Background: Globalization and Christianization

From the sixteenth century ad onwards, the expansionist politics of European Empires affected many—if not all—regions of the world, sometimes leading to violent clashes between Indigenous communities and the invading forces. This process was related to a striving to create a world economic system (Wallerstein Reference Wallerstein1974; Reference Wallerstein2004), but it also implied the expansion of a European worldview, including a particular religion: Christianity (Dussel Reference Dussel1992; Mignolo Reference Mignolo2003).

The expansion of Christianity was a multimodal process that included a multitude of aspects of social and religious life, such as practices related to ritual and religious spheres, discourses conforming to the immaterial aspects of Christianity and a set of material items related to these practices and discourses, such as buildings (churches, chapels), clothes (the dress of priests) and symbols (crosses, icons, etc.). However, the expansion and adoption of Christianity were not homogenous around the globe (i.e. Bialecki et al. Reference Bialecki, Haynes and Robbins2008; Brock Reference Brock2000; Patrick Reference Patrick1991). The colonial powers and the cultural logic of different imperial powers differed.

The Spanish occupation of America was firmly based on the desire to expand Christianity, i.e. its Catholic version, with the salvation of Indigenous people (‘savage beings’) as the main focus of the conquest (Duviols Reference Duviols1977). Dussell (Reference Dussel1992) proposed understanding Spanish expansionism as the first ‘Modern Order’ that preceded the Enlightenment and Industrial Revolution. It was based on the centrality of Christianity to order social and religious life, the force of the cross as a symbol, and centred around the idea of progress and the ‘primacy evolution’ of Europe (wording translated here from Spanish). Similarly, Deagan (Reference Deagan, Alcock, D’Altroy, Morrison and Sinopoli2001, 183) has discussed ‘how the goals of evangelization and conversion revenues were equally as compelling as the economic motifs of the venture’. In fact, the Bull of Donation issued by Pope Alexander VI in 1493 recognized a title over the land in America to Spain based on the obligation to evangelize and Christianize Indigenous people (Deagan Reference Deagan, Alcock, D’Altroy, Morrison and Sinopoli2001). Later, the first (1551) and second (1567) Concilium of Lima declared the relevance of eliminating idols and any records or practice of Indigenous religion in the Andes (Castro Reference Castro2009; Duviols Reference Duviols1977).

The Christianization of Australia differs from the South American case. As Schwarz and Dussart (Reference Schwarz and Dussart2010) point out, the presence of Christian missionaries in Australia is a more recent process than in other parts of the world colonized by Europeans. Also, the Christianization of Indigenous communities was not the primary purpose for the conquest of this territory. As a result, many of the interactions between Indigenous groups and missions were based more on a request for land and the development of new economic hubs for trade. The characteristics and intensity of these interactions were heterogeneous among Indigenous communities (Brock Reference Brock2000; Schwarz & Dussart Reference Schwarz and Dussart2010). For Patrick (Reference Patrick1991), the Christianization of Indigenous communities in colonial Australia can be labelled as derivative because it was a ‘by-product’ of the population movement from Europe to Australia. Moreover, colonial Australian Christianity was segmented into subgroups deeply engaged in nationalism, such as English Anglicanism, Irish Catholicism, Scottish Presbyterianism and German Lutheranism (Patrick Reference Patrick1991). Even if the establishment of missions deeply affected Indigenous people’s social and cultural lives, they also sometimes became refugee places where Indigenous people could avoid the worst impact of European colonialism. In remote areas such as Kalumburu in the Kimberley (Crawford Reference Crawford2001) and Oenpelli in western Arnhem Land (May et al. Reference May, Rademaker, Nadjamerrek and Gumurdul2020a), both in northern Australia, different missions and missionaries had a more accepting attitude towards Indigenous people, allowing them to continue to transmit their cultural practices and traditions to the next generation.

