Previous Next

An Archaeology of Early Christianity in Vanuatu

3

Tanna: In the Shadow of the Volcano, 1839–1868

The most striking feature of Tanna is Mt Yasur, the active volcano rising 360m above sea level on the island’s southeastern flank. Tanna is the second-largest island in TAFEA Province, covering just over 550km2. It is also the most densely populated island in the province, and today is an important political and economic centre of southern Vanuatu. Rock formations on Tanna are no more than three million years old. The island is formed of three overlapping volcanic massifs, with the oldest formations in the north and eastern parts of the island, which are late Pliocene in age. Late Pleistocene formations in the southwest include Mt Tukosmeru, the highest peak on Tanna, and the most recent volcanic activity, which is ongoing, is in the southeastern part of the island. The island has been tilting to the east throughout its formation, exposing limestone terraces along the west coast of Tanna (Carney and Macfarlane 1971; Depledge 1994).

Knowledge of Tannese prehistory is very limited at this point. In part, this is because the active volcanism on the island has deeply buried many archaeological sites, especially in the southeast. More importantly, there has simply not been much archaeological fieldwork done on the island. While Lapita ceramic sites representing initial settlement almost certainly exist on the island, none have been discovered at this point. What we do know about Tannese archaeology derives primarily from fieldwork carried out by Richard and Mary Shutler in 1964 (Shutler and Shutler 1966; Shutler et al. 2002). The Shutlers worked primarily on the west coast of the island, where they excavated two rockshelters and recorded several middens and a village site. The Shutlers’ work, which also extended to Erromango, Futuna, and Aneityum, resulted in the first radiocarbon dates from southern Vanuatu (Shutler 1973), though there is reason to be somewhat sceptical of these dates. A modest sample of artefacts were recovered on Tanna, both from excavations and given to the Shutlers as gifts. The gifted material included stone adzes, a fragment of a kawas (stone throwing club), and stone discs.

Excavations in two rockshelter sites, which the Shutlers called TaRS1 and TaRS3, recovered similar stone artefacts, as well as a variety of shell objects, including flaked Tridacna (giant clam), a Conus adze, and Conus beads. Human burials were uncovered in both caves, though only the TaRS1 burials were associated with grave goods. Radiocarbon dates from TaRS1 were primarily from the last 1,000 years, with a piece of wood charcoal from the deepest levels (2–2.3m below surface) dating to 2370+/-90BP (Shutler et al. 2002: 192–195). Re-analysis of the materials from these sites using modern techniques, particularly for dating, would be highly beneficial for drawing any further conclusions about these sequences. After the Shutlers, the next archaeological excavations on Tanna were carried out for this project, which included several relevant discoveries for Tannese prehistory.

European knowledge of Tanna begins, as with so many islands in the South Pacific, with Cook. After departing from Erromango in August 1774 (see Chapter 2), Cook continued south, drawn by the glow of Yasur, which was clearly visible from the sea by night. Cook and his crew spent about two weeks in Tanna, primarily in the Port Resolution area (Beaglehole, ed. 1969: 483–509; a more detailed discussion of this encounter appears below in the section on Port Resolution). Relations were mostly peaceful, and certainly more productive than on Erromango. Cook was able to exchange some things with local people, though there were clear tensions and eventually the killing of a Tannese warrior by one of the ship’s marines necessitated the Resolution’s departure. It is from this encounter that we get the name Tanna, which in the language of the Port Resolution area simply means ‘ground’ or ‘earth’. Johann Reinhold Forster, the ship’s naturalist, had pointed to the ground assuming he would be told the name of the island, but quite reasonably, Tannese people chose to interpret his gesture literally. A local name for the island is Ipare, and people from neighbouring islands had names for Tanna as well, for example Ekiamo in Futunese (Beaglehole, ed. 1969: 489; see also Jolly 2009). Nonetheless, the name Tanna stuck through the colonial era and into the present. Such cultural misunderstandings became fairly standard in European encounters with the island.

After Cook there were fleeting contacts with explorers, such as the Russian Golovnin in 1809, or passing adventurers such as Dillon in 1825 (Adams 1984: 32–33). The rate of European contacts increased beginning in the late 1820s with the expansion of the sandalwood trade (Shineberg 1967). In the 1840s there was a sandalwood trader on Tanna named Paddon who was married to a Tannese woman, with whom he had four children. After sandalwood came the labour trade, which drew thousands of Tannese, mostly younger men, away from the island to work on plantations in Australia, Fiji, and New Caledonia (Docker 1970; Shineberg 1999). Sandalwooding and the labour trade greatly increased Tannese interest in imported goods such as metal tools and trade cloth (Bonnemaison 1994: 39–43).

Missionary contacts began with the settlement of Polynesian teachers on the island in 1839. The Camden stopped at Tanna before taking John Williams and James Harris on their fateful trip to Erromango. The first European missionaries were Turner and Nisbet from the London Missionary Society, who arrived at Port Resolution in 1842. They stayed for seven months, eventually departing because of native indifference and increasing local tensions. The Patons, Mathesons, and Johnstons were Presbyterian missionaries on south Tanna (Port Resolution and Kwamera) from 1858–1862, though this mission collapsed catastrophically as relationships with local people deteriorated following a series of epidemics. Long-term missionary contacts began in the 1870s in the south, though many parts of the island resisted Christian incursions of any sort well into the 20th century (Adams 1984; Liua‘ana 1996; Miller 1981: 20–45, 1986: 246–425). From 1905–1925, there was a period of ‘Tanna Law’, when Christian courts were established and traditional activities such as kava drinking and dancing were sometimes violently repressed. If anything, this experiment with Christian theocracy in the New Hebrides served to entrench kastom on the island (Bonnemaison 1994; Guiart 1956).

It appears that Tanna did not suffer the same level of demographic decline from introduced diseases as occurred on neighbouring Erromango and Aneityum (Bonnemaison 1994: 44; Humphreys 1926: 1; cf. Spriggs 2007). Further research is of course needed to refine this observation. There certainly would have been some population decrease, especially after particularly virulent outbreaks of measles and dysentery from the 1840s through the 1860s (Adams 1984). However, Tanna seems to have remained relatively densely populated, and the population began to recover earlier than on neighbouring islands. Tannese people would have explained the apparent resilience of their population in contrast to those of their neighbours in magical terms, and this may relate to the longer period of resistance to missionaries and European colonisers more generally.

Ironically, that Tanna had the largest population in the southern New Hebrides made it an important administrative centre during the era of formal colonialism in the islands (Guiart 1956: 130–150). In 1887, a Joint Anglo-French Naval Commission was established in the New Hebrides, partly to protect Europeans from Melanesian ‘aggressions’. More importantly, the Joint Naval Commission was meant to stabilise relationships between French settlers from New Caledonia and British settlers from Australia. Both groups were agitating for exclusive annexation of the New Hebrides, which would almost certainly have resulted in conflict. The end result was the formation of the Anglo-French ‘Condominium’, which served (mostly dysfunctionally) as the colonial government from 1906 until independence in 1980. One of the main functions of the Condominium was the adjudication of land disputes, primarily among European settlers, but also where native people had reasons to complain, though in the latter case there was little the government could or would do (Bonnemaison 1994: 83–87; Jacomb 1914; Rodman 2001: 21–50; Van Trease 1987).

Tanna was an important location of resistance to colonial policies, both governmental and missionary. It likewise became an important site in the independence movement. In both cases, the strength of resistance was drawn from the strength of kastom on the island. Traditional beliefs and practices structured and became emblematic of Tannese attachment to ples (place) as they defined life on their islands in the colonial and postcolonial environment. In more recent decades, kastom continues to evolve on the island, with the presence of more conservative traditions existing alongside more innovative phenomena such as the John Frum ‘cargo cult’. The resistance, resilience, and innovation inherent in Tannese kastom has in some cases been said to take its current form in part because of the excesses of the missionary era, especially in the first decades of the 1900s (see Bonnemaison 1994; Guiart 1956; Lindstrom 1982, 1993). That said, there is nothing foreign about kastom on Tanna, which owes its form entirely to the knowledge, traditions, and ingenuity of Tannese people.

