Previous Next

An Archaeology of Early Christianity in Vanuatu

1

Theories, Methods, and Materials

Thus fell JOHN WILLIAMS – ‘the Apostle of Polynesia’ – a man who had won the most splendid reputation perhaps of any missionary since the days of the great Apostle of the Gentiles … died a martyr on that ever memorable day, NOVEMBER TWENTIETH, 1839 … and now in glory shines as the stars for ever and ever (Gordon, ed. 1863: 101).

The death of John Williams and James Harris on the beach at Dillon’s Bay, Erromango, in 1839 (see Chapter 2) was a pivotal moment in the history of the New Hebrides (now Vanuatu). This event crystallised the reputation of indigenous Melanesians as hostile and inherently dangerous. The New Hebrides were characterised as a place of fear, black magic, and cannibalism (e.g. American Sunday School Union 1844; Copeland 1878). At the same time, this moment provided a rallying point for missionaries determined to ‘bring light to dark isles’ in the region. Thus began a century or so of mission labour in the New Hebrides, resulting in the widespread conversion of Melanesian people to Christianity (indigenous ni-Vanuatu identified as 86 per cent Christian in the most recent census; Vanuatu National Statistics Office 2009: 25).

It is an archaeological truism that many phenomena treated as historical events are in fact better understood as moments within long-term cultural processes. The work of Christian conversion in the New Hebrides is no exception. Conversion was part of a long-term process of accommodation, adaptation, and appropriation of European things and beliefs by Melanesian people. The process was shaped by Melanesian people in ways that allowed for a great deal of flexibility in responses to the emergence of modernity and colonialism in their islands. Religious change in these islands was not simply a result of the efforts of European missionaries, but also of the indigenous Melanesians who converted, often assisting the mission endeavour as teachers (see Miller 1978). Equally a part of this story are those people who resisted the spread of mission influence, as well as those who sought to avoid contact altogether, holding tight to kastom (referring to a variety of indigenous traditional beliefs and practices).

This book demonstrates that the social process of religious change in the southern New Hebrides had an inherently material basis. The focus will be the islands of Tanna and Erromango, though reference will be made to the neighbouring islands of Aneityum, Futuna, and Aniwa (Figure 1.1), especially with regards to museum collections from this period (see Chapter 5). Broad regional or inter-site patterns are relevant for explaining the overall process of conversion, while local or intra-site variability reflects the importance of rooting interpretations to specific places where conversion happened.

The Presbyterian missionaries who established the earliest permanent footholds and largest groups of converts in these islands saw spiritual conversion and transformation of the material conditions of everyday life as inherently intertwined. It was not enough for indigenous Melanesian people to learn to pray, read the Scriptures, and attend church under missionary instruction. They were also expected, to greater or lesser extents depending on the missionary, to alter their agricultural practices (towards a monocropping system that would have been more legible in capitalist terms); their habits of dress (not just to cover more skin than was customary, but also to set aside indigenous materials in favour of imported European ones); and their daily rhythms of labour, devotion, and domestic life (particularly to closer approximate expected gender roles; see Jolly and Macintyre, eds 1989; Choi and Jolly, eds 2014). Further, native converts were expected to cast off all signs of ‘idolatry’ and other dangerous anachronisms that could interfere with the process of true conversion (see Keane 2007).

Figure 1.1 Map of southern Vanuatu, including islands and communities mentioned in the text.

Source: James Flexner

The resident Presbyterian missionaries in the New Hebrides were meant to model these changes for Melanesian people, with mission houses and missionary family life providing the civilised exemplar to which Melanesian people should aspire (though the extent to which missionaries expected native people to achieve civilisation was debatable; see Lydon 2009: 154–155). At the same time, this was not a one-sided exchange. Missionaries and their accompanying family members were deeply dependent on local and non-local resources for their survival. In the remote islands of the New Hebrides, local resources became necessary for the long-term survival of missionary families. This included material things, particularly food staples and building materials, but also the less tangible social networks of indigenous people who might be convinced to convert. Without Melanesian ‘heathens’, the missionary had no reason to travel to these islands. Conversion was more than a simple individual decision. Wider networks of people, places, and objects are implicated in the conversion process (Keane 2007; see also Latour 2005).

Besides simple reliance on indigenous people, missionaries themselves became localised, partaking of an increasingly Melanesian set of social and material references, especially in the later decades of the 19th century when longer-term missions were established. This was part of the wider process of indigenising Christianity in the New Hebrides (Flexner and Spriggs 2015). Mission houses, once showcases of civilisation, became increasingly filled with indigenous things. Further, missionary collection of Melanesian ‘curiosities’ greatly amplified the social networks of such objects, eventually resulting in their dispersal to urban centres throughout Europe and European settler societies. Fascination with the exotic was not simply about Melanesian interest in the ‘advanced’ industrial material culture of Europeans. The ‘primitive’ things of Melanesians held an allure of their own (Thomas 1991).

Conversion had a material basis because individual or group decisions to adopt Christianity in the New Hebrides were not simply a matter of internal spiritual awakening. They resulted from a network of linked things, people, social relationships, and material encounters. To examine any one individual’s decision to convert is to pull on a web of social obligations, gifting relationships, ecological variables, historical trajectories, and supernatural beliefs, all mediated through things of various sorts. This is what is meant by the idea that Christian conversion had a material basis: you can’t separate the ‘religious’ nature of conversion from the wider social networks in which conversions took place, and these networks by necessity include objects. As Fowles (2013) suggests, any attempt to separate ‘religious change’ from ‘culture change’ more broadly is part of a modernist ideological framework in which ‘purified’ forms of social institutions (politics, economics, religion) are properly ordered when they do not mix. Ironically, attempts at purification if anything intensify the proliferation of hybrid or mixed categories (see also Latour 1993). Religious change is culture change, and vice versa.