Indigenous communities were not passive recipients of European actions. Many examples worldwide show how adopting Christian traditions, materials and symbols from the expansionist colonial powers was a profound political process that implied several layers of meaning (e.g. Nabokov, Reference Nabokov2002; Rademaker Reference Rademaker2018). Brock (Reference Brock2000), for example, has compared the incorporation of Christianity between Canadian and Australian Indigenous communities, recognizing many differences resulting from the particular dynamics of local Indigenous communities and missionaries and the political agency and negotiation within these communities. With the development and momentum of Indigenous community-led research globally, we will surely get more diverse histories about such processes.

Christian-related rock art: an overview

The colonial encounters between Indigenous communities and Western Empires have left many traces in the archaeological record, even in rock art (Fig. 1). In this context, we discuss Christian-related imagery created by Indigenous artists, a seldom-discussed subject within global rock-art research (cf. Berrojalbiz Reference Berrojalbiz2015). Nevertheless, Christian-related imagery appears on all continents and in various rock-art traditions, such as in northern Africa (Kleinitz & Olsson Reference Kleinitz and Olsson2005), southern Africa (Jolly Reference Jolly, Chippindale and Taçon1998), North America (Kirkland & Newcomb Reference Kirkland and Newcomb1967; Rajnovich Reference Rajnovich2002; Turpin Reference Turpin and Hurst Thomas1989; Wilkinson Reference Wilkinson, Moro Abadía and Porr2021), Latin America (Samson et al. Reference Samson, Cooper and Caamaño2016), South America (Hostnig Reference Hostnig2004; Lewis et al. Reference Lewis, Strecker, Taboada and Steinbring1995; Martínez Reference Martínez2009; Perez & Ramirez Reference Perez and Ramirez2020; and below), Australia (see below), northern Europe (Goldhahn Reference Goldhahn, Whitley, Loubser and Whitelaw2020; Hygen & Bengtsson Reference Hygen and Bengtsson2000), western Europe (Van Hoek Reference Van Hoek1993), southern Europe (Anati Reference Anati1994) and Russia (Lahelma Reference Lahelma, Cochrane and Jones2012; Ravdonikas Reference Ravdonikas1936).

Figure 1. An example of Christian-influenced rock art, in this case, highly visible Christian calvaries from Limari Valley, our case study II from north-central Chile. (Photograph: A. Troncoso.)

However, with a few exceptions, Christian-related rock art has rarely been the preliminary focus of any research project, leaving such assemblages heavily under-studied. An apparent reason for this situation is that some early rock-art researchers were influenced by colonial and, to some extent, post-colonial research paradigms aiming to reveal the ‘true’ and ‘authentic’ belief systems of Indigenous people, i.e. as these were before the intervention of colonialism and Christianity. In such a modernistic quest, any sign of influences from outsiders, especially Christian impact on matters related to Indigenous people’s cosmologies and ontologies, were believed and considered to be inauthentic.

A majority of recorded and published Christian-related painted and engraved rock-art motifs come from North and South America (e.g. Berrojalbiz Reference Berrojalbiz2014; Cooper et al. Reference Cooper, Samson and Nieves2016; Gallardo et al. Reference Gallardo, Castro and Miranda1990; Hedden Reference Hedden2002; Lewis et al. Reference Lewis, Strecker, Taboada and Steinbring1995; Martínez Reference Martínez2009; Martínez & Arenas Reference Martínez, Arenas, Sepúlveda, Briones and Chacama2009; Querejazu Reference Querejazu1992; Rajnovich Reference Rajnovich2002; Riris Reference Riris2017; Troncoso et al. Reference Troncoso, Pascual and Moya2019; Turpin 1989; Reference Turpin2010; Turpin & Eling Reference Turpin, Eling and Berrojalbiz2015; Wilkinson Reference Wilkinson, Moro Abadía and Porr2021). This includes both simple and ornate crosses (e.g. Arenas & Odone Reference Arenas and Odone2015; Berrojalbiz Reference Berrojalbiz2014; Riris Reference Riris2017; Turpin & Eling Reference Turpin, Eling and Berrojalbiz2015; Wilkinson Reference Wilkinson, Moro Abadía and Porr2021), human figures dressed in ritual garb with paraphernalia (e.g. Berrojalbiz Reference Berrojalbiz2014; González Reference González2014; Lewis et al. Reference Lewis, Strecker, Taboada and Steinbring1995), introduced musical instruments associated with Christianity, such as the harp (Berrojalbiz Reference Berrojalbiz2014), churches and buildings (e.g. Lewis et al. Reference Lewis, Strecker, Taboada and Steinbring1995; Rivet Reference Rivet2013), as well as images of saints (Gallardo et al.1999 Reference Gallardo, Castro and Miranda1990; Lewis et al. Reference Lewis, Strecker, Taboada and Steinbring1995).