Kastom on Tanna

Knowledge about Tannese kastom is a contested field, both academically and in vernacular terms. Tanna has been the most intensely studied island in southern Vanuatu, and so there is more scholarly detail to discuss. At the same time, a habit of Tannese ‘secrecy’, where power is derived from letting people know that one holds significant knowledge without fully divulging the details of that knowledge, contributes to both outsider and insider understandings of kastom (Adams 1987; Lindstrom 1982, 1990). There are enough agreed-upon ethnographic ‘facts’ that we can lay out a general model of Tannese kastom in the century or so leading up to missionary presence on the island. As always, though, this knowledge should be taken with the caveat that it represents only a rough framework to understand some of the patterns that emerged in the colonial era. It also needs to be noted that much of the early knowledge about Tannese people was filtered through a missionary lens (e.g. Gray 1892; Turner 1861: 69–94; Watt 1895), which continued to influence the work of later ethnographic observers (e.g. Humphreys 1926: 71–72). Further complicating the matter, there are five main languages on Tanna, and possibly a dozen dialects (Nehrbass 2012). Kastom terms vary across the different areas of the island. Here I will use the Nafé language of the Kwamera area of southwest Tanna, with references to other languages where these are commonly used in other sources or geographically relevant to the area being discussed.

As on the neighbouring islands, land divisions on Tanna were called ‘canoes’ (netata in Nafé, niko in Lenakel; cf. Aneityum nelcau, Erromango lo). As we saw, Erromango was divided into six lo, each of which had a paramount chief (see Chapter 2). Aneityum was divided into approximately 55 districts, which were distributed among seven nelcau, each of which likewise constituted a ‘chiefdom’ (Spriggs 1985, 1986). Tanna, in contrast, contained at least 116 netata (Figure 3.1), without any higher-level chiefs. This has led Tannese society to be described as ‘atomistic’ in comparison to its more hierarchical neighbours (Brunton 1979). Lack of hierarchy on Tanna should not be equated with a lack of social complexity. If anything, the heterarchical nature of Tannese chiefship increases the complexity of day-to-day social and political interactions (e.g. Guiart 1956: 107–115).

Figure 3.1 Map of netata (land divisions) on Tanna, including sites and locations mentioned in the text. Note that division boundaries are approximate only and meant to be for general reference.

Source: James Flexner

European visitors to the island, from Cook to early ethnographers to contemporary anthropologists, have remarked on the limited nature of chiefly power, with decision-making largely a matter of establishing consensus among groups of men who hold chiefly titles (e.g. Beaglehole, ed. 1969: 507–508; Bonnemaison 1994: 152; Humphreys 1926: 35–36). One mid-20th-century survey on the island counted 601 ‘chiefs’ against a total population of 6,937. This comes out to one chief for every 11–12 individuals (Guiart 1956: 9). Such a profusion of chiefs, assuming that the ratio hadn’t changed much for at least 200 years or so, meant any major decisions involved lengthy discussions and negotiations among chiefs from neighbouring villages and netata, not to mention the various ritual specialists involved. Chiefly discussions would have revolved around preparing, planting, and harvesting gardens, holding ceremonial exchanges or dances, and going to war. Consumption of kava was often a central activity surrounding these chiefly occupations.

One of the main chiefly titles was Yeremwanu, which belonged to men who had the right to wear a hair ornament of hawk feathers (see Chapter 5). There were other chiefs who bore the title of Yani, including Yani Netata, district chiefs (Bonnemaison 1994: 146–156; Guiart 1956: 15–17). Among this rank the Yani en Dete follow to some extent the role of Oceanic ‘talking chiefs’, being responsible for speaking at various sorts of gatherings, and playing the role more broadly of ‘guardian of social well-being and local values and practices’ (Douglas 1996: 243). Chiefly titles were generally inherited, though where the oldest son of a chief was considered unfit, the title could be passed on to a different individual, presumably another son (Humphreys 1926: 36). Beyond passing on certain privileges, there is evidence that chiefly titles involved to some extent the Oceanic ‘Heroic I’. Those who hold a title lay claim not only to their own achievements, but those of every previous holder of that title (Lindstrom 2011).

Besides chiefs, there were various magical specialists in Tannese society. As on Erromango, magic could take on benevolent or malevolent forms. The primary goal of benevolent magic is to increase agricultural productivity, especially for yams. Yams are not only a dietary staple, but also feature in important annual exchange rituals. Magic stones, usually shaped like the crop that was to benefit from the magical activity (called nukwei narak), were kept and used by particular individuals to increase productivity. Similarly, beneficial magic could be used to bring rain or sunshine, summon fish, ensure the successful breeding and growth of pigs, and many other purposes (Bonnemaison 1991, 1994: 172–178; Guiart 1956: 63–66; Humphreys 1926: 71). There were also oracles called Narumin who were thought to have the ability to predict the future and divine the unknown (Humphreys 1926: 70).

Black magic was the domain of greatly feared sorcerers called Tupunas. All cases of illness and death were attributed either to the activities of Tupunas or malevolent ghosts called ierehma (yarmis or yermis in other Tanna languages). In the practice of black magic (narak), an item from the body, including hair, food scraps, clothing or another item, is bundled in leaves with a magic stone. If the leaves are burned, the victim perishes almost immediately, or lesser illnesses can be caused by keeping the personal effect at some distance from the malevolent stone (Bonnemaison 1994: 179–180; Guiart 1956: 69–72; Humphreys 1926: 72–73). Significantly, the magic stones of Erromango were said to come from Tanna, suggesting the practice may have spread from one island to the other (Chapter 2). Spiritual beliefs on Tanna appear to have been primarily concerned with these various types of magic and relationships to ancestral beings. Tannese people attributed the action of supernatural forces causally to the behaviours of human ritual practitioners. This is a more anthropocentric perspective when compared with neighbouring Aneityum, where spirits were treated more as autonomous beings (Douglas 1989).

There was apparently belief in a supreme being called Kwumwesin (Humphreys 1926: 71), but as with Nobu on Erromango, Kwumwesin appears to have made the world and then withdrew from all involvement with it. There are other important deities and legendary figures in Tannese cosmology, the most significant being Mwatiktiki (from the Polynesian Maui-tiki-tiki, one of the main deities whose existence can be traced linguistically to Proto-Polynesian, Maaui; Kirch and Green 2001: 243). Mwatiktiki stories in some ways parallel their Polynesian counterparts, for example in stories that say he fished the islands out of the sea. Others pertain specifically to Mwatiktiki’s arrival on Tanna, for example that he introduced pigs and ‘real’ kava to the island. Tangalua (the sea snake) is another clear Polynesian introduction that can be traced back to Proto-Polynesian (Taangaloa; Kirch and Green 2001: 245). There is a great deal of evidence for Polynesian influence on Tanna, particularly in the language and rituals pertaining to kava (Lindstrom 2004; Lynch 1996).

The everyday ritual of kava drinking is central to men’s lives, and social life more broadly on Tanna (Bonnemaison 1994: 182; Humphreys 1926: 81–83). Men retreat each afternoon to the imwarim (in the language of south Tanna; yimwayim in west Tanna; nakamal in Bislama), large, cathedral-like clearings usually surrounded by large banyan trees (Figure 3.2). The kava is prepared (traditionally the root is chewed) while men discuss the events of the day. As each drinks he makes his tamafa, a prayer accompanied by the spitting of fine droplets of saliva and kava (Lindstrom 1980). As they become intoxicated, each man retreats to his own corner of the imwarim to listen to the spirits or ancestors. It has been likened to a ‘daily dissolution of society’, in which this retreat to ‘listen’ to kava contributes to the egalitarianism of Tannese people (Brunton 1979).

Figure 3.2 Kava-drinking ground near Waisisi.

Source: James Flexner

The imwarim were important nodes on the Tannese landscape. A traditional system of ‘roads’ (conceptual as much as physical) connected imwarim to networks of hamlets and gardens. One imwarim could host men from multiple hamlets, and while each man would have a ‘home’ place to drink kava, it was common for men to move between different kava-drinking grounds (Brunton 1989: 130–138; Lindstrom 1996). These networks would become important during the annual dance festivals, toka and nao, which served as significant exchange ceremonies. During these events, brides were exchanged, and alliances for both exchange and warfare were solidified. Significant to the structure of these exchanges were the two main rival ‘moieties’ on the island, Koyometa and Numrukuen (Bonnemaison 1994: 148–153; Guiart 1956: 24–27; 90–94).

