dias, 2004) and da Silva (2002).

The broadly defined Taquara/Itararé tradition is characterised by its round-bottomed fine ceramics with incised decorations and a mixed subsistence economy based on hunting, fishing and swidden farming supplemented with gathering starch-rich Araucaria seeds (Meggers and Evans, 1983: 313-14; Iriarte and Behling, 2007: 120). Settlements consist of clustered pithouse dwellings, or casas subterrâneas in the Brazilian literature.

Individual pithouses in a group are often not contemporary, but are thought to be successively occupied (Dias, 2004: 189). To this end, a degree of seasonality has been suggested in the character of Taquara/Itararé settlements, with groups spending time between various settlement sites and their associated fields and Araucaria forest stands (Beber, 2004: 257). Most mound/enclosure complexes date from around AD 1050, possibly showing an increase in the frequency of this cultural practice from this time onwards (Iriarte et al., 2008, Table 1). The primary use of mounds/enclosures is shown to be as mortuary monuments (Beber, 2004: 260). A theorised correlation exists between increasing territoriality, monumentality and the displacement of Taquara/Itararé groups to higher altitudes in the Planalto (between 600-1200 masl) by Tupi-Guarani groups (Politis, 2008: 254-55; Iriarte et al.,2008: 948, 959).

Palaeoenvironmental data from the region indicates a major shift in the vegetation and climatic conditions of the southern Brazilian highlands starting from circa 1500 years BP in Paraná, 1140 years BP in Rio Grande do Sul and 1000 years BP in Santa Catarina (Bitencourt and Krauspenhar, 2006: 110). The pollen record shows a marked growth in Araucaria angustifolia (Paraná pine) forest, signifying a partial replacement of campos-type grasslands by mixed Araucaria forest. This change is registered alongside and in part due to the emergence of a much more wet annual weather cycle and with it a far less marked dry season. Wetter climatic conditions notwithstanding, a jump in charcoal particulates is observed with the spread of Araucaria forest, indicating firesetting activity by humans in the newly abundant tree cover (Figure 2) (Iriarte and Behling, 2007: 116-20). It was against this backdrop of an environment in transition that the Taquara/Itararé tradition flourished.

Figure 2: Pollen profile registering late Holocene vegetational changes in the southern Brazilian highlands

Figure 2: Pollen profile registering late Holocene vegetational changes in the southern Brazilian highlands. (Source: Behling and Pillar, 2007. Reproduced with kind permission of Hermann Behling and The Royal Society)
[Click on image for full-size version]

July-August 2008 Excavation at PM01

The following section provides a summarised description of the new data from PM01 retrieved in excavations between 7 July and 23 August 2008 as an abbreviated site report.

PM01, designated as the first site in the Piray Mini river valley, is located in the Eldorado Department, Misiones Province, north-eastern Argentina. The province is bordered on two sides by Brazil and on the two others by Paraguay and Corrientes Province. PM01 is part of a complex of eight circular earthen enclosures located in what is now a cultivated Araucaria angustifolia (Paraná pine) plantation circa 19 kilometres north of the city of Eldorado (Figures 3 and 4) (Iriarte et al., 2008: 949). Institutions involved in the excavations at PM01 include the University of Exeter, the Universidade Federal do Rio Grande do Sul and the Universidad Nacional de Misiones. The project is directed by Dr José Iriarte of the University of Exeter and funded by National Geographic Society.

 Figure 3: Map of PM01 (1) and surrounding enclosures shown with the Piray Mini river, North of El Dorado

Figure 3: Map of PM01 (1) and surrounding enclosures shown with the Piray Mini river, North of El Dorado. (Source: Iriarte, 2008, after Wachnitz, 1984. Reproduced with kind permission of Jose Iriarte)

 Figure 4: Location of PM01 within Misiones Province, Argentina

Figure 4: Location of PM01 within Misiones Province, Argentina. (Source: Google Earth)

The primary aim of the July-August 2008 field season was to continue excavating PM01 in the same vein as the seasons of 2006-7, following the completion of geophysical mapping (Figure 5) and test excavations on the eastern and north-western edges of the earth embankment. Conglomerations of rocks interpreted as 'earthen ovens' by the authors were discovered (Iriarte et al., 2008: 957), which warrants further investigation to expand the dataset. Refer to Iriarte et al. (2008) for a more detailed discussion of this past work. The 2008 season saw an extensive enlargement of previously excavated areas on the north-western rim (see Figure 6), in order to confirm the hypothesis of the site hosting repeated feasting events in conjunction with post-funerary rites (Iriarte et al., 2008: 959). Other archaeological research efforts have previously been focussed on the central mound (see Menghin, 1957, and Wachnitz, 1984).

