4. Concluding Discussion

For the first time archaeologists possess an accurate record of the engravings and surface of Avencal 1. PTM was instrumental in developing an improved understanding of how the engravings were produced. Through careful examination of the captured datasets we have built up an impression of the site that permits us to begin to access the technical aspects of how the engravings were produced (Sections 3.3 and 3.4) and to augment the record of its extent and content. Certain contradictions within and between the four panels defined by previous authors (Piazza 1966; Rohr 1971; Piazza and Prous 1977; Prous 1994) have, as a result, been resolved. An example of the first type is the relative topology of engravings within panels (see Figure 4 and Figure 5). Secondly, the limitations on recording important, culturally-significant detail by tracing have been bypassed by the PTM dataset. Taking these points into account, it is clear that PTM can specifically enhance our understanding of rock art sites in general. As the data capture methodology has proven useful for documenting previously-unknown engravings, the technique has clear implications both for documentation and conservation of exposed and vulnerable heritage. At this point, we consider it of paramount importance to show and make available a new and more complete record of Avencal 1 to an international audience. In this regard, the method and resulting record is likely to benefit from the incorporation of additional survey techniques, such as structure from motion multi-image photogrammetry (Chandler et al. 2005; De Reu et al. 2013), to provide the process of data capture with metric accuracy and quantitative information for interpretation.

The site and its neighbours are crucial components of our understanding of southern proto-Jê cosmology and landscape-level patterning in the region (Corteletti 2013). Although so-called 'geometric' engravings are problematic to interpret from an anthropological perspective (see Bednarik et al. 1990; Chippindale 2001), PTM permits us to shift the emphasis of analysis away from classification towards technique (sensu Pelegrin 2009). Examining the PTM data reveals to us a chaîne opératoire of techniques that were employed at Avencal 1. Engravings were seldom made on unprepared surfaces. It appears instead that most, including the largest panels, were pecked out ahead of time and the prepared surface polished further in some cases, such as the iconic 'masks' in Panel 1A. We noted in section 3.2 that the sandstone of Urubici would be a brighter colour when recently exposed and stand in contrast to the brown-grey weathering visible on the surface today. This may reflect a deliberate choice by the makers of Avencal 1 to make the engravings more striking under certain specific lighting conditions. In other words, a variety of techniques of production, including incision, polishing, and pecking were used to further the ends of the producers of Avencal 1.

A significant finding of our study is the order in which incisions were made, which we suggest may provide information on the cultural norms and technical skill that guided its production. The act of manufacture, as a series of techniques and gestures, has a socio-technological significance that goes beyond the panels as a 'finished' assemblage of engravings (Pfaffenberger 1992, 505; Méndez Melgar 2008, 146-7). Systematic documentation of these patterns can reveal the range and variation in their expression within the technological framework of southern proto-Jê rock art practices. These factors have been termed the 'grammar' of rock art (Chippindale 1992), which can afford access to previously undetected patterns both within and between different traditions. Our initial findings serve as a useful starting point for additional PTM data collection with a more inclusive sample of southern proto-Jê rock art sites, from both highland and coastal contexts. This will provide a fruitful basis advancing the study of southern proto-Jê rock art beyond the typological focus which has endured since the 1950s (Funari 1999, 27). Sites in the southern Brazilian highlands were traditionally grouped within the Southern and Geometric rock art 'traditions' (Gaspar 2003). This kind of classification is usually contradictory and inconclusive, and as a perfect example of this situation it is important to note that Avencal 1 incorporates elements of both putative traditions. Comparing the findings here with patterns observed in sites from different contexts will allow for stronger conclusions to be made on how they were produced. Studying pre-Columbian art in Brazil can lead to this corpus generating a greater impact on wider disciplinary debates regarding the technologies of petroglyph creation.

View Panel 5 with WebRTIViewer

Our focus on technique should not be viewed as an attempt to diminish the symbolic or cosmological significance of the assemblage as a whole. At the same time, the anthropological literature on Amerindian worldviews reaffirm that imagery, fauna and features of the 'natural' environment have well-documented agentic roles in indigenous societies (Viveiros de Castro 1998; Lagrou 2009). The elements of Avencal 1 are a part of a visual representational system which explains how members of a specific society thought about themselves and the world around them. This system is connected to the social organisation and cosmology of these groups. Da Silva (2001) explains the visual representation system of Kaingang people by identifying the representational role of imagery in pottery decoration, rock art sites and in the body paint of modern groups. Each moiety is distinguished by a set of rules, including their own types of motifs. These moieties constitute a dual cosmological scheme and are contrasting, opposing and complementary. For example: the Kamé moiety are known by slim, tall and open signs (téi), while the Kainru moiety are known by circular, low and closed motifs (ror). We believe this is reinforced at Avencal 1 by the Jê imagery recorded on the wall (Da Silva 2001). Furthermore, our record allows us to go deeper than previous research and hypothesise that an engraving of Jê shamanic animals such as the jaguar in Panel 6C (bookmark 2), can be identified with the Kamé moiety of modern Kaingang people (see Crépeau 2002, 119). Following this idea, one of the 'masks' (Figure 13) may be a representation of the Kamé: the zigzag face paint is identified (by Da Silva 2001) as a ra ionior motif, signifying membership of this moiety. The Kainru moiety is also represented in Avencal 1 site by closed shapes such as arrangements of dots, triangles with dots and the mosaics of squares and rhombuses (Panel 6C, bookmark 1; Panel 1A, bookmark 5; see also Figure 6). When téi and ror signs are found together, as in Panels 6, 5 and 1A, they are called ra iãnhiá signs and represent an inter-moiety alliance (visually signifying an event – marriage, for example – Da Silva, 2001, 192). Taken together, these factors reflect the role of Avencal 1 assemblage as a material expression of the symbolic dual organisation of Jê society (see Crépeau 2005; Iriarte et al. 2013).

We have demonstrated that certain sections of 'blank' rock face actually bear badly preserved petroglyphs eroded by exposure to the elements. While this discovery is notable in the context of this fieldwork and the application of RTI, it is important to note that the perception of 'emptiness' by modern observers is in itself a cultural choice by the producers of the petroglyphs. It is therefore not only an issue of taphonomy, but may be related to the landscape and geomorphology, acting as a social agent in itself (Santos-Granero 2009).

Finally, we note that some engravings are located only a few centimetres above ground level at Avencal 1. Coupled with the discovery of new engravings, this may indicate that additional ones may exist below the present surface. Debitage from surface preparation in a datable archaeological context can supply a more precise date for the site. Given the range of production techniques in the rock art that has been inferred from the PTM data, the potential recovery of tools deposited at the site could provide an independent line of evidence to support our interpretations (Fiore 1999, 279; Méndez Melgar 2008). These factors are the subject of on-going efforts.


Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

University of York legal statements

File last updated: Tue Feb 3 2015