13

Patterns of connection: The Wanigela shells revisited

Elizabeth Bonshek

Some interesting relics of a by-gone age were found at Collingwood Bay, during some excavations that were being made there, consisting of fragments of pottery and carved conch shells. The examination of these articles by specialists may help ethnologists to determine whether the present population is aboriginal or not. (Monckton 1905:11)

So reported Charles Arthur Whitmore Monckton (1873–1936), the buccaneering resident magistrate of the North East Division, British New Guinea (Lutton 1978, 1986). In 1905 he sent three engraved Conus shells (Figure 13.1), 333 pot sherds and other finds to the British Museum (Oc1905,0209.1–Oc1905,0209.330). These were unearthed during the construction of a new site for the Anglican Mission Station in Wanigela on the north coast of Collingwood Bay. Ten other Conus shells were found: Percy Money, the lay missionary who organised the relocation of the station, collected six; his superior, Rev. Chignell, collected two; Charles Seligman[n],¹ who was collecting for the British Museum, obtained one; and the Viennese ethnologist Rudolf Pöch (see also Howes, Chapter 14, this volume) obtained another, along with human remains, in 1905.

Figure 13.1. Conus shells, collected by C.A.W. Monckton in Wanigela.

Source: © The Trustees of the British Museum (Oc1905, 0209.336–338). Shared under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) licence.

The 1904 excavation represents the first archaeological dig in Papua New Guinea and provided the foundation upon which ethnologists theorised Pacific origins prior to the development of modern archaeology (Spriggs 2013). In 1971, the first of 19 additional Conus shells was found in the northern Massim islands up to 500 km away, bringing the total to 32. These shells are now located in museums and private collections (see Ambrose et al. 2012:114–115). Four of them, including one of Money’s shells (E15596B), were dated to between AD 1101 and AD 1495.²

In an analysis of the designs on the shells, Ambrose et al. (2012) argue that they represent the earliest evidence of the contemporary Massim design tradition on the New Guinea mainland. Were the mounds in Wanigela in which the shells were found created in an outpost of the progenitors of the contemporary Massim tradition? Or do they demonstrate connections through trade up until AD 1465–1495? The presence of spiral motifs and a curvilinear design aesthetic has formed the focus of this interpretation, inspired by the contemporary importance of Conus shells in the kula trade of the Massim. But must the suggestion that the Conus shells represent a continuity with Massim art styles necessarily exclude connection to contemporary art styles of the Wanigela area?

The historical context of Monckton’s finds

Charles Seligmann and Thomas Athol Joyce record that Mr Monckton informed them that the excavated site was ‘an old village site of a forgotten people’ (Seligmann and Joyce 1907:329). How did Monckton know this?

Monckton established the government station at Cape Nelson, 40 km north of Wanigela, in 1901 and reported on all progress in colonial activities to the British New Guinea Administration. The Anglican Mission outpost was established earlier, in 1899, by Reverend Abbot and his party. It was located on the beach near some villages, one of which was stockaded (Chignell 1913:18). Abbot left a year later. Money arrived in 1901 and was joined by Reverend Chignell in 1907. By 1904 Money was corresponding with Robert Etheridge (1846–1920), curator at the Australian Museum, Sydney, and making an ethnological collection for the museum (Bonshek 1989). Monckton referred to Money’s great knowledge of Wanigela and alluded to his ‘manuscript’, which had not yet been published because it was a work in progress (Monckton 1905:34).

Money wrote to Etheridge on 24 September 1904, noting he had a ‘good collection of ancient pottery’ that he would send once he had written his report (Australian Museum Archives AMS9, Letter Received M:85/1904). However, he delayed sending the materials, later explaining to Etheridge that someone had thrown his notes away. He summarised what he remembered and sent the finds, accompanied with his own classification of them. Most of the pottery fragments were picked up at Murin Creek about three miles inland and Money suspected they had washed downriver. He concluded that similar fragments found in the Wanigela mounds on the coast were brought in from Murin, because so few were discovered in the mounds. His only comment on the Conus shells was that the one with a ‘duck’ on it was ‘particularly interesting’ (Australian Museum Archives AMS9, Letter Received M:2/1906; registered into Australian Museum as E15597 and illustrated in Ambrose et al. 2012:116).

Money had enquired among the local population about the excavated material and recorded how some of the enigmatic pottery fragments might have been used (Australian Museum Archives AMS9, Letter Received M:2/1906, Money to Etheridge, 18 November 1905). He also met an old woman who said she was the sole survivor of the clan that had lived on the excavated site. But because she had grown up a mile south, she had never seen the village, nor could she speak her clan’s language. A very old man, who identified his clan, said he had known her father and her paternal clan and that the latter had not spoken Ubir, or any other language then spoken in Wanigela. His own clan had lived adjacent. Money continued:

Both of these old folk agreed on the following points – they had never seen the villages of the [woman’s clan], they had never seen anyone making pottery like the fragments which have been unearthed & had never seen perfect specimens in use. Therefore I conclude that nothing to help us can be gathered from the natives. (Australian Museum Archives AMS9 M:1/1906, Money to Etheridge, 18 November 1905)

Money’s conclusion must have fed into Monckton’s report.