Figure 5: Combined geophysical and topographical map of PM01, partially reconstructed from Wachnitz

Figure 5: Combined geophysical and topographical map of PM01, partially reconstructed from Wachnitz (1957). Blue points indicate past test excavations, red points indicate July-August 2008 excavation. (After: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)
[Click on image for full-size version]

The ultimate goal of the project investigating the archaeology of Misiones province, and the southern Brazilian highlands as a whole, is to seek new data to corroborate or refute previous assumptions on indigenous societies (e.g. the environmental determinism of Betty Meggers, 1954). By studying the activities that took place at conspicuous locales such as mortuary mounds, for example feasting, rites of passage or ancestor veneration (Dillehay, 1990; Barrett, 1996; Iriarte et al., 2008) we may gain insights into social processes leading to the emergence of complexity in intermediate societies. The Taquara/Itararé are thus far relatively under-theorised in mainstream archaeological literature despite diligent regional academics investigating the tradition since the sixties (e.g. Schmitz, 1999; but see Saldanha, 2008). PM01, as recorded by Menghin (1957), was a 'key-hole shaped' (Iriarte et al., 2008: 948) circular enclosure with a central mound. Two ringlets adjoin the main circle on the eastern and north-western edges of the rim, the latter being the larger of the two. The main enclosure (Circle I) is near-perfect circle, consisting of an earthen bank with a shallowly rising profile between 30-70 cm in height and 6 m wide. The central mound, 20 m wide and 3 m high, is the focal point of the mortuary architecture at PM01. Wachnitz's work (1984: 174) yielded the likely remains of a human cremation at the base of the anthropogenic layers in the mound. With no evidence of differential stratigraphy in the mound, this suggests a single large construction event (Iriarte et al., 2008: 956). The summit of the hill, from which the central mound commands a view of the surrounding landscape, had an avenue, 18 m wide and 400 m long, of two parallel earth banks leading up to it. Forestry activity has since erased approximately half of the southern end of the site, including the avenue (Figure 5).

Figure 6: Plan of the excavated portions of PM01 from the July-August 2008 field season, showing all excavated and recorded features

Figure 6: Plan of the excavated portions of PM01 from the July-August 2008 field season, showing all excavated and recorded features. (Provided by and reproduced with kind permission of Jose Iriarte, 2008)
[Click on image for full-size version]

As described in Iriarte et al. (2008), the 2006/7 excavations on the north-western intersection of Circles I and II returned evidence of repeated feasting events taking place on the banks of the PM01 enclosure. The archaeological signature of the practice is evident here in the form of miscellaneous fine ceramic sherds, lithic flakes, charcoal and clusters of stones in various densities. The chronological relationships between stone clusters are only known through radiocarbon dating; the anthropogenic material exhibits no stratigraphical differentiation. Furthermore, the full extent and morphological consistency of the stone clusters was not known, due in part to the small scale of the original excavation and differing results from the eastern sector of Circle I (Iriarte et al., 2008: 953). With this in mind, the 2008 season took place to enhance the presently available information on the formation, uses and attributes of PM01.


Results of Excavation

I will draw the distinction here at the start of this section between artefactual and architectural remains. For reasons of practicality, artefacts such as those mentioned in the previous section (essentially portable items) are described under the heading Artefacts. Furthermore, stone clusters will be the primary focus of the section on Major features. I acknowledge that stone clusters are fundamentally artefactual in nature; however, these features are intrinsic in the formation of the enclosure at PM01 (which will be discussed in a later section) and are therefore more practically dealt with in their own sub-section.


Taquara/Itararé burials in mound/enclosure complexes will commonly yield evidence of one or more human cremations at the base of the central mound, often accompanied by ceramics and lithic tools (Iriarte et al., 2008, 956). Conversely, a relative paucity of artefactual remains from the enclosure rim was recorded; the high acidity of the local iron-rich soils resulted in very little bone being recovered and the few remains that were found were too fragmented or damaged to be diagnostic. Stone tools were conspicuously absent from all levels. However, informal tools in part consisting of flakes with a basalt cortex were encountered. It is thought that flakes were struck off cores as needed for performing cutting tasks during periods that PM01 was in use (Oscar Marozzi, personal communication). Flakes were found in both stratigraphically low contexts with stone clusters and higher contexts with ceramic sherds, demonstrating no readily detectable patterning.

Figure 7: Profiles of Taquara/Itararé drinking cups, as recovered from PM01

Figure 7: Profiles of Taquara/Itararé drinking cups, as recovered from PM01. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Taquara/Itararé tradition pottery present in mound/enclosure contexts are generally small drinking vessels of fine material (Figure 7) suggested to have been used for drinking a maize-based beverage, based on phytolith analysis on carbonised residue (Iriarte et al., 2008: 954). The few sherds recovered this season were largely located in the levels immediately preceding stone clusters. Although typical of the tradition, very few were identifiable as types through profiles or decorations. Carbonised plant matter was the most common class of artefact encountered in the course of excavation. Where there was a sufficient quantity for 14C dating, a sample was always taken and recorded. In direct association with stone clusters, intact logs of carbonised wood were discovered in trenches B4 to B6, C5 and A4 (Figure 6).

Major Features

By far the most prominent remains at PM01, the monument itself notwithstanding, were stone clusters. Of the seven clusters (possibly eight or nine, see below, Discussion of the 2008 season) so far uncovered, two were known of from previous work (Iriarte et al., 2008: 953). Carbonised plant material was always found in the levels above a stone cluster and in a single case, within a cluster located chiefly within B7 (Figure 8) during dismantling. As illustrated in Figure 6, the clusters revealed in 2008 follow the curve of Circle I in a roughly south-west to north-east alignment.