Networks composed of missionaries and Melanesian converts, non-converts, backsliders, and uncontacted communities included the objects that Melanesians made, and those that missionaries brought with them. Archaeologists increasingly recognise that objects themselves should be included as agents within such human networks (e.g. Byrne et al. 2011; see also Latour 1987). To the networks of people and objects should be added the primary importance of place in Melanesian social networks (Rodman 1992). Places simultaneously contain the things (people and objects) of social networks, and are themselves things within the social network. In the analysis below, the importance of place (or in Bislama, Vanuatu pidgin, ples) will become clear in the ways that local people remember and engage with the locations where the work of conversion happened (see also Flexner 2014c). Mission sites, rather than being part of foreign, colonial ‘waet man’ (white man) history, are integrated into the pantheon of traditional kastom sites, just as Christianity is integrated into what it means to be ni-Vanuatu today (see also Flexner and Spriggs 2015).

How and Why Did People Convert?

The history of the Presbyterian missions in the New Hebrides is not a story of ‘triumph of the Gospels’ as one later missionary account suggests in its title (Paton 1903). The one early missionary who had great success was John Geddie, who settled on the island of Aneityum in 1848 (Patterson 1882). As a memorial pulpit in his main station at Anelcauhat reads, ‘When he arrived there were no Christians, and when he departed, there were no heathen’. Yet even Geddie’s success involved a long-term process of material exchange, local demographic collapse, and gradual integration of the church into local social networks (Spriggs 1985, 2007). On the islands of Tanna and Erromango, where this study is focused, missionary progress was much slower. Attempts at starting missions on both islands repeatedly failed over the course of the 1840s through the 1860s. The first long-term stations were only established in the 1870s, and even then it was decades before the numbers of converts began to outnumber those who held onto traditional religious practices.

In other cases, there were mass conversions, as on Aniwa where John G. Paton not only survived on the tabu ground of the sea snake god, but also dug a freshwater well. This convinced the local chief of the potency of the foreign god, and where the chief converted, so did his followers (Paton 1907 Vol. 2: 130–196). So it appears that people in the southern New Hebrides held out against converting for a certain amount of time, and then converted; sometimes more slowly and gradually, as happened generally on Tanna and Erromango, or sometimes very rapidly, as happened on Aniwa. As we develop an account of the material networks that drove the conversion process, we have to start by asking: why did people choose to convert at all? What was it about the new religion of the missionaries that led people to change what they were doing eventually when, at first, people simply ignored, drove off, or even killed the would-be agents of change?

In attacking this problem, there is a related, and fundamental problem that needs to be addressed first. The name of that problem is ‘religion’, at least as the term is typically used in modern secular scholarship as well as popular rhetoric. In his account of the archaeology of Southwestern North American Pueblo ‘doings’, Fowles (2013) provides a compelling critique of the ways that scholars have tended to analyse religion as a human social institution. In orthodox models, modern subjects separate religion from other realms of behaviour (particularly political and economic behaviours). In analysing pre-modern societies and subjects, religion increasingly becomes blurred with the appropriately purified, and in modern societies separate institutions of politics and economics. In reality, however, there is immense overlap with other categories, and the secularisation narrative is a ‘just-so’ story that masks the complexity of both ‘traditional’ and ‘modern’ religious life and practice (Fowles 2013: 1–37). The tendency of ‘religion’ to blend with other aspects of life, from subsistence to chiefly politics, likewise holds true for the New Hebrides. The indigenous term kastom (Bislama, ‘custom’) is probably more appropriate to discussing local experiences of religious change. The introduction of Christianity did not change this, as the foreign faith was simply integrated into kastom.

If religion permeates all aspects of pre-modern societies, then what is the analytical use of religion if it can’t be purified from other realms? Situations of colonial religious conversion highlight the instability of these kinds of categories. Historical archaeology shows the ways that religious conversion was part of a broader process of ‘culture change’ in the colonial era, particularly in North America (e.g. Arendt 2010; Lightfoot 2005; Panich and Schneider, eds 2014) and Australasia (e.g. Lydon 2009; Middleton 2008; Morrison et al. 2015). The emerging perspective eschews any simple transition of religious beliefs from ‘indigenous’ to ‘introduced’, as aspects of traditional belief and practice are always included as the spiritual realm is transformed. Likewise, there is no situation in which religion changes separately from other realms such as politics or economics. Christian conversion is inevitably accompanied by transformations in everyday life including aspects that leave material traces, such as dress, food, and domestic architecture. I recognise these ‘hybrid’ (sensu Latour 1993) realities, both in terms of the overlapping of cultural categories and the lack of a stable ‘purity’ of transformation in those categories. However, I frame the argument here in terms of a focus on ‘religious change’, specifically in terms of the umbrella term of kastom that I see as central to Melanesian identities, practices, and beliefs spanning the pre-Christian era through the present and into the future. In kastom, the supernatural cannot be separated from the everyday and mundane, so any changes in kastom are by necessity ‘religious’ changes.

As Keane (2007: 85) points out, one of the major problems missionaries can encounter when doing the work of conversion is determining what native practices are harmless ‘traditions’ and thus acceptable to be kept, and what are dangerous fetishisms or idolatries that need to be rooted out. It will be seen that in Melanesian kastom, it was nearly impossible for missionaries to separate out the religious practices that needed to be eliminated from the broader realm of traditional practice. This led missionaries to adopt a variety of strategies, from integration of customary beliefs into Christian practice, to attempts to violently crack down on native practices (the latter approach almost always failed spectacularly). In fact, crackdowns if anything solidified kastom as a resilient aspect of indigeneity in the colonial era. For example, the so-called ‘Tanna Law’ enforced by early 20th-century Presbyterian missionaries and their converts outlawed traditional kava drinking, dancing, and other practices. This repression of kastom resulted in widespread abandonment of church practices, a resurgence of traditional beliefs, and the emergence of the new John Frum ‘cargo cult’ on the island (see Bonnemaison 1994: 201–219; Lindstrom 1993).

Yet kastom could be remarkably flexible, employed and enacted in a variety of settings incorporating the political, economic, and religious, for different purposes in different places and different historical moments (Jolly 1994; Lindstrom 1982). Attempts to separate religion from kastom are doomed to failure because such categorical divisions do not exist within kastom. As Fowles (2013: 102) notes, Pueblo people distinguish not between sacred and secular, but ‘between the sacred and the sacred’. While indigenous people in Vanuatu recognise a category of ‘church’ (jioj) in the present, within kastom engagements with the sacred, including Christian beliefs, are generally integrated into everyday life and habits.