Tannese gardens were enchanted spaces, powered by the magic stones that made various plants grow (Bonnemaison 1991, 1994: 173–176). The primary staple crop is yam, of which many varieties are grown, including giant ‘aristocratic’ yams raised specifically for exchange. Yams are sown in mounds (takwu) formed from holes in the ground that are then heaped with a mixture of earth, burned vegetation, ash, and magic leaves (Figure 3.3). One early observer noted takwu measuring 7 feet (2m) in height and 60 feet (20m) in circumference. The yams from these mounds could grow to four feet (1.5m) in length, and weigh up to 50 pounds (22kg; Turner 1861: 87). Alongside yams were grown taro, coconut, sugarcane, and a variety of gourds and fruit trees. Agricultural surplus was used to raise pigs, which were primarily important for chiefly exchanges. Competitive feasting was significant on Tanna as elsewhere in the southern New Hebrides. This may have been one of the mechanisms that further maintained the egalitarian social structure on the island, relating to a focus on quality rather than quantity of agricultural production (Spriggs 1986: 16–18).

Figure 3.3 ‘Yam gardens, East Tanna’, Symons Collection, Australian Museum (AMS354–244).

Source: Australian Museum Archives.

The everyday ritual of kava, the annual cycles of toka, nao, and other events, such as clearing, planting gardens, and the yam harvest, structure the sense of temporality on Tanna. In the language of the Kwamera area of south Tanna, ‘the word for the day after tomorrow (neis) is the same as the word for the day before yesterday, and the word for indefinite future (kwumweisin) is the same for the indefinite past’ (Lindstrom 2011: 146). This suggests a ‘timeless’ or ‘cyclical’ element to time on the island. At the same time, Tannese people do recognise a sense of historical progress and rupture. This is marked by legendary events in the deeper past, such as the arrival of Mwatiktiki, whose footprint is located on the coast near Kwaraka; or the arrival of the canoe carrying the magic stones that resulted in the formation of Koyometa and Numrukuen. It is also marked by the reverberations of European contacts, both direct, as with the arrival of European explorers or missionaries, and more broadly, as in the ‘stolen war’ of Shipimanwawa (Bonnemaison 1994; see also Chapter 4). This type of ‘historicity’ (Ballard 2014) parallels in some ways the archaeological approach used here, where specific events are measured against patterns of everyday life over the long term (see also Braudel 1980).

A Tannese Village: Anuikaraka Before the Mission Era

Thus far, we have explored mission encounters in the New Hebrides from the perspective of mission houses and the surrounding landscapes. Melanesian people appear where they interacted with the mission, either in the form of evidence for local materials, exchanges, and local labour, or as named individuals in missionary accounts. Archaeological remains of Tannese village sites in the area of Anuikaraka and Kwaraka provide a significant counterpoint (Flexner et al. 2016c). These sites are located within 2km of the mission remains in the Kwamera area (Figure 3.4). The remains from this area, which span several centuries, offer an opportunity to catch glimpses of everyday life in a Tannese village before, during, and after the period of mission encounters on the island. While these sites may not be completely ‘typical’ for Tannese villages, they do offer a certain baseline, and exhibit patterns that would have been broadly similar at least across the southern part of the island. These sites offer an opportunity to ‘span the prehistory/history divide’ (Lightfoot 1995), a necessary endeavour to understand southern New Hebrides missions and Melanesian landscapes from a broader temporal as well as spatial perspective.

The archaeological landscape of Anuikaraka and Kwaraka consists of a series of stone walls, enclosures, and mounds, covering an area of roughly 400m of coastal plain, extending slightly over 100m inland from the sea (Figure 3.5). The archaeological landscape is bisected by Komaru stream. Komaru also serves as the boundary between the two netata in the area: Umairarekarmene to the southwest, and Neaimene to the northeast. The area of Anuikaraka and Kwaraka has been surveyed and all the surface features mapped. A limited number of features were excavated, but knowledge of site stratigraphy, function, and chronology could be greatly improved with further research. The form of surface features and the excavation work that has been done has offered valuable information about these sites, and long-term landscape archaeology on Tanna more broadly (Flexner et al. 2016c).

Figure 3.4 Archaeological features, Kwamera area, south Tanna.

Source: James Flexner

Figure 3.5 Plan map of survey area, Kwaraka and Anuikaraka.

Source: James Flexner

For an overview of the landscape, we will start in the northeast of Kwaraka (Figure 3.6), then move southwest through to Anuikaraka. The first feature we encounter is a stone-walled enclosure, sunken slightly below the surrounding ground surface (Figure 3.7). This is the old imwarim (kava-drinking and dancing ground) Irumien. Within the feature there are several stones worth noting. There is a grinding stone with impressions from the processing of natapoa (in Bislama; scientific name Terminalia catappa) nuts in the eastern, seaward wall. Further south along this wall there is a row of stones embedded in the ground surface at the base of the wall. It is not known what function they have, but considering their location inside the imwarim, it is likely that they had some kind of ritual significance for the men who would drink their kava in this place. Further south, there is a flat stone inside of the rounded terminus of this wall, next to the southern entrance to the imwarim. This stone served as a refuge stone for people fleeing conflict. If the stone could be reached ahead of one’s pursuers then the imwarim offered protection (Flexner 2014c: 15–17).

Figure 3.6 Surface features, Kwaraka.

Source: James Flexner

Continuing on from Irumien, we enter a landscape of stone walls running parallel to the seashore, and punctuated with mounds of stone and earth. The walls may have served as stormbreaks against major surges on the sea side. The easternmost set of walls (W1-B and W3-B in Figure 3.6) form a low terrace of sorts for the village. Below this, there was another imwarim, which was cleared to make way for a football field in recent history. The mounds are remembered locally as being built to raise houses and storage areas, as the ground surface could become extremely muddy during the rainy season (usually lasting from November until July). These features are mostly unexcavated at this point. A shovel test pit (STP17) excavated in the area of ‘Wall 3’ (W3-A in Figure 3.6) contained charcoal, bone, shell, glass, and metal artefacts, suggesting a midden deposit in this area (Flexner et al. 2016c). A group of three round stone-faced mounds (M2, M3, and M4) are particularly well-preserved. These become of interest in the missionary and post-missionary era at Kwaraka, and will be returned to below. Inland from Kwaraka, there is a collection of smooth yam stones (nukwei nuk) of fine-grained basalt (Figure 3.8). These would have provided the magic to make the yams grow in the inland gardens connected to the hamlets and imwarim of this area.

Figure 3.7 Imwarim (kava-drinking ground) at Irumien.

Source: James Flexner

Figure 3.8 Nukwei nuk (yam stones).

Source: James Flexner

Two of the well-preserved stone-faced mounds in the western part of Kwaraka (M3 and M4 in Figure 3.6) bear a striking resemblance to contact-era house mounds from Samoa (Green and Davidson, eds 1974). This raises the possibility that this kind of construction technique was introduced to Tanna either directly by Samoan teachers, or by Aneityumese who learned from the Polynesians (note once again that the Aneityumese were themselves accomplished builders in stone, see Spriggs 1981). At this point, further excavation work is needed to test this hypothesis. This site is extremely significant for the history of mission contacts in southern Vanuatu, because of its association with important oral histories, and its great archaeological potential (see Flexner 2014c: 12–18; Flexner et al. 2016c).

Figure 3.9 Surface features, Anuikaraka.

Source: James Flexner

Across the rocky channel of the intermittent Komaru stream (which flows during heavy rains and throughout the wet season), the features continue into Anuikaraka (Figure 3.9). The stone walls and mounds in Anuikaraka are less well-preserved than in Kwaraka. Where Kwaraka is flat, Anuikaraka is relatively steeply sloped. Further, the walls tend to run perpendicular to the seashore, directed downslope. The reason for this is unclear, though channelling rainwater downslope may offer a reason for this. At least some of the walls (W10 and W11) flanked a traditional road leading to another imwarim, which has also been lost to much more recent road-building activities. The enclosure in the northern area of the site (E1 in Figure 3.9) is associated with the chief Iarisi, who was an important individual in early mission history on Tanna.

Figure 3.10 Mound 9 plan and TU6 stratigraphic profile, Anuikaraka.