Colonial and Christian imagery in rock art shows heterogeneous situations, suggesting the relevance of local dynamics to understanding the emergence and characteristics of such rock art. Among other things, this is demonstrated in the significant variations in subject matter and the fact that the attributes of visual productions were affected and unfolded differently (i.e. images, themes, techniques, emplacement, etc.). As a result, in some areas, Christian-related rock art has a focus on bravery, aggression and conflict (Turpin & Eling Reference Turpin, Eling and Berrojalbiz2015), while other cases are centred on symbols and saints (Arenas & Odone Reference Arenas and Odone2015; Arenas et al. Reference Arenas, González and Martínez2019; Martínez Reference Martínez2009). In addition, adopting this new set of images could occur in the context of the continuity and maintenance of many traditional visual cultures and the creation of rock art (Berrojalbiz Reference Berrojalbiz2014). The ongoing Samburu rock-art tradition in northern Kenya, for instance, created by lmurran warriors of different age-sets, is biographical, and it is not apparent if and how it conveys any transcendental elements from either Samburu or Christian ontologies (Goldhahn et al. Reference Goldhahn, Labarakwe, Skoglund and Westergren2021). Technological transformations have sometimes been recognized due to the introduction of new tools, mainly iron (Wilkinson Reference Wilkinson, Moro Abadía and Porr2021). Even more, the combination of previous ‘traditional’ images with new visual items proper to Christianity has also been recognized (e.g. González Reference González2014; Martínez Reference Martínez2009; Rajnovich Reference Rajnovich2002; Riris Reference Riris2017; Wilkinson Reference Wilkinson, Moro Abadía and Porr2021). Some ongoing belief systems and practices involving rock art in northern Europe during the 1700s to 1900s were declared as blasphemy, whereafter these were defaced or destroyed by notorious righteous priests (Goldhahn Reference Goldhahn, Whitley, Loubser and Whitelaw2020) or ‘Christianized’ and transformed into new images with Christian connotations (Lahelma Reference Lahelma, Cochrane and Jones2012). In this context, it should also be noted that some Indigenous groups have reinterpreted their traditional rock-art imagery based on introduced Christian beliefs (e.g. Jolly Reference Jolly1996).

Although many discussions have been raised considering the visual transformation of rock art through the study of particular sites and traditions, less attention has been paid to the geographical distribution of Christian-related rock art and how these new marking activities were related to existing places with rock art. Berrojalbiz (Reference Berrojalbiz2014) provides an exciting exception. He found that Christian-related imagery was marginalized within the local landscape in modern México, except at one site, Cueva de las Monas, where Christian and Indigenous imagery were ambiguously blended. Berrojalbiz (Reference Berrojalbiz2014, 90) notes that the new ‘Religion is treated in a peculiar manner’ and differently across the landscape and through time. At older sites, Catholicism is barely represented and is not the central feature, though, in his own words, ‘elements of Indigenous religion are … of greater relevance’.