Melanesian people had their own reasons for engaging with Christianity, or not. It is of course difficult to speak to the motivations of any particular person or group in the past, especially when writing across the boundaries of tradition and culture. Not growing up with kastom, I am still something of a neophyte, so my ideas on the topic must be taken with a grain of salt against possible alternative interpretations by people in the Melanesian communities where I worked on this project (see Preface). What I can speak to authoritatively is the archaeological evidence that speaks to the materiality (or materialisation; Bell and Geismar 2009) of conversion processes. Regardless of any individual reason (or rationalisation) for converting, and how specifically people integrated Christianity into beliefs and rhetoric about kastom, the archaeological evidence indicates that the process involved transformations of material things.

To say that conversion was a material process is not an end-point to the discussion. It offers a challenge to interpret why people in southern Vanuatu chose to indigenise Christianity. A number of alternatives present themselves, which will be explored in the chapters to follow. One possibility that has been offered for eventual widespread conversion is a demographic shift (Bedford, pers. comm.). Massive population declines following European contacts in Oceania are well established, and southern Vanuatu was no exception to this pattern (Adams 1984; Humphreys 1926; McArthur 1981; Speiser 1922; Spriggs 2007). So one possible argument is that as populations were severely decimated, people moved to coastal areas, and people whose villages had essentially been wiped out aggregated around the mission stations. Having seen most of their friends and relatives perish within a few generations, people were less resistant to the possibilities offered by the new religion of the missionaries.

Another possibility would be that the new economic opportunities offered by the mission drew people in at least initially. Fascination with European goods and the ability to engage with new trading networks led people to begin interacting with missionaries, eventually resulting in their integration into communities of Christian converts. This possibility has been suggested at the Te Puna Mission in New Zealand (Middleton 2006), though it is clear that Maori were interested in trade on their terms, including trading for things that were generally discouraged by the Church Missionary Society (CMS), such as alcohol and firearms. These two possibilities, demographic change and economic opportunity, are driven largely by external phenomena (introduced diseases and foreign trade goods).

While such factors certainly had a role to play in driving the conversion process, we also have to consider what internal factors may have shaped religious change on Erromango and Tanna. One of the clear patterns in mission historiography is that without the consent of local chiefs, missionaries had no chance of settling or surviving in Melanesian communities. So why would a chief allow a missionary to settle in his territory? The trading hypothesis offered above is an obvious answer, though it actually appears to be only a small part of the story, especially when weighed against archaeological evidence for European trade goods in native settlements (Chapter 3; Flexner et al. 2016c). Perhaps more valuable were the advantages chiefs gained over other local chiefs. The prestige of having foreign visitors in one’s territory may have made missionary settlement appear worthwhile for some chiefs, especially if that settlement was assumed by Melanesians to be temporary in nature. It also appears that some chiefs sought to use missionaries to support their side or assist in brokering peace during conflicts over land and chiefly titles.

Finally, by the end of the 19th century, mission presence may have been seen as a buffer against a colonial regime that was by turns neglectful and deeply abusive of the indigenous population. This is somewhat ironic, as on Tanna in the 1860s John G. Paton, one of the unsuccessful missionaries on that island, caused several villages in the Port Resolution area to be bombarded by the British Navy in retaliation for the missionaries being driven away, an event known as the ‘Curaçoa affair’ (Adams 1984: 150–167). By the time the Anglo-French ‘Condominium’ was established in 1906, missionaries could offer some protection by formalising land claims for indigenous communities (of course only for those who had converted), and mediating with government forces. This is one of many tensions in the missionaries’ position in Melanesian society, however, as they were both forces for the incursion of state power (Flexner 2014b), and one of the early options for indigenous people hoping to deflect or resist that power.

Ultimately, conversion was a complex process shaped by multiple causal factors. All of the above were important to the indigenisation of Christianity in the southern New Hebrides. One final point about the economic basis for conversion: just as conversion could hardly be considered a ‘triumph of the gospels’, it was also not simply a result of Melanesians becoming good capitalists through an interest in consumer goods. Melanesians are well known for having a ‘partible’ view of material culture (Gosden 2004; Strathern 1990). They did not divide things and people into separate categories (nor really did or do Europeans, including Presbyterian missionaries, but that is another point to be raised later). Thus interest in foreign things was not based on their ‘value’ from a capitalist perspective. Rather, things were valued for the networks in which they were enmeshed. Missionary objects were fascinating for their connections to foreign people, places, ancestors, and spirits. They provided an opportunity for Melanesians to amplify the social-material networks of which they were a part.

Why Archaeology?

That conversion was a material process is abundantly evident in the mission houses, church buildings, and assemblages of artefacts found in archaeological sites on Tanna and Erromango (Chapters 2–4). That the process inflected the material culture of both sides, European and Melanesian, is evident in the relatively small number of indigenous artefacts found on mission sites, and more dramatically in the large assemblages of native things found in Western museum collections (Chapter 5). Because material things were so important to the conversion process in the New Hebrides, archaeology provides the most analytically powerful toolkit for understanding the theoretical problems outlined above.

The approach used here is archaeological in two senses. First, it is concerned with the emergence of present patterns by searching for their origins or roots in the relatively deep past (Sahlins 2004: 43; see also Olivier 2011). Second, the basis for arguing about the how and why of conversion on Erromango and Tanna is based primarily on the archaeological record, which is then informed by documentary evidence, rather than the other way around. The archaeological record includes the expected array of old buildings, stone ruins, stratified deposits, and artefacts. Oral traditions and documentary evidence are also understood archaeologically (see Beaudry, ed. 1988; Beck and Somerville 2005; David et al. 2012). Oral histories can be systematically mapped onto the landscape in the same way as stone terraces or scatters of potsherds. By the same logic, ‘natural’ features such as stones or trees related to kastom stories are an important part of the record of mapped archaeological sites (see Flexner 2014c).