Source: James Flexner

One of the round, conical mounds (M9 in Figure 3.9) held a kastom stone on the surface, which was associated with women’s magic relating to the toka dance. It was said to have been placed there by Iarisi to contain its power. A 1x4m trench was excavated across one side of this mound to better understand its stratigraphy and age (Figure 3.10). The mound was constructed of an undifferentiated deposit of stone cobbles, boulders, and earth. The construction fill contained several smooth, cobble-sized pieces and smaller fragments of red ochre, pig bones, Cypraea (cowrie) and Turbo shell fragments, and the distal end of a fine-grained basalt adze blade. These materials, particularly the pig bone and ochre fragments suggest a ritual deposition of some sort. Pigs were extremely significant to chiefly exchanges on Tanna, and red ochre was important as a pigment, especially as body paint. Ochre was often traded from neighbouring Aneityum Island, which was connected to this area by a major kastom ‘road’ (canoe voyaging route).

Radiocarbon dates indicate that the mound was constructed in the mid-17th or 18th century (Appendix D), almost certainly before Cook set foot on Tanna. This feature may have been constructed as part of a ritual feasting event. The remains of sacrificed pigs, food debris, and red ochre were heaped together, possibly as a way of closing a major ceremonial event or as a sacrifice to propitiate ierehma (ancestral beings). At this point, stone features such as this are known only from Anuikaraka and Kwaraka. In part, this reflects the lack of archaeological fieldwork overall on Tanna. However, this kind of landscape may be unique to the southern part of Tanna, which has close connections to Aneityum, an island well-known for large-scale stone construction (Spriggs 1981, 1986). Future research on Tannese settlement patterns will be needed to clarify this. Regardless, these features offer a valuable dataset for understanding at least some of the practices that existed prior to missionary arrival on Tanna. Annual cycles of agricultural production, community and island-wide ceremonies, and the daily rituals of kava at the imwarim are integral components of Tannese kastom. These practices existed for centuries before missionary arrival and, as will be seen, continued well beyond the period when missionaries were no longer active on Tanna.

Contacts in the Fish’s Tail: Port Resolution

I named the Harbour, Port Resolution after the Ship, as she was the first who ever entered it

— James Cook, 1774 (Beaglehole, ed. 1969: 508)

The Resolution was not the first long-distance maritime vessel to land in Port Resolution, as Oceanic sailing canoes would have regularly visited the naturally sheltered harbour for millennia before Cook’s arrival. As seen above, south Tanna had important exchange ‘roads’ leading to Futuna, Aniwa, and Aneityum. It may be this geographic feature, and the historical precedent set by Cook, that partly explains why so many ‘contact events’ took place in the Port Resolution area. The landform that created the much-desired harbour is locally called Nipikinamu (literally ‘the fish’s tail’; Figure 3.11). The Port Resolution area spans three netata (see Figure 3.1), a significant fact for later complications in missionary work.

Figure 3.11 View along ‘the fish’s tail’, Port Resolution. The small outcrop on the far right of the peninsula is named ‘Captain Cook’.

Source: James Flexner

Cook at Port Resolution

The Resolution sailed into the harbour for which it would be named on 5 August 1774. As was typical of initial encounters with Pacific Islanders during Cook’s voyages, the initial moments were marked by exchange of materials, including cloth, medals, and nails for coconuts, yams, and other crops. Cook notes in his diaries that the landing party sent the following day to establish friendly relations and to collect water and wood was met by two parties of armed men (possibly representing Koyometa and Numrukuen or two of the netata in the area). On the second landing, there were at least 1,000 people present, and apparently this number grew with each subsequent landing. This landing was also met with an arrangement of reeds on the shore, in the middle of which were bundles of food (Beaglehole, ed. 1969: 482–484). It is possible that Cook and his crew were seen as ierehma (ghosts) who had returned over the sea. As such they were dangerous, and it is likely that this offering was meant to propitiate them in the hope that they would simply leave (feeding the ierehma is a traditional kastom work; Bonnemaison 1994: 178–179).

Superficially friendly relations were established through the help of an old man named Paowang, a local mediator who must have been a high-ranking chief. Among other gifts, Paowang received a pair of Tahitian dogs from Cook. Keeping to an established structure for controlling such encounters, Cook used firearms, first muskets then larger cannon, to frighten the Tannese when it appeared that they had the upper hand. The guns were never to be fired at people except as a last resort, and apparently this remained the case for the first part of this encounter. Despite his interest, Cook was generally unable to trade for clubs or other arms, though he was able to collect a few objects in the encounter (see Chapter 5). After a few days, Cook befriended a young man named ‘Wha-a-gou’ (a Polynesian-sounding name, suggesting possibly a visitor from Aniwa or Futuna), who was persuaded to come on board the Resolution. Though they won some friends among the coastal dwellers, Cook and his party were generally prevented from travelling inland. Specifically, they were not able to walk to Mt Yasur, the volcano that had drawn them to Tanna in the first place (Beaglehole, ed. 1969: 486–493). It is possible that crossing the boundary to the netata of Yanekahi was tabu for some reason. Or perhaps the idea of Cook’s crew as unknown entities walking to Yasur, a highly sacred place, was simply seen as too potentially dangerous. Regardless, every attempt to reach the volcano was turned into a wild goose chase that inevitably meandered back to the small area of beach used by the crew of the Resolution as a landing place.

On the 19 August 1774, Cook’s relationship to people in Port Resolution was tragically and irreparably damaged when a sentry shot and killed one of the local people. Even though the men (and apparently some women) of the area went about constantly armed, and occasionally made menacing gestures, there had not been a single attack during the two weeks the Resolution was at harbour. All it took was a single thoughtless moment by one of the ship’s marines to shatter the tentative peace. The following day, Cook decided to weigh anchor and summarily departed (Beaglehole, ed. 1969: 499–500). What this encounter would have meant to Tannese people is difficult to know precisely, though it certainly left an imprint on social memory. In the 1840s, the missionary George Turner recorded an oral tradition interpreting the events, in which Cook came to Tanna to attack a Tupunas in the Port Resolution area, after which he departed. ‘For the Tannese, the event assumed significance in terms of a different cultural framework; as it turned out in political terms, it became an instance of their control over the European’ (Adams 1984: 31). Cook’s presence has been sedimented into the landscape via toponyms, from the misnomer island name ‘Tanna’ (Jolly 2009), to a sandstone outcrop from which he is said to have made cartographic measurements (Figure 3.12).

Mission Contacts from Williams to Watt

Missionary endeavours on Tanna began with John Williams and Jacob Harris in November 1839. The earliest contacts occurred in the Port Resolution area. As elsewhere, these encounters are an integral part of the cultural landscape in Port Resolution (Figure 3.12). The most notable event that occurred involving the initial visit of the London Missionary Society (LMS) men was a Tannese chief spitting down the throat of Mr Harris (Lindstrom 1980: 228). When Williams and Harris travelled on to Erromango the following day, they left three Samoan missionaries, Mose, Lalolagi, and Salamea, at Port Resolution. They were joined in 1840 by Pomare and Vaiofaga. Shortly after the new arrivals landed, there was a major outbreak of disease, which was blamed on the foreign god of the missionaries. This was in spite of the fact that the Samoans themselves became ill during the same epidemic, and Salamea and Pomare both died in the event. The Tannese cut off contact with the mission. In 1841, two more teachers, Faleese and Apolo, joined the struggling mission, while Lalolagi returned home to Samoa (Latai forthcoming; Liua‘ana 1996: 52). The imwarim of Yakuperang, now referred to as ‘Samoa’, is likely the place where these teachers lived during their first two years on Tanna.

Figure 3.12 Port Resolution area site locations.

Source: James Flexner

The Scottish LMS missionaries Turner and Nisbet arrived in Port Resolution in 1842 along with the Cook Islands teacher Kapao (Liua‘ana 1996: 52). Within a few days of their arrival, the area was thrown into a state of upheaval after local people were attacked, apparently without reason, by the crew of an American trading vessel. Turner and Nisbet spent roughly seven months on Tanna. Their first act was to build a ‘sixty feet weather-boarded cottage’ from materials they had brought on the ship (Turner 1861: 7). This may have been the first prefabricated building in the New Hebrides, and would have been one of the earliest in the region (see Flexner et al. 2015: 266–267). Theft was a great concern, both during the construction of this building and after, though Turner suggests that the practice was ubiquitous and not particularly targeted at the missionaries (Turner 1861: 8).