Several studies in the Andes in South America have shown a heterogeneity of places marked using Christian-related imagery, including pre-Hispanic rock-art sites, chullpas (funerary towers of the Late Intermediate Period) and churches (Arenas & Odone Reference Arenas and Odone2015; Reference Arenas and Odone2016; Hostnig Reference Hostnig2004; Rivet Reference Rivet2013). Other studies have shown how the distribution of colonial rock art was deeply related to the spatial placement of Spanish installations in the landscape and its impact on local Indigenous communities (Arenas et al. Reference Arenas, González and Martínez2019; González Reference González2014).

As we have demonstrated in this brief global review, rock art related to Christianity has been recognized in many parts of the world, and it shows key differences due to the local colonial contexts and processes. At the same time, many studies have focused on the characterization of isolated rock-art sites rather than regional studies discussing the different layers of attributes associated with creating Christian-related images in rock art. Such artworks have often been left without remark and research, which created a huge need for further research.

In the following, and to start to address this situation, we offer a comparative analysis of two case-studies of rock art created by Indigenous communities responding to colonization and Christianization: Christian-related rock art in western Arnhem Land in northern Australia and in the southern Andes in north-central Chile. Generally speaking, in northern Australia, there appears to be a notable absence of explicit Christian imagery in locations where Christian missions have been present, and rock art was continuously created in the surrounding landscapes (Goldhahn & May Reference Goldhahn and May2019). This pattern may, however, be due to a lack of research in many remote areas. In contrast, in north-central Chile, Christian images are frequently documented in rock art, but there is a lack of regional studies showing the main trends and signification of such manifestations (cf. Troncoso Reference Troncoso2022; Troncoso et al. Reference Troncoso, Pascual and Moya2019). Finally, a comparative view can shed light on the engagement of local traditions, vernacular landscapes, Indigenous communities, and the impact of European powers and Christianity.

Study area I: western Arnhem Land, Australia

For Australia, our research focuses on rock art from western Arnhem Land due to the availability of comprehensive data based on over 15 years of fieldwork. The Wellington Range is home to several different Indigenous clan groups, including the Maung, the Traditional Owners of the Namunidjbuk clan estate where most of the rock-art sites we discuss are located (Fig. 2). This estate covers more than 100 square kilometres and has been occupied for over 30,000 years (Wesley et al. Reference Wesley, Litster and O’Connor2018). It is home to a vast array of rock-art styles and traditions (Taçon et al. Reference Taçon, May and Lamilami2020), including many contact period rock paintings and stencils (e.g. Frieman & May Reference Frieman and May2019; May et al. Reference May, Taçon, Paterson and Travers2013; Reference May, Taylor and Frieman2020b; Taçon et al. Reference Taçon, May, Fallon, Travers, Wesley and Lamilami2010; Wesley & Viney Reference Wesley and Viney2016). This large assemblage of rock art allows us to explore the presence and absence of Christian iconography in context with a broad range of introduced subject matter, especially sailing vessels, firearms and feral animals.

Figure 2. Map showing the general location of key places named in the text. (Map: ANU CartoGIS.)

Within this painted landscape, one site stands out as a history book for local Indigenous people—Djulirri. This shelter is home to the largest cluster of introduced subject matter in rock art documented in Arnhem Land and, likely, Australia (Dix Reference Dix2021; May et al. Reference May, Taçon, Wesley and Travers2010; Taçon et al. Reference Taçon, May, Fallon, Travers, Wesley and Lamilami2010; Fig. 2). Other examples mentioned in this paper are drawn from other researchers’ published and unpublished reports from across western Arnhem Land including Kakadu National Park.