There is an abundant written record for the Presbyterian New Hebrides missions, including primary sources such as missionary correspondence, land records, and published biographical works (e.g. Gordon, ed. 1863; Gunn 1914; Paton 1903; Paton 1907; Patterson 1864, 1882; Watt 1896). These works not only have to do with the work of conversion, but often include some of the earliest detailed ethnographic observations of the New Hebrides (Gray 1892; Inglis 1854; Robertson 1902; Turner 1861). While missionary descriptions of kastom must be taken with a grain of salt, they also contain significant grains of truth. These accounts are extremely important for understanding New Hebrides cultures in the 19th century. Missionary ethnographies must be read critically, but they can provide valuable information both about traditional cultures and the prejudices and worldviews of the missionaries themselves (Douglas 2001). Published missionary works are supplemented with other documentary evidence such as drawings, photograph albums, and magic lantern slides. There are valuable historical accounts of mission history in the New Hebrides that rely primarily on such documents that inform my scholarship (e.g. Adams 1984, 1998; Lubcke 2009; Miller 1978, 1981). Documents such as missionary correspondence provide a record of what people thought or said they were doing, but they are simultaneously ‘things’ in the sense that certain documents could be used to act on the world (as with land records), or to tell stories about exotic people and places (as with photograph albums).

These documentary and oral historical things are just as much a part of the archaeological record. There are more prosaic reasons for relying on an archaeological approach to better understand the conversion process on Tanna and Erromango. Colonialism was an inherently material process, in which people, domesticated and commensal organisms, objects, and ideas encircled the globe. As they moved around, these things formed new networks, at times transcending or even dissolving or hybridising cultural differences, at others creating marked resistance movements that threw such differences into starker relief. Much that happened as part of this process was simply never written down, or what was written reflected the cultural values and biases of those doing the writing as much as anything else.

Historical archaeologists have long asserted the importance of materials-based approaches to literate societies (Andrén 1998). The history of the discipline in many ways follows the broader trajectory of archaeology from antiquarian concerns (Harrington 1955), to ‘processual’ functionalist accounts of past societies (South 1978), structuralist approaches concerned with identifying meaning in the archaeological record (Deetz 1996), and more ‘post-processual’ approaches concerned with identifying conflict, resistance, and power in colonial interactions (Hall 2000). Colonialism has become one of the main concerns of modern historical archaeology, in Anglophone literature largely focused on encounters and interactions between Europeans and indigenous peoples throughout the world over the last five centuries (e.g. Cusick, ed. 1998; Gosden 2004; Kirch 1992; Lawrence and Davies 2011; Lightfoot 2005; Murray, ed. 2004; Mills 2002; Rubertone 2000; Silliman 2005).

One of the major problems that has arisen from historical archaeologies of colonialism is the issue of indigenous resistance to European colonial incursions. Recently, scholarship has converged on the issue of persistence and continuity in indigenous practices, including in part a sustained critique of issues of authenticity as they pertain to native communities (Flexner 2014a; Lydon and Burns 2010; Panich 2013; Silliman 2009). At the same time, there is an ample literature on the creativity of indigenous peoples in response to colonialism, particularly emphasising the ‘hybrid’ forms that emerged from indigenous appropriation, adaptation, imitation, or remixing of foreign things (e.g. Card, ed. 2013; Harrison 2006; Torrence and Clarke, eds 2000). As Silliman (2014) notes, probably the best approaches to hybridity follow the idea that the ‘pure’ categories (such as native/introduced) don’t really exist in practice. What we are in fact observing is the proliferation of hybrids that emerges from the act of creating categories (Latour 1993; see also Flexner 2014a).

It is easily acknowledged that indigenous people did not disappear, despite 19th-century predictions to the contrary. That indigenous practices survived, adapted, and transformed through five centuries of colonial violence is indisputable from the available archaeological evidence, not to mention the resilient indigenous communities that continue to survive in the contemporary world. The questions that remain are highly relevant but contentious: What does it mean to be indigenous (or not)? Who gets to make authoritative statements about indigeneity? Is it enough to simply assert that the world’s native peoples ‘have history’ or ‘have agency’?

These are not problems that will be easily solved by scratching away at the dirt and using broken bits and pieces from past lives to determine degrees of change and continuity in the archaeological record, tied up as it is with living peoples’ identities, lives, politics, and so on. On the other hand, archaeologists absolutely can contribute important evidence to these discussions since often the important data points in the colonial period cannot be gleaned from the written record alone. Assertions about ‘invented’ traditions or ‘inauthentic’ claims to indigenous rights for political or economic reasons can be thoroughly undermined in the face of the material traces of native resilience and adaptation during the colonial period.

While it is not a major point in this book, the authenticity question relates to colonial processes of religious change. Westerners may consider conversion to Christianity as an element of inauthenticity in indigenous cultures. At the same time, experience of the mission setting by one’s ancestors in some indigenous communities can be an integral component to claims of authenticity (as is sometimes true for Aboriginal Australians, McNiven and Russell 2005: 226–227; see also Lydon and Burns 2010). Questions of indigenous authenticity tend to be most marked in European settler societies (United States, Canada, Australia, New Zealand), where the settler population has to deal with the uncomfortable reality of a dispossessed population of colonised people who often continue to survive as the poorest inhabitants of the lands from which they were often illegally and violently removed. An ideology of terra nullius, in which the indigenous population never ‘really’ owned the land, and disappeared anyway, provides the ideological prop for maintaining such a status quo (e.g. Gosden 2004: 114–152).

Archaeologies of Christian missions have important implications for this discussion. If it was in the process of becoming Christian that indigenous people lost their authentic religion, among other practices, this should be visible in the archaeological record. Yet what historical archaeologists find repeatedly is the integration and subsumption of Christian missions and missionaries into indigenous landscapes and communities, not the other way around (e.g. Ash et al. 2010; Middleton 2003; Panich and Schneider, eds 2014). In the Bay of Islands, New Zealand, missionaries at Hohi and later Te Puna were reliant on Maori interest in exchange relationships for their acceptance into the community. Further, the presentation of Maori children for education on the mission was used as a way to draw missionaries and the European goods they could access into Maori networks of reciprocal social obligations. Producing Christian Maori was a secondary outcome of that process (Middleton 2003, 2006, 2008; Smith 2014; Smith et al. 2012, 2014).