Almost immediately after their arrival, Turner and Nisbet witnessed a battle taking place directly on the mission ground. They endeavoured to stop the fighting, but were unable to do so. Within a few months, there appeared to be renewed interest in Christianity among local people. A printing press was set up, and a few hymns composed in the local language. Emboldened, Turner and Nisbet sought to move inland towards the volcano, but were resisted on that front. Almost inevitably, another epidemic, this time of dysentery, struck in the area. The Tupunas vocally blamed the missionaries and their god as the source of the outbreak. A local chief, Teman, had started as a supporter of the mission but switched sides and perished shortly thereafter. Within a few days, war was declared upon the missionaries and their remaining supporters. As the situation deteriorated, Turner and Nisbet decided to abandon the mission on Tanna. In the last act, Iāru, a chiefly supporter of the mission, asked for the gun that the missionaries had brought with them to turn the tides. The missionaries refused. Another supporter, Kuanan, had his club taken by opposing warriors in an ambush, a significant loss. After a few days of planning, the missionaries had to flee in the night on the brig Highlander under Captain Lucas, returning to Samoa (Turner 1861: 11–68).

One of the main challenges of early mission work on Tanna was lack of understanding of local culture. Language was a major challenge, as Polynesian terms could not be simply translated into Tannese languages. This led to a number of difficulties in attempting to explain Christian beliefs to local people, especially in translating terms for ‘god’, ‘devil’, ‘spirit’, and so on. Further, the missionaries inadvertently placed themselves in the middle of a conflict between the local tribes. Neraimene and Nepikinamame, eastern tribes led by Viavia, formed one side. Kaserumene and Yanekahi to the west, led by Lamias, were the other. When epidemic disease broke out, Turner and Nisbet were seen as sorcerers on the Neraimene/Nepikinamame side. By refusing a propitiatory gift from Kaserumene, Turner and Nisbet only showed that this was indeed the case (Adams 1984: 56–69).

While there would have been temporary relief at Turner and Nisbet’s departure, the epidemics continued. This caused some local people to change their minds about the potency of the missionaries’ god, believing that the missionaries’ settlement, objects, and rituals were indeed tabu. Polynesian teachers returned to Tanna in 1845. The Samoans included Adamu and Iona, who fled from Aniwa, along with Ioane, Petelu, Pita, and Tagifo, from Tutuila, accompanied by their wives. Some of these missionaries were settled among the Kaserumene and Yanekahi to try to avoid the earlier mistake of apparent missionary partisanship. In 1846, the Polynesians were joined by two more Samoans, Lefau and Vasa, and three Cook Islanders, Marugatanga, Upokumanu, and Rangia. But by this point, epidemic disease had broken out again. Those Samoans who didn’t succumb were accused of sorcery. Vasa was clubbed to death by Neraimene warriors. The fortuitous appearance of a whaling ship allowed the majority of the teachers to escape to Aneityum. Upokumanu, who was protected by a Yanekahi yeremwanu named Kapahai, appears to have stayed for a time before returning to Aneityum (Adams 1984: 70; Liua‘ana 1996: 54–55).

From 1846–1853, Polynesian teachers continued to be settled at Port Resolution, with almost no success. A smallpox outbreak in 1853 finally ended the LMS endeavour at Port Resolution (Liua‘ana 1996: 55–56; Miller 1978: 36–37). Archaeologically, almost nothing is known about this period, aside from the oral traditions associated with the imwarim Yakuperang. No archaeological excavations have been carried out at Port Resolution at this point. A brief survey recorded a number of archaeological surface features, mostly relating to later mission contacts in the area (Figure 3.12). In 1858, the Scottish Presbyterian missionary John G. Paton settled on Tanna along with the Canadian John Matheson. Paton and his wife settled at Port Resolution, while Matheson and his wife settled in the Kwamera area. Samuel Johnston, another Canadian, joined the Tanna Mission in 1860. Within two years, Johnston, John G. Paton’s wife Mary, and the Patons’ son died. (Adams 1984: 99–115; Miller 1981: 20–35; Paton 1907 Vol. 1; Patterson 1864). Like their LMS predecessors, the Presbyterians struggled to win converts on Tanna, owing largely to difficulties in translating Biblical concepts, and lack of understanding of Tannese kastom. Paton showed himself to be particularly hard-headed, and prone to making these kinds of errors (e.g. Adams 1984: 112–113). In 1862, following a series of deadly measles outbreaks, and a destructive cyclone, missionaries were again condemned as dangerous foreign sorcerers, and forced to flee the island (Adams 1984: 116–149).

The surface remains in the Port Resolution area relating to these early contacts include the grave site of Paton’s first wife and child, as well as Rev. Johnston, all of whom died of illness on the island. These graves were located next to Paton’s first mission house in Port Resolution. The house had been built of wattle and plaster, and apparently contained, ‘A grand piano, china dinner, and silver cutlery services’, typical markers of civilised domestic life (Adams 1984: 100). After these deaths, Paton chose to relocate to a more ‘healthful’ location up the hill (Paton 1907 Vol. 1: 128–138). The site is marked by a minor rubble scatter, and not much else, though apparently gardeners in the area occasionally find pieces of earthenware. The site would merit future test excavation. Nearby is an old mango tree said to have been the first in the area, planted by the missionaries in the late 1850s. These features all cluster on the south side of the harbour.

Figure 3.13 Schematic of the later mission compound at Port Resolution (1: mission church; 2: mission house; 3: printing house; 4: Agnes Watt grave; 5: boat landing). Forms and locations of features are approximate only.

Source: James Flexner

The features on the east side of the harbour relate primarily to the later missionary tenure of the Watts, Scottish missionaries supported by the New Zealand Reformed Church who were settled on Tanna in 1869 (Lindstrom 2013; Miller 1981: 38–39; Watt 1896). These later contacts will be the focus of Chapter 4, but bear brief description here as the features are presented in Figure 3.12. No clear structural remains are present at this point as the mission site is now also the location of the Port Resolution Yacht Club. The general layout of the mission could be roughly reconstructed based on local social memories (Figure 3.13). Agnes Watt died in 1894 and is buried next to the old church. Her marble headstone came from the manufacturers Bouskli and McNab of Auckland, reflecting both the New Zealand connections of the mission, and the increasing availability of heavy materials from overseas in this period. Construction activity on the site uncovered a lid from an iron ship’s tank bearing the mark of ‘JOHN BELLAMY BYNG ST. MILLWALL’. These kinds of tanks were often used in colonial Australia as water tanks or boilers (Pearson 1992). Presumably this would have been the case at Port Resolution as well.

Two more features are worth noting, as they relate to more recent integration of mission heritage and kastom in the 20th century (Flexner and Spriggs 2015). The bell from the Watts’ church at Port Resolution has been placed in a tree in one of the central areas of the current village. The bell bears the inscription, ‘COME UNTO ME. [ST.] PAUL’S, GLASGOW. 1890. JOHN C. WILSON & Co FOUNDERS. GL[ASGOW]’ (Figure 3.14). The tree stands near the site of the church erected by the Watts in 1891 (Watt 1896: 321–324). The other feature is the grave site of early church elders, including ‘BARAUN’ (also sometimes spelled Braun), who was one of the teachers helping the Watts at a new station established in Ikurupu (Watt 1896: 315). The continued significance of the graves of Tannese elders reflects the importance of Melanesian integration of Christianity into local life. Local curation of artefacts from the mission period, such as the bell, shows that material from this era holds continued relevance. This is especially notable in the Port Resolution area, where local reinterpretation of modernity has taken the form of the John Frum ‘cargo cult’ (Lindstrom 1993; Tabani 2010).

Figure 3.14 Old mission bell in a tree, Port Resolution.

Source: James Flexner

The Mathesons at Imua

John and Mary Matheson settled in the Kwamera area in the netata of Umairarekarmene in 1858 as part of an effort to revitalise the Tanna Mission. Their initial house location was on the coast, where they built a three-room cottage measuring 40x15 feet (13x5m) with a view of Aneityum and easy access to fresh water (Patterson 1864: 381–382). In 1859, after recurring illnesses, the Mathesons decided to quit the mission station in Kwamera for the cooler climate of Aneityum. Presumably they also had better access to healthcare and other resources in Geddie’s Mission stronghold at Anelcauhat. Eventually they spent a short time at the neighbouring mission station of Umetch, and a rainy season on Erromango (Patterson 1864: 397–445).