Study area II: north-central Chile, southern Andes

Colonial rock art has been widely described in several regions of the southern Andes, and many examples of Christian imagery can be recognized (e.g. Gallardo et al. Reference Gallardo, Castro and Miranda1999; González Reference González2014; Martínez Reference Martínez2009; Martínez & Arenas Reference Martínez, Arenas, Sepúlveda, Briones and Chacama2009). For Martinez (Reference Martínez2009), the imagery of colonial times in the Andes consists of a reduced number of subjects reproduced across many regions, and Christian images are essential in this assemblage. Crosses, calvaries, and priests are the most common motifs, but churches and some Christian writing have also been recorded. Christian crosses would play a central function in this assemblage due to their recurrence in the landscape, but also for their role as a critical symbol of Christianity (Arenas & Odone Reference Arenas and Odone2015; Martínez Reference Martínez2009). A particular image recognized in some regions of the southern Andes corresponds to Santiago (St James), the patron saint of Spain, who provided protection for the conquest of America (Gallardo et al. Reference Gallardo, Castro and Miranda1990; 1999) .

Our case study focuses on rock art from north-central Chile. This region has several east–west hydrological basins from the Andes mountains to the Pacific Ocean. We concentrate our case on recorded rock art in four river basins: Choapa, Combarbalá, Limarí and Elqui, covering c. 400 sq. km (Fig. 3). This territory has been occupied for more than 14,000 years, mainly by hunter-gatherer groups, except for the last 1000 years when agricultural communities have been recognized in the archaeological record. Although the bulk of rock art in this region is dated to pre-Hispanic times, c. 3500–1540 ce, petroglyphs and rock paintings from colonial times have also been recognized (Troncoso Reference Troncoso2022; Troncoso et al. Reference Troncoso, Pascual and Moya2019), including subjects such as horse-riding scenes and different images related to Christianity. Two sets of colonial rock art can be identified: on one side, a hunter-gatherer rock-painting assemblage with some equestrian scenes, although some Christian crosses could have been painted; on the other, agrarian Diaguita communities, which were the leading producers of rock art in late Pre-Hispanic times, created colonial and Christian-related imagery. Our study focuses on this second set.

Figure 3. Map showing north-central Chile (southern Andes) and our second case study area. (Map: A. Troncoso.)

Results I: western Arnhem Land, Australia

The results of our studies in western Arnhem Land are very clear—Christian religious iconography is almost entirely absent. Despite documenting thousands of rock-art sites ranging across hundreds of kilometres, our researchers have found little evidence that Christianity made its way into the rock-art traditions of Arnhem Land. Importantly, this is not the result of rock-art traditions ceasing upon contact with outsiders. On the contrary, rock-art creation appears to have accelerated in this region following contact with outsiders and continued long after missions had been established (e.g. Frieman & May Reference Frieman and May2019; May et al. Reference May, Taylor and Frieman2020b).

Some rare examples of Christian-related rock art can be found at Djulirri, a secluded site in a remote area of the Wellington Ranges (Fig. 4). Even if you have been here many times, the site is hard to identify from a distance, which makes it unlikely that any European would find it without an Indigenous person’s help. The Christian influence consists of a series of English letters made from beeswax. It could be argued that the artist is attempting to write ‘Amen’ (Fig. 5, top left; see also Dix Reference Dix2021, 146). Moreover, a writing slate, presumably used at the Oenpelli Church Missionary Society school (May et al. Reference May, Rademaker, Nadjamerrek and Gumurdul2020a) or Goulburn Island Methodist Mission school (Emilsen Reference Emilsen2016), was also found at this site, suggesting community members were practising new skills being taught by missionaries (Fig. 5, bottom left).

Figure 4. The Djulirri rock-art site. (Photograph: S.K. May.)

Figure 5. (Top left) Beeswax rock art from Djulirri possibly including the word ‘Amen’; (bottom left) Writing slate found at Djulirri; (right) Painting possibly depicting a priest; the bottom close-up photograph has been enhanced with the dStretch lrd-extension. (Photograph: S.K. May.)