Missions in Aboriginal Australia tended to be more centralised and institutional, at least physically (Middleton 2010; Sutton 2003; for a critique of this see Morrison et al. 2015: 98–99). However, these missions were no less shaped by the indigenous landscapes where they were set, a pattern visible throughout Aboriginal Australia and Australasia more broadly (Lydon and Ash 2010). Aboriginal people were drawn to missions for complex reasons, having to do with new opportunities for trade, expanding social networks, dispute resolution, and escape from settler violence on the colonial frontier (e.g. Lydon 2009; Morrison et al. 2015: 100–101). What is clear from both Australia and New Zealand is that to enter the mission was by no means to abandon one’s indigeneity (see also Lydon and Burns 2010; McNiven and Russell 2005).

In Vanuatu, questions of authenticity do not arise in the same ways, since the majority of people are indigenous Melanesians, who have largely retained land rights and ownership (though there is a complicated history to this; Van Trease 1987). However, the significant fact is that Christian conversion in no way undermines the extent to which ni-Vanuatu can be considered indigenous. Rather, Christianity and kastom do and did go hand-in-hand for Melanesian people who converted, and their descendants who largely continue to practice Christianity today (Flexner and Spriggs 2015). This project is a result of indigenous interest in Christian archaeological sites on Tanna and Erromango, which are not seen as separate from any other category of kastom places. There is no reason to assume that the indigenous relationship to Christianity would be any different in the European settler societies. The either/or Christian/Native division represents yet another artificial binary that masks a more complex hybrid reality.

In addition to questions of indigeneity, archaeologies of colonialism necessitate an exploration of the identities of the colonisers. European Christianity involved a set of embodied material practices that had to be enacted every day by its practitioners, especially in colonial settings where such practices were thrown into relief against those of non-Europeans or non-Christians (e.g. Chenoweth 2009, 2012). Missionaries would have identified primarily as European (British, and particularly Scottish), even if they were born in one of the settler societies. While the majority of missionaries came from what is now Canada, the area was in the 19th century the Lower Provinces of British North America. Scottishness may have been a significant component of people’s identities (a pattern visible in the present, though the history of this is complicated; Vance 2005, 2011). In entering the mission field, to what extent did missionaries hold on to their European-derived identities? What aspects of the mission encounter changed their habits and worldviews, and what aspects of continuity in material practice are visible? In such an analysis, the object is to offer the same treatment to Melanesians and missionaries. In the case of Europeans, the authenticity of their identities is rarely questioned, and the goal here is not to say necessarily that the missionaries became anything else (the identity and political issues at stake would be different anyway). Rather, I am interested in the possibility of applying the same metrics of continuity and change to the colonisers that were traditionally used in studying the colonised (e.g. Ramenofsky 1998). Archaeology provides the best means to understand this dynamic.

Outline of Mission History in the Southern New Hebrides

A brief summary of the history of Presbyterian missions on Tanna and Erromango is provided to give readers a basic temporal framework and cast of characters for the archaeological analyses that follow. More specific, detailed accounts drawn from the documentary sources left behind by particular missionaries will be given in the relevant chapters to follow. The London Missionary Society (LMS) was the earliest organisation active in the southern New Hebrides. Mission history in the New Hebrides began on a tragic note with the death of John Williams and James Harris at Dillon’s Bay on 20 November 1839. The importance of this event for shaping the history of Christianity in the New Hebrides cannot be understated, especially when considering the long-term patterns of resistance, conversion, and guilt as experienced by people on Erromango (Chapter 2; see also Mayer et al. 2013). On this same trip, Williams established colonies of Samoan teachers, themselves recent converts to Christianity, on Aneityum and Tanna (Liua‘ana 1996). Many of these died no less tragic, but less widely acknowledged martyrs’ deaths on these islands.

Following from Williams and Harris, there was basically sustained missionary contact with the southern New Hebrides over the ensuing century (Table 1.1). The LMS continued to send Polynesian teachers in the early 1840s (Liua‘ana 1996). There was also a short-lived mission station on Tanna, inhabited by the LMS missionaries Turner and Nisbet in 1842, though this was abandoned after less than one year. While the LMS would remain somewhat active in the New Hebrides, the Presbyterian Church became the main Protestant missionary supporter in the New Hebrides (Miller 1978), alongside the Anglican Melanesian Mission, which was active in the northern and eastern islands of the group (Armstrong 1900; Hilliard 1978).

Table 1.1 European missionaries on Tanna and Erromango, 1839–1920.

Island

Missionary Name

Year(s)

Place

Erromango

John Williams

1839

Dillon’s Bay (1 day)

Erromango

George and Ellen Gordon

1857–1861

Dillon’s Bay

Erromango

Hugh and Christina Robertson

1872–1904

Dillon’s Bay

Erromango

James Gordon

1864–1872

Dillon’s Bay/Potnuma (post-1868)

Tanna

John Williams

1839

Port Resolution (1 day)

Tanna

Turner and Nisbet

1842

Port Resolution (7 months)

Tanna

Thomas Neilson

1868–1882

Port Resolution

Tanna

Joseph Copeland

1858–1862

Port Resolution

Tanna

John G. Paton

1858–1862 [1903]

Port Resolution, Kwamera [later Aniwa]

Tanna

John and Mary Matheson

1858–1862

Kwamera

Tanna

William and Agnes Watt

1869–1894

Kwamera

Tanna

Frank Paton

1896–1902

Lenakel

Tanna

John C. Nicholson

1903–1917

Lenakel

Tanna

William and Elizabeth Gray

1882–1894

Waisisi

The first significant long-term mission settlement in the New Hebrides was established on Aneityum in 1848 by Rev. John Geddie, with the support of the Presbyterian Church of the Lower Provinces of British North America (now the Presbyterian Church of Canada). Geddie was a resident missionary on Aneityum for over four decades, and his was without question the most successful early mission in the New Hebrides, in the sense that he was able to achieve widespread local conversion to Christianity during a long tenure on the island (Patterson 1882). The mission station at Anelcauhat boasted a large stone masonry house and church (Jones 2013), and remained the main stronghold for the Prebyterian Church in the New Hebrides until the 1890s. During his time working on the island, Geddie not only converted a significant percentage of the population, he could also claim to have planted ‘a school in every district’ of the island, providing the Aneityum Mission with the strongest early network of mission stations and native teachers (Spriggs 1985). Arguably, Geddie’s success stemmed largely from his willingness to learn Aneityumese language and kastom, and to adapt mission practice appropriately (see Flexner and Spriggs in press). As suggested above, the notion of missionary ‘success’ should be tempered against the more hybrid realities of religious and cultural change in the Melanesian context.