Having recovered, the Mathesons resumed their labours on Tanna in April 1860. The missionaries moved their house inland to higher elevation, which was thought to be less likely to cause illness (see also discussion of G. Gordon House, Chapter 2). Nonetheless, they continued to suffer to greater and lesser degrees for their remaining time on Tanna (Patterson 1864: 446–466). The Mathesons were under the protection of a local chief named Kapuku. As elsewhere on the island, there were major outbreaks of measles and dysentery, and the aforementioned cyclone in 1860. Kapuku was one of the ones who became severely ill during one of the measles outbreaks, and afterwards avoided the mission. The Mathesons insisted that only their God could cause or cure illness. Once again, the missionaries cast themselves as Tupunas, dangerous sorcerers. When John Matheson asked local people about their hostility, one response from the yeremwanu Kati was that as long as God didn’t know about their heathenism, he couldn’t punish them for it, so by avoiding the missionaries, they avoided illness (Adams 1984: 119–121). Eventually, the position of the missionaries became unsustainable, and with ongoing epidemics, failing health, and increasing local hostility, including the burning of the church building at Kwamera, the Mathesons were forced to flee from Tanna, along with the Patons, on 2 February 1862 (Adams 1984: 143–145; Miller 1981: 32–33; Patterson 1864: 494–498). Both of the Mathesons died of illness shortly after this, Mary a few weeks later on Aneityum and John on Maré in the Loyalty Islands of New Caledonia in June 1862.

Imua Mission Excavations

The place where the Mathesons’ second house was built is called Imua, a local orthography of ‘Samoa’ in honour of the teachers who were settled in the area from 1854 onwards. Kapuku, who despite his avoidance remained a friend of the mission, is buried near the house site (see Figure 3.4). The remains of the mission site consist of single-course stone alignments visible on the surface (Figure 3.15). Artefacts, primarily transfer-printed refined earthenwares, were also present on the surface. The stone alignments appear to group into two ‘structures’, though site stratigraphy suggests that the western alignments were likely garden boundaries around which the house collapsed. Visiting the site in 1869, Agnes Watt noted that the ruins of the house, garden paths, and broken dishes were still visible on the surface, and the lemon trees the Mathesons had planted were still bearing fruit (Watt 1896: 91).

Excavations at the house (to be referred to as ‘Imua Mission’ below, to distinguish it from the earlier Matheson house, which has yet to be discovered) consisted of a 2x2m test unit in the eastern structure, and a 4x4m area consisting of four contiguous units in the west. The unit in the eastern structure (TU3) had a simple stratigraphic sequence, with a very dark brown topsoil layer overlying a layer containing mortar fragments and artefacts, which overlay a clay subsoil (Figure 3.16). The ceramic assemblage from TU3 appears to be somewhat distinct from the western area, suggesting functionally different use of space. Specifically, TU3 contained a chamberpot and mineral water bottle, so this part of the house may have been used for rituals relating to personal hygiene. The western excavation area showed a similar stratigraphic sequence (Figure 3.17), but the concentration of artefacts was much higher across most of the unit. Excavation of the topsoil revealed that the dense concentration of artefacts was bounded on the southwestern corner (Figure 3.18), and was not correlated with the stone alignments. For this reason we doubt that the surface stones to the west were building wall footings, though they may have supported or framed low walls or fences.

Figure 3.15 Plan of Imua Mission.

Source: James Flexner

Figure 3.16 Stratigraphic profile, TU3, Imua Mission.

Source: James Flexner

Figure 3.17 Stratigraphic profiles, TU1/2/4/5, Imua Mission.

Source: James Flexner

Figure 3.18 Plan of western excavation area after removal of the topsoil revealed the full extent of stone alignments, and concentration of building materials and artefacts from the mission house at Imua.

Source: James Flexner

Even before the Mathesons abandoned the site, the house and other mission buildings, including a church and storehouse closer to the coast, had been badly damaged in a storm (Patterson 1864: 482). After the house was abandoned, the structure and whatever materials remained inside would have deteriorated rapidly. The deposit of artefacts and lime mortar fragments in the western excavation area at Imua Mission was the highest recovered from any site excavated for this project. This is interpreted as a ‘destruction layer’, which formed after the house had been abandoned. The high concentration of materials reflects in part the suddenness with which the Mathesons departed, leaving most of the household goods behind (though there is some evidence that a few valuables were placed in a boat house that a later observer found unplundered; Patterson 1864: 497–498).

In some places, ceramic sherds were stacked on top of each other, and groups of lime mortar fragments clustered together where larger sections of wall had fallen down (Figure 3.19). A 2x2m area of this deposit in the northwest (PN196, TU5) contained more ceramic artefacts than had been recovered from all of the other excavations on Tanna and Erromango mission sites combined. If anything, artefact concentration appears to increase towards the north and west, suggesting the house was located in this direction. There is evidence for postdepositional movement of artefacts and mortar fragments, primarily as a result of the activities of burrowing animals, including rats and crabs, as well as tree root growth. There was some vertical tilting of ceramic sherds. Stratigraphically, the stone alignments overlie some of the artefact concentration, though this is interpreted as occurring because of animal burrowing and tree roots moving things under the stones.

Figure 3.19 In situ concentration of lime mortar fragments and ceramic sherds, Imua.

Source: James Flexner

Future research could pinpoint the location of the house and any other buildings on the site, as well as recovering more of the very rich artefact assemblage from the site. The house was made of wattle and lime mortar, probably set directly onto compacted earth. Because of the aforementioned formation processes (Schiffer 1987) at the site, the house structure will probably appear as a concentration of mortar fragments and artefacts rather than a discrete architectural ‘feature’. The stone alignments on the surface are thus more likely interpreted as garden paths or other features, though the eastern structure has not been ruled out as a possible building.

Imua Artefact Assemblage

The artefacts from Imua Mission provide an astonishing glimpse into mission life on Tanna for the brief period that the house was inhabited by the Mathesons. The amount of material recovered from the site far outweighs what was recovered from any of the other excavated mission sites from Tanna and Erromango. Lime mortar was only sampled from the site, because of limitations on how much could be brought back to the National Museum in Port Vila for analysis. Over 30kg of mortar was removed from the site and analysed. Of this, roughly 28 per cent by weight (8.5kg, N=175) had beam impressions from the wattle structure of the house (Figure 3.20). Another 7 per cent still retained the original whitewash from the wall surfaces, with 13 of these fragments coming from a single concentration within the western excavation area.

Figure 3.20 Lime mortar fragments with wattle impressions, Imua.

Source: James Flexner

Other architectural materials were abundant as well, including 827g of window glass fragments. Again this is far more than was recovered from any other site, even taking into account the slightly larger excavation area at Imua (20m2 total). A total of 1,550 nails and nail fragments were recovered. The identifiable nails were primarily of the hand-wrought type (N=1422, 3.76kg), though there were also a few machine-cut nails (N=11, 16.5g), and modern round wire nails (N=30, 69g). The latter are thought to be evidence for more recent activities around the site long after its abandonment, also shown by some 20th-century glass vessels found on the surface. Of the wrought nails, roughly 16 per cent by count (N=232) showed some evidence of being bent, including 56 that are thought to have been intentionally bent into hooks (Figure 3.21).

Figure 3.21 Hand-wrought iron nails, possibly bent intentionally, Imua.

Source: James Flexner

Very few glass artefacts aside from the window glass shards were recovered at Imua. Not counting the two 20th-century examples already noted, glass was limited to 13 colourless and aquamarine shards of thin-walled bottles. Notably, two of the colourless bottle fragments included embossed writing, one simply ‘ON’ or ‘NO’, and the other ‘… ERION …’ Two conjoining sherds of an aquamarine bottle include the letters ‘… IC’, suggesting ‘TONIC’. It is certainly likely that the Mathesons would have used some kinds of medicine or tonics typical of the 19th century considering their apparent ill-health. Thus far, there is no clear evidence for alcohol consumption at Imua. As is typical, faunal remains were dominated by shellfish, including the usual taxa of Turbo, Cypraea, Trochidae, Conidae, Nerita, and Arcidae. A few taxa not found on other mission sites, including Ostreidae and Triton were also present. The Triton is intriguing, as it was used not only as a food source but for large shell trumpets (called bubu in Bislama). The presence of several large Triton fragments suggests that such a trumpet may have indeed been used to call people to meetings at the Mathesons’ Mission.