The importance of spirals

Neither Money nor Monckton were ethnologists. It fell to Seligmann, Joyce, Etheridge and Pöch and subsequent theorists to build arguments for the origins of New Guineans in the Pacific, based on the comparison of spiral motifs on shells, potsherds, lime spatulas and stone monoliths (Spriggs 2013). Spirals continued to attract attention in the 1970s, with archaeologists suggesting connections to Dong Son motifs from southern China and northern Vietnam dating to 2,000 years ago. In his overview, Spriggs summarised: does the prehistoric culture of Collingwood Bay hold the key to the immediate origins of the art styles of the Massim and confirm its Dong Son inspiration? (Spriggs 2013:9–10).

Between 1967 and 1969 Brian Egloff excavated new sites at Wanigela, establishing dates of between 1,000 to 500 years BP for the pottery fragments he dug up near the site of Money’s excavations. However, he did not find any shells. Together with the dates for similar pottery found by Vincent Kwebu in the Massim (Spriggs 2013:10), the Conus shells were estimated, by association, to be 1,500 to 1,000 years BP. Extending beyond the immediate region, Meredith Wilson (in Spriggs 2013:10) has suggested that three of the Conus shells and an incised monolith found in Goodenough Bay are most closely linked to a spiral-based tradition from East New Britain and New Ireland engraved rock art, forming a part of a widespread Austronesian Engraved Style spreading into Milne Bay Province by 2,000 years BP (Spriggs 2013:10). The discovery of additional incised Conus shells in New Caledonia, in Lapita contexts dating to some 3,000 years ago, pushes their antiquity further back. Again Spriggs asks: Are the shells direct descendants of Lapita practice? Or are they a reinvention of such practices to reproduce Southeast Asian Dong Son designs on nonmetal artefacts?

Theorising Wanigela designs

The dates associated with the four shells place them within Phase 2 (Expansion Phase, AD 950–1450) of the three-phase development proposed for Massim prehistory. However, the Wanigela shell fell closer to the date range associated with Phase 3, known as the Refuge Phase (AD 1450–1850) (Ambrose et al. 2012:128–129).

During Phases 1 and 2 the people of the northern Massim were using pots made of clay originating from Wanigela and Goodenough Island (Ambrose et al. 2012:129), while their stone, used in burial practices, originated from Woodlark Island. In Phase 2, the people of the Trobriand Islands stopped importing stone. In Phase 3 evidence of trade connections between the New Guinea mainland and the islands of the northern Massim disappeared and the latter turned southwards for inter-island trade and engagement. How do the designs on the shells relate to the dates established?

Ambrose et al.’s (2012) stylistic analysis of form and surface design used contemporary Massim designs as a reference point. Large Conus shells are available in the Massim: they are important in the production of mwali (armshells) in the regional exchange known as the kula and the spiral is used in contemporary Massim woodcarving. Thus, contemporary Massim carving motifs were used as defining characteristics of the prehistoric group.

The designs were classified as ‘framing’, characteristic of the Massim, or ‘all over decoration’, characteristic of Wanigela (2012:120, Fig. 10). Five motifs – circles/spirals, ‘inward scrolls’, bird figures (2012:116, Fig. 4e), concentric circles (2012:119, Fig. 7c) and ‘face’ motifs (2012:116, Fig. 4c) – were identified as diagnostic of the contemporary Massim style.

Shells with rectilinear designs were considered untypical, or aberrant to contemporary Massim style. These might be designs imitative of the Collingwood Bay style and together with the later date for the Wanigela shells, these might represent the process of disconnection from the south.

Collingwood Bay style

If we open up an analysis of Collingwood Bay style to include different object types made by women and men, an expansion upon rectilinear/geometric motifs emerges. Continuing the use of contemporary analogy, the upturned dish in Figure 13.2 and the woman’s barkcloth in Figure 13.3 show asymmetrical scrolls, meandering curvilinear style and geometric elements coexisting.

Meandering lines, hooks, scrolls, concentric circles and ‘S’ shapes occur on pots and barkcloth in Percy Money’s collection (Bonshek 1989:114, 116, 120, 122–123, 138, 176 and 178).³ Frank Hurley’s (1924:110) photograph of public mourning in Wanigela depicts two widows, one hidden beneath a cloth adorned with meandering designs and a second under a cloth with what is probably a crocodile motif on it.

Figure 13.2. Upturned dish.