In the middle of the 1850s, the Presbyterian Church sought to expand upon its success on Aneityum, establishing George and Ellen Gordon as missionaries on Erromango in 1857 (Gordon, ed. 1863), and Reverends John G. Paton, John Matheson, and Samuel Johnston with their wives on Tanna in 1858 (Paton 1907 Vol. 1; Patterson 1864). The Canadian Church continued to be the main source of missionaries and support for these missions, though Paton was from Scotland. These were to be relatively short-lived endeavours. By 1860, Johnston was dead of malaria and the Tanna Mission had to be temporarily abandoned because of increasingly hostile local communities. After a brief resettlement in 1861, the Tanna Mission was completely abandoned in February 1862 as the Mathesons and Paton were unable to successfully re-establish footholds on the island (Adams 1984: 144–145; see also Chapter 3). On Erromango in 1861, George and Ellen Gordon were killed at Dillon’s Bay by warriors from a neighbouring village. George’s brother James Gordon followed on Erromango, arriving at Dillon’s Bay in 1865. By 1868, James Gordon had resettled to Potnuma, near the large village of Port Narvin on the east side of the island. In 1872, James would also die a martyr’s death at the hands of local people (Robertson 1902; see also Chapter 2).

Early missionary work did not follow an easy road to conversion. These short-lived missions did have some factors in common. Missionaries generally had a poor understanding of local customs, had to learn local languages from scratch, and often suffered because of their lack of knowledge. Paton and colleagues were chased off Tanna because the mission had inadvertently placed itself in the middle of a local land dispute. James Gordon may have been killed for serving poisonous fish at a local feast. The 1850s–1870s on Tanna and Erromango could be considered an ‘adaptive’ period, when missionaries were still learning how to survive in the island environment (Birmingham and Jeans 1983). That said, later missionaries would credit the struggles of these earlier missionaries to win converts during this period with their ultimate successes.

The missionaries’ fortunes would change by the early 1870s. This period saw the establishment of long-term missions, with the Watts at south Tanna (Watt 1896) and the Robertsons at Dillon’s Bay (Robertson 1902). Both families would stay through the turn of the century, successfully converting large numbers of local people. Further, the archaeological record shows that during this period missionaries began to invest in larger, more elaborate, more permanent infrastructure in their settlements. In 1882, Rev. William Gray, an Australian missionary, joined the Watts on Tanna, settling at Waisisi.

During this time the mission also expanded to the Polynesian Outliers of Futuna and Aniwa (Capell 1958). Rev. William Gunn, a Scot and medical missionary for the group, would establish his primary settlement on Futuna in 1883 (Gunn 1914). John G. Paton returned to the New Hebrides after his failure on Tanna, eventually being sent by the Church to Aniwa (Paton 1907). The story of Paton’s success on Aniwa is relevant to a pattern in missionary settlement in indigenous landscapes. As often happened, Paton was allowed by the Aniwans to settle in a poisoned, tabu area inhabited by the local sea snake deity. After Paton not only survived in this spiritually dangerous place, but successfully dug a freshwater well, he achieved a mass conversion of the populace on Aniwa. This episode is notable in that missionary successes could come about because of resilience and survival in spiritually dangerous places. In other cases, as will be seen, missionary illnesses or deaths may also have been attributed to the inability of their deity to protect them from local spirits, inhibiting the conversion process.

By the 1890s, the Presbyterian Church had permanent mission stations established on Tanna and Erromango, and had expanded into the central and northern islands of the New Hebrides. This expansion was matched by an expansion of connections between the mission stations and global networks of consumer materials. In the west Tanna Mission, established at Lenakel by John G. Paton’s son Frank in 1896 (Paton 1903), the initial station looked much like newly established stations elsewhere: limited infrastructure and buildings primarily of local materials. However, within two decades, the Lenakel Mission would expand to include a large prefabricated church manufactured as a kit in Sydney, a large hospital, and ‘model village’ of native converts (Chapter 4; Flexner et al. 2015). The west Tanna Mission, then, represents an ultimate stage in mission endeavours in the New Hebrides, which was bolstered by the ability to draw widely on the global networks of industrialised Europe and European settler societies. Buoyed by an increasing sense of triumph, the missionaries overreached, attempting on Tanna to eliminate traditional ceremonies, dancing, and kava drinking completely. The result was a backlash in which the missions and model villages were largely abandoned, people retreated to kastom, and rumours of a new ‘cargo cult’ began to spread, especially in the southeastern part of the island (Bonnemaison 1994: 226; see also Guiart 1956: 163; Lindstrom 1993).

Presbyterians were not the only missionaries in the New Hebrides. The Presbyterian Church worked closely with the Anglican Melanesian Mission (Armstrong 1900; Hilliard 1978), sharing resources, particularly mission ships, and amicably dividing geographic responsibilities, with the Anglicans proselytising to the islands in the north and east, and the Presbyterians to the south and west (Miller 1978: 142–146). In contrast, the Protestant missionaries were deeply distrustful of, even antagonistic towards, the Catholic Church and its missionaries, though they were not really active in the New Hebrides until after 1885 (Miller 1978: 139–141). The Presbyterians not only distrusted Roman Catholics for their loyalty to the Pope, but probably worse, they were closely aligned with the French, who were seen as competing with British imperial ambitions in the region. By the early 1900s, various other evangelical sects began working in the New Hebrides, notably the Church of Jesus Christ of Latter Day Saints (LDS, or Mormon Church), and the Seventh Day Adventists (Bonnemaison 1994: 67–80).