Artefacts relating to personal adornment included six buttons (Figure 3.22). Four consisted of round metal alloy backings with dark blue glass paste jewels. There was also a plain copper alloy disc, which could have had a decorative covering that did not preserve, and a copper alloy shank with a mother of pearl disc. A single handmade bone button with three holes was also recovered. Missionaries attempted to maintain Victorian standards of dress in the field, while also attempting to encourage changing habits of dress among local people. Mary Matheson may have been particularly driven to cover the body as a way of suppressing its dangerous passions, especially in contrast to the nakedness of the Melanesian people who surrounded the mission (Adams 1984: 99–100). At the same time, increasingly localised frames of reference in bodily adornment are apparent in the form of a Nerita shell that has been perforated on one side to form a simple pendant (Figure 3.23).

Figure 3.22 Buttons from Imua (top row, from left: three iron with blue glass paste jewel, one copper alloy disc; bottom row, from left: handmade bone button with three holes, mother of pearl disc with copper alloy shank).

Source: James Flexner

Figure 3.23 Nerita shell pendant, Imua.

Source: James Flexner

The most impressive finds from Imua were the ceramic artefacts (Flexner and Ball 2016). Included among these was a ceramic sheep’s head (Figure 3.24). This is thought to have come from a child’s toy. Such artefacts are generally rare from mission sites in Australasia (Middleton 2008: 219–220). A porcelain doll was recovered from Ebenezer Mission in southeast Australia. Such artefacts were meant to habituate indigenous female children to Western standards of domesticity (Lydon 2009: 138–139). In this case, the sheep may also have related to an idealised pastoral British identity, not to mention the other obvious Biblical references to the ‘Lamb of God’. It should also be noted that the Mathesons had a baby girl in 1861, though she only lived for a few months (Patterson 1864: 493–494). It is possible that this object was originally intended for the Mathesons’ child, then used for native children before being left behind when the house was abandoned.

Figure 3.24 Ceramic sheep’s head, Imua.

Source: James Flexner

The ceramics from Imua house likewise have a distinctly pastoral theme. A total of 2,739 sherds weighing 16.6kg were recovered from Imua. The sherds have been reconstructed to a minimum of 36 vessels (Table 3.1). Of these, 26 are of the dense, white refined earthenware generally called ‘whiteware’ with variations of the same blue transfer-printed design. The motifs are pastoral, featuring sheep, cattle, and castle ruins (Figure 3.25). These are arranged in a central vignette, and then elements of the design are repeated in cartouches around the rim. Two of the vessels bear the stamp ‘ARCADIA’, and the remaining 24 pastoral vessels bear an identical pattern or related variation. It was typical for only a small portion of vessels in a transfer-printed set to bear stamps (Samford 1997: 1). Arcadia was produced by J. & M. P. Bell & Co., a Glasgow pottery active from 1842–1928 (Brooks 2005: 69; Coysh and Henrywood 1982: 24). The pottery is a projection of idealised British pastoral landscapes (see Brooks 1999), and possibly the Scottish identity of the missionaries. John G. Paton served as a city missionary in Glasgow before departing for the New Hebrides (Adams 1984: 88–94; Paton 1907: 53–84), though Bell ceramics were common throughout Australasia so there is no reason to assume these vessels came directly with Paton.

Table 3.1 Reconstructed vessels from Imua (minimum number of vessels).

Vessel Frags.

Context

MNV Count

Short Description

Transfer Group

VF10-19

TU1/2/4/5

2

Flow Blue bowl

Flow Blue

VF8

TU1/2/4/5

1

Linear cup

Linear

VF2-7

TU1/2/4/5

2

Linear plate

Linear

VF9

TU1/2/4/5

1

Brown stoneware storage jar or jug

NA

VF41

TU1/2/4/5

1

Handled vessel, hand-painted floral motif

NA

VF35-36

TU1/2/4/5

2

Arcadia round shallow bowl

Romantic-Pastoral

VF63/64

TU1/2/4/5

1

Repeated motif, Arcadia round shallow bowl (one of these two might go with VF35)

Romantic-Pastoral

VF21-27

TU1/2/4/5

2

Large rectangular platter, Arcadia

Romantic-Pastoral

VF40, 42

TU1/2/4/5

1

Moulded rim bowl, smaller, Arcadia

Romantic-Pastoral

VF43

TU1/2/4/5

1

Pastoral storage bowl, interior (cattle/cartouche) and exterior (architecture), Arcadia

Romantic-Pastoral

VF62, 69-73

TU1/2/4/5

1

‘Mezzanine’/Meat Strainer, Arcadia variant

Romantic-Pastoral

VF63-64, 66-68

TU1/2/4/5

2

Possibly fragments of the same open vessel base, repeated boat motif on non-mending sherd, also repeating motif on VF63/64. VF110 rim might go with this set. VF66 and VF68 are clearly the same pattern, though slightly overlap (so 2 vessels min). Possibly an Arcadia variant, but not definite

Romantic-Pastoral

VF65

TU1/2/4/5

1

Very blurry base with slightly different pastoral motif, again possibly Arcadia but not certain

Romantic-Pastoral

VF44-61, 75-150

TU1/2/4/5

14

Most common pastoral motif, minimum vessels based on non-repeating motif of base fragments. There are slightly sharper and slightly blurrier (blurry is represented by VF55, VF56 of the bases) versions of the pattern, more variable on the rims. Rims indicate at least three rectangular platters, at least four shallow bowls or plates, and of course there would be more of each. The round shallow bowl/plate was the more common form, so the estimate is an MNV of 10 round ones and 4 rectangular.

Romantic-Pastoral

VF1

TU1/2/4/5

1

Black ‘JOHN’ saucer

Scriptural

Sherd 2103

TU3

1

Selter’s Mineral Water bottle, stoneware

NA

Sherd 2104

TU3

1

Undec. whiteware bowl

NA

VF37-39

TU3

1

Chinese Fountains chamberpot

Oriental

Figure 3.25 ‘Arcadia’ transfer pattern, Imua.

Source: James Flexner

Other transfer-printed vessels from Imua include a chamberpot from TU3 with a pattern identified as ‘Chinese Fountains’ (Figure 3.26a), which was produced by Elkin Knight and Bridgwood in England from 1822–1846 (J. Adamson, pers. comm.; Williams and Weber 1978: 110). There were at least two plates and one cup with a linear design of blue and seafoam green (Figure 3.26b). A small, fine-moulded saucer bore the only black transfer-printed design, which consists of a geometric border and the name ‘JOHN’ in the middle (Figure 3.26c). This may have been part of a set of children’s teawares, with the names of the gospels (Matthew, Mark, Luke, and John) printed on the saucers. These kinds of moralising children’s wares were fairly common in the Victorian era (Williams and Weber 1978: 534–570). There were at least two bowls of the type known as ‘Flow Blue’ (Gaston 1989), bearing a transfer pattern that was intentionally over-inked so that the blue colour bled. The motifs on the Flow Blue bowls are quite similar to the Chinese Fountains pattern (Figure 3.26d).

Figure 3.26 Other transfer patterns from Imua (see text for details).

Source: James Flexner

There were only three vessels that did not bear transfer printing. There was a single undecorated whiteware bowl from TU3. The remaining two vessels are utilitarian stonewares. The mouth of a brown glazed stoneware jug is represented by three sherds. None of the body of this vessel was recovered. The other stoneware vessel is a cylindrical bottle bearing a stamp on the shoulder that consists of a lion surrounded by two concentric circles, with the word ‘SELTERS’ within the circles. Below this the letters ‘[H]ERZOGTH[UM]’ were stamped (Figure 3.27). This bottle contained mineral water from a famous spring in the Taunus Mountains of Germany. These were common trade items in the 1700s and 1800s (Lockhart 2010: 98). A recent find from a Polish shipwreck indicates that these bottles were sometimes refilled with alcoholic beverages (PAP 2014). Considering the ill health of which the Mathesons complained throughout their time on Tanna (Patterson 1864), it is more likely that in this case the mineral water was consumed for its medicinal properties.

Figure 3.27 Selter’s stoneware mineral water bottle from Imua.