Figure 8: Disassembling a cluster to obtain charcoal

Figure 8: Disassembling a cluster to obtain charcoal. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Following guidelines by White (1980), some brief generalised descriptions of clusters are provided. All recorded clusters were found at a depth between 1.20 and 1.25 m, with the exceptions of the clusters located in D-2, D-4 and E-3, which were found at 1.30 m. This was to be expected from the greater height that the enclosing bank rises to in this area. No clusters consisted of more than a single layer of stones, and rarely exceeded a total surface area of circa 1.5 m2. Apart from various stones and pebbles scattered around the clusters, which probably happened as part of the site formation process, rocks in clusters were densely packed together. Fill between individual stones consists primarily of earth, with fragments of charcoal interspersed. Charred logs were found in association with and on top of some clusters. There were no instances of thermal fracturing, indicative of a basalt or similar mineral (Marozzi, 2008, personal communication). Fragments of burnt earth were occasionally encountered, although commonly they did not occur in concentrations greater than a few pieces each less than a centimetre across. At least three major patches, located largely within A4, D3 and E2, were discovered, the latter two being new discoveries in 2008. The burnt earthis distinctive due to its hardened character and density when compared to the surrounding soils. The largest of such formations (E2) is comparable in size to some stone clusters. Processes leading to the creation of this type of feature are not well understood, nor are the exact relationships between patches and stone clusters. They may represent the remains of hearths or bonfires used alongside or instead of clusters. Finally, there are two isolated cases of post holes in C8 and C4. Neither were more than 10 cm in diameter and of similar depth. Until more are located and recorded, it is difficult to evaluate their relationship to the site as a whole and whether they are isolated features or the remnants of larger structures.


Discussion of the 2008 Season


The 2008 field season has produced new evidence to expand further the current framework of the Taquara/Itararé tradition. Site PM01 has already been excellently described and discussed by Iriarte et al. (2008). Recent evidence from the July-August 2008 season supports their interpretation of PM01 as a funerary monument where an individual vested with importance may have been buried under the central mound in a single event followed by repeated feasting over several generations. The archaeology shows a degree of cultural continuity between Taquara/Itararé mound construction and later ethnographic descriptions of Kaingang commemorative feasting at mortuary monuments (Iriarte et al., 2008: 957-58; de Masi, 2009: 111). The discussion presented here will seek to build on the conclusions of Iriarte et al. (2008), but also to raise questions based on the new data detailed previously in this study.

Stone clusters, broadly similar in morphology, have been found in both the eastern and western edges of the site (Iriarte et al., 2008: 953). The embankment where Circle I and II converge has revealed a wealth of data on the character and spatial relationships between clusters. Visible to us now, through new data and radiocarbon dating, is the chronological superimposition of a variety of features over a period of time close to a century and a half in length. Depending on the working definition of what constitutes a discrete cluster, interpretation can vary as to the exact number of excavated or partially excavated clusters, so that there appears be to between seven and nine earthen ovens. The cluster(s) linked by only one rock, occupying C1 and D1 are a pertinent example, as is the thus far unknown full extent of the cluster(s) in D-2, D-4 and E-3. More data needs to be incorporated before it is possible to discuss site and cluster formation with accuracy or authority. Research in this direction can include creating predictive models of site formation that can aid the expectations of later excavations at similar sites.

The central mound was almost certainly raised in a single construction event, but only concerns the discussion given here in relation to the surrounding enclosure. The notion of successive creation of clusters has previously been put forward by Iriarte et al., (2008: 954) but the present condition of the enclosure is not fully explained.

Figure 9: Upper anthropogenic levels of PM01. Stones of this character were encountered across the excavated area in the upper strata

Figure 9: Upper anthropogenic levels of PM01. Stones of this character were encountered across the excavated area in the upper strata. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Frequent revisiting of the monument is probably implicit in the formation of the enclosure, and consequently, I propose that earthen ovens were fired and utilised for cooking more than once. At the risk of devolving into reductionist arguments, this is in part due to the labour expenditure involved in transporting rocks in the quantities needed for a typical stone cluster of 1 to 2 m² (in excess of 60 rocks, each roughly the size of a closed fist) from the Piray Mini river, roughly 2 km away. A large quantity of fuel is required for efficient firing (Dering, 1999: 665). Re-use of stones already located on-site is plausible, given their source and mass. If there were no motives to maintain ovens in the enclosure beyond cooking foodstuffs, cooking would conceivably have taken place on hot coals instead (Thoms 2009, 587). It is also suggested that the loose aggregations of rocks associated with charcoal, lithics and bone in the occupational strata preceding the compact stone clusters (Figure 9) are the result of periodic removal in order to expose clusters for cooking and subsequent backfilling to maintain the integrity of the enclosure. In this view, little to no stratigraphic differentiation between clusters and variation in 14C dates between clusters support this argument; there is no reason to suspect all the clusters were re-used in every visit to PM01. Furthermore, analogous patterns of re-use can be found in the archaeology of North American pre-Hispanic groups (Figure 10), albeit not in ceremonial contexts, giving an overall idea of how the enclosure formed.