Today, it can be said that Vanuatu is, for the most part, a Christian society. That said, Christianity did not replace kastom, rather it was simply incorporated into Melanesian tradition in Vanuatu, as happened throughout the region (see Taylor 2010; Tomlinson and McDougall, eds 2012). What follows is an archaeological account of the process of conversion over the first eight or nine decades of mission work in the New Hebrides. The ‘success’ of Christianity should not be taken as a foregone conclusion. Throughout the history of conversion local people resisted missionary settlement, missions failed more or less catastrophically, apparently sincere converts defected, and missionaries perished as a result of their efforts. In many cases, communities did not convert, and today there remains a significant population of non-Christians in Vanuatu who still maintain the beliefs of their ancestors, or have developed new ones in response to colonialism, such as the John Frum ‘cargo cult’ of Tanna (Lindstrom 1993).

Neither can we say that the work of conversion was a completed project. Throughout the archaeology of early Christianity in the New Hebrides, we find evidence of what missionaries would have seen as dangerous mixing of heathen and Christian religious beliefs and practices. Indeed, many of the more divisive debates that took place within the Church during the missionary era revolved around the question of which Melanesian practices could be kept, and which had to be eliminated among convert communities (see also Keane 2007). In many cases, it has been argued that apparent converts simply hung the trappings of Christianity on extant beliefs and practices (e.g. Bonnemaison 1994: 211). Archaeology has an important role to play in illuminating how this situation came to be. As I will conclude, the discoveries of this project hold important implications for how the question of religious change in Vanuatu might be considered in centuries to come.

Archaeological Methodology

Mission archaeology projects in Oceania have generally focused on a single mission station, which is then sometimes tied into a wider landscape (e.g. Dalley and Memmott 2010; Ireland 2010; Lydon 2009; Middleton 2008; Smith et al. 2012). One of the innovations of this project was that archaeological survey covered multiple sites, allowing for documentation of variability across time and space (Flexner 2013; see also Ash et al. 2010; Morrison et al. 2010), as well as presumably cultural variability (the research area covers four different language areas). Work was spread out among seven mission landscapes on two islands: Port Narvin, Cook’s Bay, and Dillon’s Bay on Erromango; Lenakel, Waisisi, Port Resolution, and Kwamera on Tanna (Figure 1.1). In many of these landscapes, there are multiple mission sites. For example, both Kwamera and Dillon’s Bay have an early mission abandoned in the 1860s, and a later, longer-duration mission spanning the period from the 1870s to the early 1900s.

While mission sites were the focus, one of the goals of research was to situate the missions in terms of the indigenous landscapes in which they were embedded (Appendix A). A corollary to this involved recording local social memories, both of mission sites and older sites associated with kastom stories. In some cases, as in the villages at Kwaraka and Anuikaraka on south Tanna (Chapter 3; Flexner et al. 2016c), the kastom sites were immediately recognisable as archaeological, with the landscape consisting of stone mounds, walls, and enclosures. In other cases, interpretation of archaeological features such as rock art sites on east Erromango could be augmented with local oral traditions. Landscape features that would be considered ‘natural’, such as trees and stones, were documented where they were associated with particular social memories. These ‘memory places’ represented the largest category of features documented in archaeological survey (Table 1.2). Where a group of memory places were linked to a common story, the resulting assemblage is referred to as an ‘event landscape’. Event landscapes offered a useful way of capturing the local sense of ‘historicities’ (Ballard 2014) regarding both mission-related and indigenous happenings in the recent past (Flexner 2014c). As I’ve noted elsewhere, ‘Event landscapes are places where social memory is both constructed and performed as people follow in the footsteps of historical characters, spirits of the ancestors, or supernatural beings’ (Flexner 2014c: 8).

Recording of indigenous perspectives of history as written on the landscape augmented the more traditional archaeological techniques that formed the bulk of fieldwork for this project. A few technical details about field methods should be noted when interpreting the data in the following chapters. Initial surveys documented site locations, architectural forms, surface artefacts, and other relevant details in and around mission sites. Mission houses were identified early on in the project as significant places of contact and interaction between missionaries and Melanesians (Flexner 2013: 16–20). A program of test excavations covering areas ranging in size from 1m x 1m to 4m x 4m was carried out at five of the mission houses (George Gordon House and Robertson House at Dillon’s Bay; James Gordon House at Potnuma; Matheson House/Imua at Kwamera; Watt House at Kwamera), and at the native villages Kwaraka and Anuikaraka. Excavations were carried out stratigraphically, following natural layers where possible, and all sediments were sieved using 0.5cm mesh. Sediment samples were taken at each of the excavated sites, and all samples and artefacts are curated at the Vanuatu National Museum in Port Vila.

Recovered artefacts were analysed in detail by material class, and the results were entered into a relational database. These objects and their spatial contexts provide critical data about the everyday interactions with things that shaped the conversion process in the New Hebrides Presbyterian missions. In addition to the excavated material, which was almost entirely European (or European-derived, as in the case of locally produced lime mortar), a survey of museum assemblages was carried out at museums in Australia, New Zealand, Canada, and Europe. These materials provide a sample of artefacts from the other side of mission exchanges, things that were given to missionaries within the context of ongoing conversion and indigenisation of Christianity.

Table 1.2 Types of landscape features recorded during archaeological survey.

Feature Type

Cook’s Bay, Erromango

Dillon’s Bay, Erromango

Eastern Erromango

Kwamera, Tanna

Kwaraka, Tanna

Lenakel, Tanna

Port Narvin, Erromango

Port Resolution, Tanna

Potnuma, Erromango

Waisisi, Tanna

Grand Total

Burial Ground

 

 

 

 

 

 

 

 

 

1

1

Cave

 

1

1

2

Church

 

1

1

2

2

1

1

8

Hospital

 

5

5

House

 

1

1

2

Land Marker

 

1

1

Memorial

 

1

2

2

4

1

1

11

Memory Place

 

7

4

1

2

1

2

1

3

21

Mission House

1

2

1

1

2

1

1

9

Mission Infrastructure

 

4

1

2

3

2

12

Nakamal

 

1

1

3

5

Native Village

 

2

2

Other

 

1

1

Petroglyph Site

 

1

2

3

School

 