The second example at Djulirri depicts a non-Indigenous man, which may relate to mission work in the region. While several paintings represent non-Indigenous people, with the long nose being a key feature and the typical colonial ‘hands-on-hips’ pose another, one painting appears to show a clerical collar. It may represent a missionary (Fig. 5, right). In a 1957 publication by archaeologist Fredrick D. McCarthy, an Axel Poignant photograph of Djulirri features the painting in Figure 4. It is labelled ‘B, paintings of modern subjects have been added to the Arnhem Land galleries. Here are shown ships, an aeroplane and a missionary painted over forty years ago’ (McCarthy Reference McCarthy1957, 184). It is evident that Poignant was led to this interpretation by the local Indigenous people he travelled with in the 1950s; this further enhances our argument that this represents a missionary. Interestingly, the first missionaries arrived on nearby Goulburn Island in 1916 (Emilsen Reference Emilsen2016), which is why Poignant’s dating of the painting in question, having been created ‘over forty years ago’, suggest it relates directly to these new arrivals (Fig. 6). The Methodist minister Reverend James Watson also stopped at the buffalo-shooting camp of Oenpelli on his way to Goulburn Island, with Djulirri situated between the two locations (see Figure 2).

Figure 6. (Top) A church service on South Goulburn Island c. 1917. (Photograph: attributed to Edward Reichenbach: Libraries and Archives NT https://hdl.handle.net/10070/742333); (bottom) Group photo outside the early Oenpelli mission church, c. 1930. (Unknown photographer: Northern Territory Archives Service NTRS694 P1 Box 2 Item 545.)

It should be noted that there were earlier Christian representatives in the area. For instance, Italian Catholic Missionary Angelo Confalonieri was based at Port Essington on the Cobourg Peninsula (near Djulirri) from 1846 to 1848 (Girola & Pizzini Reference Girola, Pizzini and Marsilli2013). He interacted extensively with local Aboriginal people and learned some of the local language, but had little success with conversions to Christianity (Girola & Pizzini Reference Girola, Pizzini and Marsilli2013; Spillett Reference Spillett1972, 147). Given the exact date of the discussed painting, and the fact that Poignant was guided to this site by local informants who had first-hand knowledge about the rock art at the site, we are confident that the discussed images are from the second part of the 1910s.

It is important to note that Dix (Reference Dix2021, 122) recently found that much of the contact period rock art at Djulirri was associated with the mission era, including a flurry of white pigment rock art found clustered on one panel, which he dates from 1918 to the 1940s. These scenes of activity include people with boxing gloves, ceremonial activities, and Arabic numbers (see also May et al. Reference May, Taçon, Wesley and Travers2010). They do not feature any Christian iconography per se, but should be noted in the broader context of mission-related-era rock art.

Another example, which we only know about from an unpublished report, comes from Kakadu National Park. During fieldwork in the 1980s, one of the senior Traditional Owners of this area, Nipper Kapirigi (c. 1910–1987), showed rock-art researcher George Chaloupka a rock painting depicting an Oenpelli (Gunbalanya) missionary (Chaloupka Reference Chaloupka, Chaloupka, Kapirigi, Nayidji and Namingum1985, 41). The painting was in Kapirigi’s Badmardi clan Country in an area known as Baluwurru. The painting, situated on the top of an escarpment in a remote area where few outsiders would pass, was described as a person wearing a European hat and with upraised arms (Chaloupka Reference Chaloupka, Chaloupka, Kapirigi, Nayidji and Namingum1985, 41).

During their conversations, Kapirigi shared stories explaining why Badmardi people may have been interested in the Oenpelli mission: ‘their brother … had been left there to be taught the Christian way’ (Chaloupka et al. Reference Chaloupka, Kapirigi, Nayidji and Namingum1985, 169). The brother’s name was not stated. Chaloupka was told that the missionaries gifted their father Nagulidj with newly introduced objects and tobacco in return for leaving their brother to learn the Christian way. After that, the rest of the family headed back to their Country. When Nangulidj soon thereafter returned to pick up his son, the missionary refused to give him back. In the words of Chaloupka (Reference Chaloupka1983, 91):

While it is not recorded whether the painting at Balawurru directly relates to this episode, there is a good chance it does. If this is the case, the missionary depicted will likely be the Reverend Alfred Dyer (Cole Reference Cole1972; May et al. Reference May, Rademaker, Nadjamerrek and Gumurdul2020a). Such a harsh encounter would have left an enduring legacy, and the rock painting may have been used to retell the story and educate others about their experiences. Without Kapirigi’s first-hand knowledge and Chaloupka’s record-keeping, this painting’s association with Christianity would remain unknown.