Figure 10: Examples of profiles from the remains of multiple-use earthen oven mounds, bearing similarity with PM01 enclosure profile 

Figure 10: Examples of profiles from the remains of multiple-use earthen oven mounds, bearing similarity with PM01 enclosure profile. (Source: Thoms, 2009. Reproduced with kind permission of Alston Thoms)

Quoting Maybury-Lewis (1974: 42), Iriarte et al. (2008, 957-58) note that 'the Shavante, in common with other tribes, value meat and maize as the basis of all ceremonial prestations'. Alcoholic beverages and maize are a central component of ritual activity in many, but not all, groups of linguistic stock. Some commentary is needed on the proposition put forth by Iriarte et al. (2008: 957) that a weak alcoholic fermented maize beverage was consumed, based on ethnographic, palaeobotanical and ceramic evidence. While the forms of pottery from PM01 indicate consumption of maize-based foodstuffs due to residue analysis and the individual 'serving size' of the ceramics (José Iriarte, personal communication; see also Figure 7), there is little evidence of maize beer brewing facilities at PM01. Modern practices of the Kaingang indicate that production took place in perishable containers made from hollow Araucaria trunks, and importantly, that this process was carried out with specific reference to mortuary rites (Veiga, 2000: 7). Furthermore, production methods documented elsewhere in the ethnographic record generally indicate that the process may generally take between three days and a week to complete. The resulting beverage should be consumed within three days of fermentation being initiated to avoid it spoiling (Moore, 1989: 685-86; Nicholson, 1960: 297). On a speculative level, production of an alcoholic drink provides clues as to how long any given period of feasting activities would last; assuming a single production and subsequent consumption event, by inference a maximum period of seven days appears appropriate for a festival or feast to have lasted. Alcoholic beveragesare bound up in systems of ritual value across Amazonian contexts (Maybury-Lewis,1974: 42; Uzendozki, 2004; Hornborg, 2005: 604). Therefore it may be noted that more data to this effect in the context of Taquara/Itararé would be of further value to the discussion at hand.

The reconstructed spatial layout of PM01 is the last point touched on in this section. Following broad trends in Taquara/Itararé monumental architecture, the site follows a roughly north-south orientation, with at least one entrance and an enclosure of ellipsoid, rectangular or circular shape. Circle II is the the largest extraneous structure associated with the main Circle I, and like other such architectural elaborations, is located in the western portions of the site (Beber, 2004: 254-55; Iriarte et al., 2008: 948-51; De Souza, 2007). While mounds are not ubiquitous in all enclosures, funerary mounds always have an associated enclosure. Neither are all Taquara/Itararé burials in mounds; cave sepulchres were also commonly used for the deposition of the dead (Beber, 2004: 259; Dias, 2004: 190). The size of the complex at PM01, the corresponding scale of activities and the significance of PM01 within the broader framework of Taquara/Itararé complexity is worth contextualising further.


The conceptualisation of feasting as a cultural practice is an important touchstone for the arguments put forth in this report. Feasting, as defined here, is communal consumption of food and/or drink in a context that diverges significantly from everyday consumption practices. Feasting consists of that activity where communal eating and drinking happens alongside ritual practices and performances. Working with this definition provides scope for variability, but also a strict criterion setting such practices aside from regular daily intake of foodstuffs (Dietler and Hayden, 2001: 3; Brown, 2001: 370).

Given that feasting locales are often connected to prominent ritual structures, it should be profitable to address mortuary feasting as a stage for transformation and negotiation of social and cultural norms. To this effect, feasts are recognised to be central elements of specific biographical events in individuals' lives, up to and including death. It follows that they have the potential to be detected archaeologically through qualitative and quantitative differences in consumed foods, spatial segregation in these patterns, unusual modes of food preparation or architectural elaboration for dramaturgical effect (Palmer and van der Veen, 2002: 199; Dietler and Hayden, 2001: 9; Kertzer, 1988: 40). Ethnoarchaeological and historical records indicate that the act of providing food and drink, especially if alcoholic, in exchange for labour donations is 'a nearly universal pattern' (Lau, 2002: 280) in intermediate societies. Often this is the primary manner which individuals acquire the moral authority to influence group decisions, wield power or indeed, resist the coercion of others. Considerable symbolic and/or political capital may be exerted to provide an appropriate quantity of food for a feast, but a successful feast also regenerates any labour obligations held towards the host. Requirements for the successful execution of large-scale feasts include planning and forethought, especially with regards to the labour and time donations necessitated (Dietler, 2001: 80; Barrett, 1996: 403).

Returning to the subject of death, the various rituals that accompany the treatment of the dead in many societies are laden with symbolism and powerful appeals to emotion. It has clearly not escaped scholars' attentions that from prehistory to the present day individual claims to power have been affirmed or contested based on the relationships the dead had to the living (Fleming, 1973: 189; Dietler, 2001: 71). Foodstuffs are ingested and, in the literal sense, communicate potent symbolism and become embodied material culture during specific rites. Successful invocation of these symbols and messages reinforces group solidarity and the socio-political status quo, as well as claims to territories through a 'genealogy of place' linked to the interred dead (Barrett, 1996: 396; Lau, 2002: 281; Renfrew, 1979). By considering feasting activity in this light, and cross-referencing the data presented in this report, there is an intention to demonstrate that the formation of PM01 is a result of routine ritual activity that took place during its use.