1

1

Tabu Site

 

3

2

5

10

Grand Total

1

16

9

6

7

15

4

14

4

18

94

Outline of the Book

This book is primarily divided by place. The chapters on each of the islands where the project took place will involve a discussion of kastom as it likely appeared in the decades leading up to the arrival of the missionaries, based on missionary accounts, early ethnographies, and contemporary social memories. Of course these interpretations of kastom are limited by the source material, and so should be read with an understanding that alternate perspectives are possible and indeed reasonable, particularly among ni-Vanuatu. One chapter will focus on analysing and interpreting fieldwork data from Erromango (Chapter 2). Fieldwork data from Tanna is divided into two chapters, covering earlier and later periods of mission life on the island (Chapters 3 and 4). Discussions will include detailed outlines of event landscapes, survey results, and stratigraphic excavations. In discussing the excavations, I avoid artificially dividing the ‘finds’ from excavated ‘sites’ (Lucas 2001: 75). Discussions of artefact assemblages will take place alongside the narrative about places where the objects were found, with some comparative discussion as warranted, and more general synthesis in the concluding chapter.

After outlining some of the relevant details of kastom on Erromango, Chapter 2 deals head-on with the event landscapes that record the deaths of early missionaries on the island. These dramatic events were sedimented into local places through social memories, and have shaped Erromangan experiences through the present, even to the point where 21st-century Erromangans have held reconciliation ceremonies with the descendants of martyred missionaries (Mayer et al. 2013; Naupa, ed. 2011: 90–102). The primary goal of archaeological research on Erromango’s colonial period is to move away from the rhetoric of the ‘martyr isle’, and to begin to understand the patterns of everyday life that shaped the history of interaction with outsiders on the island. If missionaries died on the island, these were events that account for a miniscule percentage of the actual span of time covered between when John Williams and James Harris were killed in 1839, and the Robertsons left after a long missionary career on the island in the early 1900s. Landscape archaeology on the island shows some of the ways that the missionary settlement reflects the changing relationships between outsiders and Erromangans, while the recovered artefacts provide data reflecting the material interactions that took place on the island.

Chapters 3 and 4 focus on Tanna. While Tanna did not see the dramatic and widely publicised martyrdoms of Erromango, the mission likewise saw early challenges, catastrophic failures, and eventually the establishment of long-term stations and fairly widespread conversion. Here, too, excavations of mission house sites provide an illuminating picture of life in the mission, and interactions with the Melanesian community over a period covering nine decades. In the Kwamera area of south Tanna, we get some of the most useful insights into everyday life in the native villages around the mission sites at Kwaraka and Anuikaraka, settlements occupied for a time period predating missionary arrival through the post-WWII era, after which the Presbyterian missionaries had mostly departed the southern New Hebrides. Significantly, occupation overlaps with the mission houses at Imua and Kwamera, just 2km away. At the coastal stations on east (Waisisi) and west (Lenakel) Tanna, we see a later period of mission history where missionaries were increasingly able to draw on global capitalism to acquire materials for their work (Chapter 4). Lenakel represents the last major Presbyterian Mission station established on Tanna, a massive infrastructural investment pulling on truly global resources, a far cry from the initial beachheads established on these islands a half-century earlier.

If fieldwork on Tanna and Erromango tells us about the material that was brought to the islands, museum assemblages can tell us about the things that travelled from the islands, collected by missionaries as ‘curiosities’ during their tenure in the New Hebrides. Missionary collections will be the focus of Chapter 5. Collections both augment what is found archaeologically, as they often contain indigenous material culture unlikely to survive locally on archaeological sites, and provide valuable data about indigenous agency in the curiosities trade (Flexner 2016b). These objects (wooden clubs, barkcloth, bows and arrows, grass skirts) blur the line between ethnographic and archaeological material, being contemporaneous with the earthenware, nails, and shells excavated on mission sites. They are just as much artefacts as the fragments found in the ground. Just as stratigraphic location provides a context for excavated material, the museum collections provide a context for understanding the social lives of missionary collected objects in the Western world.

Finally, Chapter 6 provides a synthetic overview of the big ideas that these materials speak to. Questions of identity, materiality, and religious change arise when studying the process of Christian conversion in the New Hebrides from an archaeological standpoint. Things writ large, from individual ceramic dishes to houses, to places marked on the landscape either physically or in social memory, all played a role in the indigenisation of Christianity on Tanna and Erromango. One of the major points here is that the category of ‘religion’ could not be separated from kastom. If kastom is taken as an all-encompassing category for social life in Melanesia, a ‘total social fact’ (Mauss 1990 [1954]), then it follows that Christianity as a new kind of engagement with the supernatural, a new ideology, and a new way of engaging with everyday life, would simply be subsumed into and adapted to fit kastom, rather than the other way around. It is almost certainly not what the Protestant missionaries would have intended (Keane 2007). But, it provides a remarkable case study of the ability of indigenous people to take things from the colonial world and make them their own as they adapted to life in an age of colonialism.

It might be said that the outline of theory and method presented above overstates the case. Because material things are deemed important, I will present an analysis of materials relating to Christian conversion, and then conclude that objects mattered in the New Hebrides missions. But this would be to miss the point of such an analysis. Physicists studying the universe believe physics is fundamental to understanding its origins and the way it works, so they use the methods and theories of physics to answer their questions. As archaeologists, we should not be shy about asserting the primacy of archaeological approaches when addressing archaeological questions. The point isn’t simply to assert that ‘things matter’, but to try to use a specific case study to understand why they matter, how they inflected patterns of everyday life in situations of contact and colonialism. Of course, many complementary approaches to this topic would be possible. Besides the historical studies mentioned above, the question of how and why Melanesians converted to Christianity could be beneficially addressed from the perspectives of sociology, psychology, ecology or evolutionary biology. Each approach has its benefits and limitations. The chapters that follow represent an initial account of systematic archaeological study of mission sites on Tanna and Erromango and the material processes at work that shaped the experience of religious conversion both among Presbyterian missionaries, and among the Melanesian communities in which they settled. This material is essential for understanding the 19th-century Melanesian universe of the southern New Hebrides.


Previous Next