Two paintings from western Arnhem Land are said to represent churches that are worth mentioning in this context. The first is from Djuwarr in Balawurru, Kakadu National Park (see Jelínek Reference Jelínek1989, fig. 342). Anthropologist Jan Jelínek (Reference Jelínek1989, 487) noted:

In 1948, Charles Mountford (Reference Mountford1956, 161–2, pl. 43D; Fig. 7) also documented a painting said to be a church at Inagurdurwil near the Oenpelli (Gunbalanya) mission. He states:

Figure 7. The ‘church’ painted at Inagurdurwil. (Mountford Reference Mountford1956, pl. 43D.)

It is unclear if the information given to Jelinek and Mountford was accurate. Both paintings are atypical. Still, given that two examples were recorded at different times and for different paintings, they may be genuine. If this is the case, we have two different artists choosing to paint churches in the rock art of western Arnhem Land. This is especially important given that buildings in rock art are extremely rare in Aboriginal rock art (cf. May et al. Reference May, Taçon, Paterson and Travers2013; Paterson Reference Paterson, Croucher and Weiss2011).

Result II: north-central Chile (southern Andes)

The situation in north-central Chile is the opposite of western Arnhem Land. Colonial rock art and Christian images are easily recognized in the landscape (Figs 1, 8, 9). Furthermore, rock art production decreased during colonial times (Troncoso Reference Troncoso2022; Troncoso et al. Reference Troncoso, Moya and Ivanovic2023). Our work in this area has recorded c. 5000 engraved panels and around 400 of these show contact rock art, including Christian images, corresponding to c. 8 per cent of the registered sites.

Figure 8. (Top left) A Christian cross engraved within a pre-Hispanic double-cross, site La Tranca del Diablo, Limarí Valley; (bottom left) Clothed humans, probably priests, site El Sauce, Limarí Valley; (right) Human anthropomorph with Christian crosses, site El Colihue, Combarbalá Valley. (Photographs: A. Troncoso.)

Figure 9. (Left) Christian inscription at Quebrada Pinto 2, Elqui Valley; (right) The coexistence of pre-Hispanic motifs and a Christian cross at Tranca del Diablo, Limari Valley. (Photographs: A. Troncoso.)

Three types of Christian imagery can be recognized. The most popular refer to symbols proper to Christianity, corresponding to crosses and calvaries. While crosses are basically drawn as two lines crossed, calvaries show a more complex expression, including triangular or rectangular bases and different kinds of crosses (Figs 1, 8–10). Although both motifs can be dated to colonial times, Arenas and Odone’s (Reference Arenas and Odone2015) research suggests that calvaries are more recent and visually aggressive on the panels than crosses (Fig. 1, cf. Figs 8–10).

Figure 10. Different relations in Christian rock art from north-central Chile. (Left) The coexistence of pre-Hispanic motifs and a Christian calvary at Quebrada Lucumán, Choapa Valley; (right) superposition of a Christian calvary on pre-Hispanic motifs at Panulcillo 2, Limari Valley. (Photographs: A. Troncoso.)

In a few cases, crosses took advantage of pre-Hispanic motifs (Fig. 8, top left). In these cases, a retouch was made to remark on the idea of a Christian cross. Churches and images of Saint Santiago are absent in this study area.

Anthropomorphic images interpreted as priests are scarce. These images correspond to human beings using long robes (Fig. 8, bottom left), but in any case, we can recognize a composition suggesting a pilgrimage or something similar. However, a particular scene occurs at the El Colihue site, where a human being handles a couple of Christian crosses (Fig. 8, right). This scene is also interesting because the human figure is framed among two zoomorphic figures, and the crosses face the animals, suggesting some protection. Similarly, a linear appendix conforming to an inverted Christian cross emerges from the groin of the human figure as a kind of phallus (Fig. 8, right). Is it a moral discourse relating sexuality with sin?