Source: James Flexner

The vessel forms from the transfer-printed vessels consisted primarily of shallow bowl forms generally referred to as ‘soup plates’ (Figure 3.28; Brooks 2005: 47). There were also at least two large rectangular platters of Arcadia ware. Other vessel forms included a ‘Mezzanine’ or meat strainer, which was a flat vessel that would have sat on top of a platter, with holes in the body to allow the juice from meat or fish dishes to drip through (Brooks 2005: 49). These dishes were an important component of Victorian dining rituals. However, they likely were used to serve overwhelmingly local ingredients and meals. The typical British diet of bread, butter, bacon, and tea was probably replaced with laplap (a kind of starchy pudding made from grated tubers mixed with water and other ingredients and cooked in an earth oven), yams, and local seafood on most occasions. Overall, this remarkable assemblage of ceramics provides a glimpse into the ways that the missionaries performed a civilised Scottish Victorian identity on a daily basis, while also trying to adapt to life on Tanna. Dining rituals would have been internally important to the missionaries and any European visitors they hosted (for example, Mary Matheson was John Geddie’s niece, and Geddie as well as the missionary John Inglis visited; Patterson 1864: 895). Equally important, though, these objects may have offered an opportunity to engage with potential converts as foreign ‘curiosities’ for Melanesian people. Arcadia ware especially may have been an expression of a nostalgic homeland that could be a reminder for their owners, and a storytelling device for cultural others.

Figure 3.28 Ceramic vessel forms, vessels are whiteware with Arcadia transfer-pattern unless otherwise indicated.

Source: James Flexner

The Mission Period at Anuikaraka

There was a chief named Iarisi (also spelled Yarisi and Jarisi in various missionary sources) who was a Yani en Dete in the Kwaraka area in the 1850s. One of Iarisi’s chiefly duties was to organise and lead canoe voyages to exchange with neighbouring Aneityum Island. The Tannese would carry grass skirts, yams, and other goods, and in exchange, would receive taro, mats, red ochre, and other items from Aneityum. When a voyage was being arranged, the men from Kwaraka would go to one of the nearby imwarim, where they would drink kava. While intoxicated, they would use the spirit of the kava to send a message to their counterparts on Aneityum, who would prepare for the arrival of a canoe the next day. Iarisi’s canoe was named Paru. It was kept on a special stone mound with a ramp when not in the water (Figure 3.29).

Figure 3.29 Canoe mound where Paru rested when not at sea, Kwaraka.

Source: James Flexner

On one of his voyages to Aneityum early in the 1850s, Iarisi met some of the Aneityumese converts to Christianity, and possibly one of the European missionaries. Of course, Iarisi would also have heard of and possibly even met the missionaries who had settled in the failed Port Resolution settlement over the course of the 1840s. This trip to Aneityum must have made a positive impression on Iarisi, who decided to open the way for Christianity to come to the Kwaraka area. Likely it was this connection that allowed for the settlement of Aneityumese teachers at Anuikaraka in 1854 (Miller 1978: 37; Murray 1863: 158; Turner 1861: 451). The missionaries assumed that Iarisi’s interest in Christianity was directly related to the new material wealth accumulating among Aneityumese Christians (Gordon, ed. 1863: 127–128).

Excavations in ‘New Kwaraka’

Not all of the residents in Kwaraka were so enthusiastic about the new religion that Iarisi brought with him. The epidemics that had racked the Port Resolution area had been blamed on the missionary-sorcerers, so perhaps there was fear that inviting such people to Kwaraka would only bring the same fate. Perhaps there were also worries that the missionaries were going to take away the magic stones, as had happened on Aneityum (Crook et al. 2015). The loss of nukwei nuk, the stones that made the yams grow, would have been nothing short of disastrous for the district. The tension between the pro- and anti-missionary factions eventually led Iarisi to take a small group of followers across Komaru stream to set up a small settlement bounded by a stone fence in Anuikaraka, remembered in oral traditions as ‘New Kwaraka’ (Figure 3.30). As noted above, this is just on the other side of the division between neteta, in Umairarekarmene.

Figure 3.30 Detail plan of excavation locations in Enclosure 1 (‘New Kwaraka’), Anuikaraka.

Source: James Flexner

When the first European missionaries arrived at Kwamera in 1858, they purchased land not from Iarisi, but from Kapuku and Kati, about 2km, or less than an hour’s walk, to the southwest. Iarisi was remembered in friendly terms by the Mathesons as a supporter of the mission (Patterson 1864: 382). The land record from Inglis, Paton, and Copeland includes a strip of coastal plain stretching from Anuikaraka to the stream of Mimretam (Numretam), next to the current mission station in Kwamera. This massive area was exchanged for red and white calico, handkerchiefs, an axe, and some fishhooks (Vanuatu National Archives Land Record 31 S.I.1). That Iarisi was not included in the purchase suggests the missionaries were largely unaware of who the actual landowners of the area they claimed to have ‘purchased’ were. Land transactions from this era were generally characterised by serious misunderstanding on both sides, partly stemming from practical barriers such as language and literacy, but also more subtle issues that had to do with how ‘ownership’ of land was understood from a cultural perspective (Flexner 2015; Van Trease 1987).

To better understand the material conditions of Iarisi and his fellow early converts, we excavated a series of five 1x2m and 2x2m test units across the enclosure where the Yani en Dete was said to have moved his house in New Kwaraka (Figure 3.30). Stratigraphy across the site consisted of a thin topsoil layer, which overlay a very rocky layer in the western units (TU1, TU3, and TU5), and a sandy loam in the eastern units (TU2 and TU4). In TU2, which was partly excavated to a greater depth, there was another rocky layer underlying the sandy loam, probably evidence for periodic deposition of larger stones during flooding events from Komaru. We can’t completely rule out the possibility at this point of some of the stone being architectural in TU1 and TU3, which had very dense concentration (up to 50 per cent of the sedimentary matrix) associated with high charcoal concentration. However, it appears that the sedimentary layering in the units is primarily a result of periodic flood deposits that created the terrace surrounded by the stone wall (Figure 3.31).

Figure 3.31 Stratigraphic profiles, TU1/2/3/4/5, Enclosure 1 (‘New Kwaraka’), Anuikaraka.

Source: James Flexner

Very few artefacts were recovered from these excavations. The vast majority of collected material consisted of charcoal and burned coconut shells (N=1,239, 200.5g). A small amount of shell was found (two fragments of Cypraea in TU1 and seven of Turbo in TU3). An adze blade fragment with the distal and proximal ends missing was found in the rocky deposit of TU3. Post-contact materials were likewise fairly rare at New Kwaraka. There were unidentifiable fragments of rusted iron in the rocky layer of TU3, but only a small amount (N=102, 18g). These could have come from a single tool or metal vessel. A single piece of amber bottle glass, likely from the 20th century, emerged from TU2. The lone artefact that probably dates to the missionary era is a fragment of a clay pipe bowl from TU4, though it should be noted that clay tobacco pipes remained in use well into the 20th century on Tanna.

Overall, the archaeological remains at New Kwaraka do not indicate huge amounts of European trade goods filtering from the missions to nearby indigenous settlements. This interpretation must be qualified with the fact that periodic flooding at this site may have impacted preservation, and that many trade goods, such as cloth, would be extremely unlikely to preserve archaeologically in this environment. It is also possible that the duration of occupation at the site was brief, with Iarisi and his followers returning to Kwaraka after a few years. Even so, the evidence seems to suggest that Iarisi was not the grasping materialist he was made out to be. Shovel test pits excavated around Anuikaraka likewise did not yield any European artefacts. One possibility is that when the Yani en Dete did receive gifts from the missionaries, these were then re-gifted, and would have circulated across Tanna and into neighbouring islands, which would result in a low density of European artefacts on the site.

This, then, was village life on Tanna in the mid-19th century. The missionary presence was causing certain kinds of internal tensions, and even minor changes in settlement patterns. Introduced materials were filtering into local communities, but apparently in very small amounts. Alliances between local communities and loyalties among chiefly individuals shifted in response to these contacts (Adams 1984). Patterns of everyday life, though, remained largely the same. The yam gardens were still planted, men still gathered at the imwarim to drink kava each afternoon, toka and nao dances were held. People still relied on the yam stones for a good harvest, and feared narak. This was likely true both for people who were holding tight to kastom, and people who were apparently enthusiastic converts to the new religion. Missionaries weren’t so much transforming life on Tanna, as some Tannese people were incorporating missionary things and ideas into kastom.


Previous Next