Concluding Summary

The Taquara/Itararé tradition and PM01 in context

Without the efforts of scholars working in the highlands of southern Brazil and the theoretical models for the Taquara/Itararé tradition that have resulted from their work, writing this dissertation would be a far more strained affair. There is great substance to the academic papers, studies and frameworks that have been produced (represented here in part by Menghin, 1957; Wachnitz, 1984; Schmitz, 1999; Beber, 2004; Dias, 2004; Bitencourt and Krauspenhar, 2006; Saldanha, 2008; Iriarte et al., 2008; and de Masi, 2009). Bearing this in mind, this section offers my own observations, interpretations and critiques of archaeological research on Taquara/Itararé cultural tradition.

A single funerary interment appears to have taken place at the central mound of PM01 (Wachnitz, 1984). The chief mode of interaction with PM01 and its associated symbolism, however, was through the regular ritual consumption of food and drink. Ritualised interactions between strictly delineated corporate groups, in this case inferred to correlate with Kaingang moieties (Iriarte et al., 2008: 956; Métraux, 1947: 149), assisted in the maintenance of a socio-political order based on authority tied to ancestor worship. In a practical sense, then, Taquara/Itararé mortuary rituals are modes of social action 'trapped in a web of symbolism' (Kertzer, 1988: 9). This is accentuated by the communal taking of food and drink in a form of destructive gift exchange; ritually consumed foodstuffs cannot be recirculated immediately, but must be produced anew or reciprocated to the host by other means (Dietler, 2001: 73). There is therefore a substantial link between the remains of the dead, mortuary architecture and the veneration of both through ritualised feasting activity (Hastorf, 2002: 305).

The longevity of the ritual use of PM01 (measurable in generations) stems from its durable character as a mnemonic device through which one could experience the past. Repetitive ritually charged mortuary rites aids in the transformation of the interred dead into a genuine ancestral figure (Dillehay, 1990: 231; Rowlands, 1993: 141; Bradley, 1998: 90, 121; Hastorf ,2002: 307). Feasting is implicated in the working definition of ritual activity here; it is emphasised here that this monument was not merely made for feasts and festivals, but created and recreated through feasts and festivals. By viewing material culture, monuments and landscapes in the southern Brazilian highlands as active participants in the ritual dynamics of the Taquara/Itararé, a more holistic characterisation of the past can be produced (Saldanha, 2008: 87).

Concluding comments

Monumental architecture can be considered indications of increased social and economic separation between people, territorial markers, means to assert dominance over important material or symbolic resources and by-products of internal social strategies (Dillehay, 1990: 238-39). Social actions carried out by individuals always happen in relation to external entities and the material consequences of these actions are how archaeologists recover and study the past (Johnson, 2000: 214). PM01 is embedded in a ceremonial landscape of natural and cultural forces intent on relaying a message to observers and participants in ritual (Dillehay, 1990: 226), as evident by its central location in a complex of mound/enclosures and prominence at the head of a hill. Archaeology should be concerned with the material remnants that past peoples left of their actions and understandings of the world. In doing so, research must go beyond only recognising that actions have taken place from physical traces and actually understanding their performance, means of execution and historic context (Silverman, 1995: 1216; Barrett, 2000,:61; Barrett ,2001: 152).

Arguments and models of complexity presented in recent years have often designated societies as 'transegalitarian', or somewhere in the margins between egalitarianism and fully stratified hierarchical systems of lineal rule. There are also known cases of large and complex but ' essentially egalitarian' societies in archaeology (Chilton, 2005: 152). Based on the case study of PM01 presented here, the late Holocene context of the Taquara/Itararé tradition is a nuanced one. The historical circumstances of environmental change and external pressure from other groups are highly relevant for concluding arguments. The remains of feasting at PM01 represent mortuary practices with a pan-generational degree of temporal depth and cultural influence. Furthermore, the earliest available calibrated 14C date for PM01 fixes it after the identified start of these processes and towards the peak of the occupation of the southern Brazilian highlands (Iriarte and Behling, 2007: 125). PM01 is shown to have been built after these processes were underway.

Permanent structures erected for the purpose of feasting may reflect an intentional, institutionalised practice of supplying foodstuffs in groups for political leverage, as opposed to opportunistic provisioning of meals at culturally appropriate times and places (Dietler and Hayden, 2001: 9; Hayden, 2001: 53; Dietler, 2001: 71). Constantly renewed ties to ancestors and appeals to symbolism limit the conceivable alternatives by ceremonial participants, and thus the possibility for societal upheaval. This leads to a form of socio-political organisation being envisioned for the Taquara/Itararé that could have had deep roots through re-enactment and regular societal reproduction. Emergent socio-political complexity, here meaning weakly differentiated forms of inherited power, was the norm during the time PM01 was occupied and experienced. Change in cultural patterns, however, not constancy, is of importance in tracking the emergence of complex cultures and political elites (Hastorf, 2002: 306). When this occurred in the southern Brazilian highlands remains an open question for research and beyond the scope of this document, but an issue that would be interesting to pursue further as more sites are illuminated.