No evidence of Christian writing has been documented. Although some inscriptions have been found on many sites, they mostly correspond to marks made by Creole people during the nineteenth century. An exception is an inscription about a hex to someone who has made an incantation. This inscription is associated with Christian crosses (Fig. 9, left). However, the Creole people could have made this inscription at the end of colonial times.

A particular trend of Christian imagery in the region is the scarcity of superposition over previous images (>1 per cent). On the contrary, they frequently co-exist on the rocks with earlier pre-Hispanic engravings in a non-disruptive relation (diachronic solidarity, following Arenas & Odone Reference Arenas and Odone2015). As a result of this situation, we found the bulk of colonial and Christian rock art on places marked by Diaguita communities in the pre-Hispanic era, suggesting the perdurance of a spatial tradition of marking places and rocks (Figs 9, 10). This continuity could also explain some visual attributes of Christian rock art in this area.

As we have seen, while crosses are the most frequent image, scenes such as pilgrimages or priests are scarce. This situation is in tune with Diaguita pre-Hispanic rock-art tradition because that set of rock art is characterized by the absence of scenes, a low number of anthropomorphic images, but a high frequency of non-figurative rock art (Troncoso Reference Troncoso2022). In fact, crosses within circles and double-crosses are motifs often documented in Diaguita pre-Hispanic rock art. In this context, we can recognize a reduction in rock-art production during colonial times in the area (Troncoso Reference Troncoso2022; Troncoso et al. Reference Troncoso, Moya and Ivanovic2023). A continuity of marking places and rock occurs, introducing new visual elements within a local visual tradition (Fig. 9, right).

The spatial distribution of colonial and Christian rock art shows that the majority of the sites are located in ravines far away from the central fluvial valleys where human settlements were concentrated in the late pre-Hispanic period and colonial times (Troncoso et al. Reference Troncoso, Moya and Ivanovic2023).

Finally, although Christian images in the area are adjusted to a visual and spatial pattern, we can recognize some differences in local production, suggesting that the colonial and Christian processes unfolded heterogeneously in different times and spaces (cf. Fig. 10). This is particularly the case when it comes to depictions of calvaries and crosses. Calvaries were incorporated in rocks marked previously, but in non-aggressive ways (no superimposition and no visual primacy). However, in some places, we found large and highly visible calvaries, which dominate previous rock markings (Fig. 1). Christian crosses also show a significant variability within the region. This heterogeneity is not only visual but also spatial. While crosses co-exist visually with previous images in some sites, in others, we find superimpositions that obliterate pre-Hispanic motifs (cf. Fig. 10). Although we cannot explain the dynamics of this heterogeneity, they would probably be related to the different relations established between local communities and the Spanish conquerors and the distance to their settlements. A preliminary result suggests that highly visible calvaries and crosses obliterating previous images are distributed in lower parts of the valleys, places more impacted by Spanish conquest.

The presence of crosses in rock-art sites along the Andes was initially interpreted as a political strategy used by Spanish conquerors to Christianize local sacred places (see discussion in Arenas & Odone Reference Arenas and Odone2015; Martinez Reference Martínez2009; Martínez & Arenas Reference Martínez, Arenas, Sepúlveda, Briones and Chacama2009). However, recent research suggests that the presence of crosses and calvaries in rock-art sites is the result of more complicated historical and colonial processes, including the appropriation of Christian symbols by Indigenous communities. Our findings are in tune with this hypothesis, though they show a heterogeneity of visual and spatial strategies related to the use of crosses and calvaries in rock-art production, suggesting, in some cases, the appropriation of these images and their incorporation into Indigenous visual discourses; but other situations exist, where such images were engraved on the rocks marking a disruption and a break with previous visual discourses.