I would like to thank the following:

Dr José Iriarte for including me on the project which would eventually become the focus for this article and for giving astute commentary on earlier drafts, as well as kindly providing materials for inclusion.

Seán Goddard for drawing the plan of the 2008 excavation presented in this article. The Municipality of Eldorado for their hospitality and superb aid with accommodation and logistics during the 2008 field season.

The Direccíon General de Patrimonio Cultural y Museos del Gobierno de la Provincia de Misiones, Ruth Poujade and Julia Argentina Perié for granting permissions to carry out investigations in Eldorado, Misiones.

Oscar Marozzi, Jonas Gregorio de Souza, Luísa Nunes D'Ávila, Marcelo Silva Sanhudo and Avelino Gambim Jr. for keeping the 2008 season in Misiones interesting.

Research at site PM01 was funded by grants from the National Geographic Committee for Research and Exploration (CRE 7853-05) and the University of Exeter Exploration Fund.

Two anonymous reviewers for their helpful comments and suggestions towards improving this report.

I take full personal responsibility for any errors or omissions in this report; they are my own.


List of Illustrations

Figure 1: Map of approximate distributions of pre-Hispanic cultural groups, including the Taquara/Itararé tradition. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Figure 2: Pollen profile registering late Holocene vegetational changes in the southern Brazilian highlands. (Source: Behling and Pillar, 2007. Reproduced with kind permission of Hermann Behling and The Royal Society)

Figure 3: Map of PM01 (1) and surrounding enclosures shown with the Piray Mini river, North of El Dorado. (Source: Iriarte, 2008, after Wachnitz, 1984. Reproduced with kind permission of Jose Iriarte)

Figure 4: Location of PM01 within Misiones Province, Argentina. (Source: Google Earth)

Figure 5: Combined geophysical and topographical map of PM01, partially reconstructed from Wachnitz (1957). (After: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Figure 6: Plan of the excavated portions of PM01 from the July-August 2008 field season, showing all excavated and recorded features. (Provided by José Iriarte, 2008. Reproduced with kind permission of Jose Iriarte)

Figure 7: Profiles of Taquara/Itararé drinking cups, as recovered from PM01. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Figure 8: Disassembling a cluster to obtain charcoal. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Figure 9: Upper anthropogenic levels of PM01. (Source: Iriarte et al., 2008. Reproduced with kind permission of Jose Iriarte)

Figure 10: Examples of profiles from the remains of multiple-use earthen oven mounds, bearing similarity with PM01 enclosure profile. (Source: Thoms, 2009. Reproduced with kind permission of Alston Thoms)



Excavation Methodology

A total surface area of 35 m2was excavated. The directed efforts of the 2008 season were based on the 6×1m trench (A to F -1) backfilled from previous seasons. Before any new trenches could be opened, the older backfilled excavations were re-exposed. A new numbering system was developed for 2008 by Oscar Marozzi. Levels were arbitrarily set and measured at 5 centimetre intervals; with no discernible difference in soil character in the anthropogenic layers being apparent, this was the chief method of recording depth and progress. All spatial measurements were taken with a levelling instrument and measuring rod.

The basic unit of the excavation was 1×1 metre trenches, which were most often worked by a single excavator using shovel-skimming and trowelling techniques, the former to clear topsoil and the latter to uncover more delicate remains at lower depths. Where extra fine work was needed (e.g. removal of a charred log or cleaning a stone cluster), two individuals could work together on a single archaeological feature with spatulas and picks, in order to expedite the process and hasten the descent to the next level. Disregarding topsoil and backfilled trenches, all soil was sieved through a ¼ cm mesh screen. The moisture content of the soil led to small artefacts (e.g. sherds) being embedded in clods of sticky soil and care had to be taken when sieving to recover any small lithics, sherds, burnt earth or bone that may have passed the excavator's notice.

Written records were made exclusively by Oscar Marozzi, noting the depth, northing and easting of (1) isolated rocks vertically presaging the appearance of stone clusters, (2) larger concentrations of charcoal and burnt logs, (3) uncommon or extraordinary finds like ceramic sherds or knapped chert (4) any other notable feature, such as post holes and large patches of distinctive burnt earth. Digital photographs accompany the written records, annotated with the site name, depth, trench number and an indication of north.


Readily portable artefacts were placed in individual see-through zip-lock bags annotated with the same information as present on photographic records. In addition, the artefact type and northing/easting were noted. Charcoal was wrapped in aluminium foil and not handled with bare hands during recovery, so as to avoid contamination in the event the sample would be used for radiocarbon dating. The collection of finds is currently curated at Museo Casa del Fundador, Departamento de Cultura, Eldorado.



[1] Philip completed his BA (Hons) in Archaeology at the University of Exeter in 2009, and is currently reading for his MA in archaeology at the same institution with a focus on the archaeology of lowland South America.



Barrett, J.C. (1996), 'Neolithic and Early Bronze Age Mortuary Practices', in Preucel, R.W. and I. Hodder (eds), Contemporary archaeology in theory, Cambridge: Blackwell, pp. 394-405

Barrett, J. (2000), 'A thesis on agency', in Dobres, M-A. and J. Robb (eds), Agency in Archaeology, London: Routledge, pp. 61-68

Barrett, J. (2001), 'Agency, the duality of structure, and the problem of the archaeological record', in Hodder, I. (ed.), Archaeological Theory Today, Cambridge: Polity Press, pp.141-64

Beber, M.V. (2004), O sistema do asentamento dos grupos ceramistas do planalto sul-brasilero: o caso da Tradiçao Taquara/Itararé, Available at, accessed 8 April 2009

Behling, H. and V. D. P. Pillar (2007), 'Late Quaternary vegetation, biodiversity and fire dynamics on the southern Brazilian highland and their implication for conservation and management of modern Araucaria forest and grassland ecosystems', Philosophical Transactions of the Royal Society B, 362, 243-51

Bitencourt, A. and P. Krauspenhar (2006), ' Possible Prehistoric Anthropogenic Effect on Araucaria angustifolia (Bert.) O. Kuntze Expansion During the Late Holocene', Revista Brasileira de Palaeontologia, 9 (1), 109-16

Bradley, R. (1998), The significance of monuments: on the shaping of human experience in Neolithic and Bronze Age Europe, London: Routledge

Brown, L. A. (2001), 'Feasting on the Periphery: The Production of Ritual Feasting and Village Festivals at the Cerén Site, El Salvador', in Dietler, M. and B. Hayden (eds), Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and Power, London: Smithsonian Institution Press

Carr, C. (2006), 'Scioto Hopewell Ritual Gatherings: A Review and Discussion of Previous Interpretations and Data', in Carr, C. and D. T. Case, Gathering Hopewell: Society, Ritual and Ritual Interaction, New York: Springer Science + Business Media, pp. 463-79

Chilton, E. (2005), 'Farming and Social Complexity in the Northeast', in Pauketat, T.R. and D. Loren, North American Archaeology, Oxford: Blackwell, pp.138-60

da Silva, S. B. (2002), 'Dualismo e Cosmologia Kaingang: O Xamã e o Domínio da Floresta', Horizontes Antropológicos, 8 (18), 189-209

Dering, P. (1999), 'Earth-Oven Plant Processing in Archaic Period Economies: An Example From a Semi-Arid Savannah in South-Central North America', American Antiquity, 64 (4), 659-74

de Masi, M. A. N. (2009), 'Centros Cerimoniais do Planalto Meridonal: Uma Análise Intrasítio', Revista de Arqueologia, 22 (1), 99-113

de Souza, J. G. (2007), 'Significados da morte: interpretando as estruturas funerárias de Pinhal da Serra (RS) e Anita Garibaldi (SC)', Anais do XIV Congresso da Sociedade de Arqueologia Brasileira (CD-ROM)

Dias, J. (2004), A Tradição Taquara e sua Ligação com o Índio Kaingang, available at, accessed 8 April 2009

Dietler, M. and B. Hayden (2001), 'Digesting the feast - Good to eat, good to drink, good to think: an introduction', in Dietler, M. and B. Hayden (eds), Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and Power, London: Smithsonian Institution Press, pp. 1-22

Dietler, M. (2001), 'Theorizing the Feast: Rituals of Consumption, Commensal Politics, and Power in African Contexts', in Dietler, M. and B. Hayden, Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and Power, London: Smithsonian Institution Press, pp. 65-114

Dillehay, T. D. (1990), 'Mapuche Ceremonial Landscape, Social Recruitment and Resource Rights', World Archaeology, 22 (2), 223-41

Dillehay, T. D. (2007), Monuments, empires, and resistance: the Araucanian polity and ritual narratives, New York: Cambridge University Press

Fleming, A. (1973), 'Tombs for the Living', Man, New Series, 8 (2), 177-93

Hastorf, C. A. (2003), 'Community with the ancestors: ceremonies and social memory in the Middle Formative at Chiripa, Bolivia', Journal of Anthropological Archaeology, 22, 305-22

Hayden, B. (2001), 'Fabulous Feasts: A Prolegomenon to the Importance of Feasting', in Dietler, M. and B. Hayden, Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and Power, London: Smithsonian Institution Press, pp. 23-64

Heckenberger, M. J., J. B. Petersen and E. G. Neves (2001), 'Of Lost Civilizations and Primitive Tribes, Amazonia: Reply to Meggers', Latin American Antiquity,12 (3), 328-33

Hegmon, M. (2005), 'Beyond the mold: Questions of Inequality in Southwest Villages', in Pauketat, T. R. and D. Loren (eds), North American Archaeology, Oxford: Blackwell, pp. 212-29

Hornborg, A. (2005), 'Ethnogenesis, Regional Integration and Ecology in Prehistoric Amazonia: Towards a System Perspective', Current Anthropology, 46 (4), 589-607

Iriarte, J. and H. Behling (2007), 'The expansion of Araucaria forest in the southern Brazilian highlands during the last 4000 years and its implications for the development of the Taquara/Itararé Tradition', Environmental Archaeology, 12, 115-27

Iriarte, J., J. C. Gillam and O. Marozzi (2008), 'Monumental burials and memorial feasting: an example from the southern Brazilian highlands', Antiquity, 82 (318), 947-61

Johnson, M. (2000), 'Self-made man and the staging of agency', in Dobres, M-A. and J. Robb (eds), Agency in Archaeology, London: Routledge, pp. 213-31

Kertzer, D. I. (1988), Ritual, Politics and Power, London: Yale University Press

Kolb, M. J. (1994), 'Monumentality and the Rise of Religious Authority in Precontact Hawai'i [and Comments and Reply]', Current Anthropology, 35 (5), 521-47

Lau, G. F. (2002), 'Feasting and Ancestor Veneration at Chinchawas, North Highlands of Ancash, Peru', Latin American Antiquity, 13 (3), 279-304

Lazzari, M. (2005), 'The Texture of Things: Objects, People and Landscape in Northwest Argentina', in Meskell, L. (ed) Archaeologies of Materiality, Oxford: Blackwell, pp. 126-61

Marrero, A. R., W. A. Silva-Junior, C. M. Bravi, M. H. Hutz,M. L. Petzl-Erler, A. Ruiz-Linares, F. M. Salzano and M. C. Bortolini (2007), 'Demographic and Evolutionary Trajectories of the Guarani and Kaingang Natives of Brazil', American Journal of Physical Anthropology, 132 (2), 301-10

Maybury-Lewis, D. (1974), Akwẽ-Shavante Society, New York: Oxford University Press

Meggers, B. J. (1954), 'Environmental Limitation on the Development of Culture', American Anthropologist, 56 (5), 801-24

Meggers, B. J. (2001), 'The Continuing Quest for El Dorado: Round Two,' Latin American Antiquity, 12 (3), 304-25

Menghin, O. F. (1957), 'El poblamiento prehistórico de Misiones', Anales de Arqueología y Etnologia, 12, 19-40

Métraux, A. (1947), 'Social Organization of the Kaingang and Aweikóma According to C. Nimuendajú's Unpublished Data', American Anthropologist, 49 (1), 148-51

Moore, J. D. (1989), 'Pre-Hispanic Beer in Coastal Peru: Technology and Social Context of Prehistoric Production', American Anthropologist, 91 (3), 682-95

Moseley, M. (2001), The Incas and their Ancestors, London: Thames & Hudson

Neves, E. G. (1998), 'Twenty years of Amazonian archaeology in Brazil (1977-1997)', Antiquity, 72 (277), 625-32

Nicholson, G. E. (1960), 'Chicha Maize Types and Chicha Manufacture in Peru', Economic Botany, 14 (4), 290-99

Palmer, P. and M. van der Veen (2002), ' Archaeobotany and the social context of food', Acta Palaeobotanica, 42 (2), 195-202

Politis, G. (2008), 'The Pampas and Campos of South America', in Silverman, H. and W.H. Isbell (eds), Handbook of South American Archaeology, Berlin: Springer Scienceand Business Media, pp. 235-62

Renfrew, C. (1979), Investigations in Orkney, London: Societies of Antiquaries

Rowlands, M. (1993), 'The Role of Memory in the Transmission of Culture', World Archaeology, 25 (2), 141-51

Saldanha, J. D. (2008), 'Paisagem e Sepultamentos nas Terras Altas do Sul do Brasil', Revista de Arquelogia, 21, 85-95

Schmitz, P. I. (1999), 'Arqueología do Planalto Sul-Brasileiro', Revista de Arqueología Americana, 17-19, 57-74.

Siegel, P. E. (1999), 'Contested Places and Places of Contest: The Evolution of Social Power and Ceremonial Space in Prehistoric Puerto Rico, Latin American Antiquity, 10 (3), 209-38

Silverman, H. (1995), 'Review: Rites of Revitalization', Science, 268 (5214), 1215-17

Silverman, H. (2004), Andean archaeology. II, Art, landscape and society, Oxford: Blackwell

Thoms, A. V. (2009), 'Rocks of ages: propagation of hot-rock cookery in western North America', Journal of Archaeological Science, 36 (3), 573-91

Uzendoski, M. A. (2004), 'Manioc Beer and Meat: Value, Reproduction and Cosmic Substance Among the Napo Runa of the Ecuadorian Amazon', Journal of the Royal Anthropological Institute, 10, 883-902

van den Bel, M. (2008), 'Late Archaic cooking pits in coastal French Guiana – Results of a preventive archaeological excavation at the junction of the coastal Pleistocene savannahs and the Precrambrian shield', Unpublished excavation report, Institut national de recherches archeologiques preventives (INRAP)

Veiga, J. (2000), 'Cosmologia Kaingang e suas práticas rituais', in XXIV Encontro Anual da ANPOCS, 23 October 2000, Petrópolis, Brazil

Wachnitz, G. A. (1984), Prehistoria Altoparanaense Preguarani, Misiones, Argentina: Imprenta Seyfried

White, J. R. (1980), 'A Closer Look at Clusters', American Antiquity, 45 (1), 66-74


To cite this paper please use the following details: Riris, P. (2010), 'Expanding the Taquara/Itarare Model: Feasting and Social Complexity in the Southern Brazilian Highlands', Reinvention: a Journal of Undergraduate Research, Volume 3, Issue 1, Date accessed [insert date]. If you cite this article or use it in any teaching or other related activities please let us know by e-mailing us at