Source: Author’s collection (acquired 2003).

Figure 13.3. Woman’s barkcloth, acquired by Percy Money.

Source: Photo by Ric Bolzan, AMS/M1711/4. Courtesy of Australian Museum Archives (E13157).

The use of recognisable animal forms (such as the bird on the Wanigela shell, E15597) appears twice in Money’s collection (a crocodile, Bonshek 1989:126, 168). The woman’s barkcloth collected by Rev. Abbot in 1899–1900 (Figure 13.4) shows a reptile together with variations on a theme of concentric circles:⁴ squared-off circles divided into four, star-like arrangements, together with hooks, single ‘V’ and double ‘V’ shapes.

The detail of a design on a man’s barkcloth (Figure 13.5) shows the typical structured, repeating segments (or ‘framing’) used on many cloths (Figure 13.6 and also Figure 13.3). Further, Money identified lines that informants told him were a ‘path’ and these effectively separate areas of design on pots too (Bonshek 1989:204, 206). This is consistent with the division of the surface into three, illustrated by the northern Massim shell known as ‘Imdeduya’ (Ambrose et al. 2012:119, fig 9c).

The designs on the Conus shells resonate with contemporary designs in the Massim but they also have elements that resonate with contemporary Collingwood Bay, which is not restricted to one visual aesthetic.

Figure 13.4. Registration slip for a woman’s barkcloth, acquired by Rev. Abbot.

Source: © The Trustees of the British Museum (Oc1901,0723.44). Shared under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) licence.

Figure 13.5. Registration slip for a man’s bark loincloth, collected by Rev. Abbott.

Source: © The Trustees of the British Museum (Oc. 1901, 0723.35). Shared under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) licence.

Figure 13.6. A man’s barkcloth, acquired by Percy Money.

Source: Photo by Ric Bolzan, AMS391/M01754/04. Courtesy of Australian Museum Archives (E16335).

Motifs, connection and movement

Money named many of the motifs on the objects he collected, and recorded clans and their language association (Bonshek 1989:203–208). His notes refer to Ubir, Oyan and Onjob clans in Wanigela as well as the Maisin of Uiaku living to the south.

Several of the motifs are clan designs: baifafaro in Ubir (Bonshek 2008). Baifafaro also include cultural practices particular to a clan. Other designs are not prescriptive and are placed on objects used for local and regional trade or exchange.

The exchange of pots made by women was (and remains) central to this exchange network. Money noted exchanges for barkcloth made by the neighbouring Maisin women in Uiaku. This exchange occurred despite the ability of Wanigela women to make barkcloth (which, in 2001–03, they say was stolen from them through sorcery). Designs circulated within and beyond Wanigela regardless of different languages and clan affiliations (43.5 per cent of the designs on the pots and cloth in the Money collection overlap, Bonshek 1989:83). Motifs cut across different media, clans, languages and villages. This concurrence reflects a history of connection between groups, not only within Wanigela but throughout the region.

The connections between people are extensive. No longer common knowledge, up until World War II shell necklaces (known in Ubir as nunug) were acquired by some clans through long-distance voyages following traditional paths (known in Ubir as eta), facilitated by a series of trade partners located at various villages along the south coast. Each clan had a series of trading partners: some clans looked to the south, while others looked to the north.

This complexity is obscure in the historical records, although glimpses emerge in the administration’s reports on tribal warfare in the area. Money signposts connections between groups in his observations on the exchanges between Maisin and Wanigela. But baifafaro were also gifted to secure alliances and provide protection. In contemporary Wanigela there are some designs that several clans have the right to use and these are evidence of connections and engagements (Bonshek 2008). Designs do not necessarily indicate membership of a group via linear descent or language affiliation but diffuse across boundaries via social connection and political negotiation.

Today Wanigela is home to 51 patrilineal clans, paired as senior and junior brothers, belonging to four language groups. They are connected by marriage and alliances that formerly governed raiding and warfare (Bonshek 2005). Not all the clans that have ever lived in Wanigela continue there today. Knowledgeable people can recount up to seven generations back, suggesting that all clans present today were present in Money’s time.

The clans moved into Collingwood Bay at different times, arriving from inland and from the north and south, some by foot and some by canoe. Each has their own account of migration into the bay (Bonshek 2005). Mackenzie Asor (1974), an Ubir, Sabarar clansman, recounted the story of the culture hero Dararuk and the movement of the clans in mythic time. As a boy Dararuk became so unhappy at the death of his pet that he cried inconsolably. So great was his grief that the clans of his village left in despair, heading off in all directions, but especially to Tanam (Cape Nelson) and Gorof (Cape Vogel). Abandoned and exposed to sorcery in his coastal village in Collingwood Bay, Dararuk left and headed towards Goodenough Island – at that time joined to the mainland – where he was adopted by a kind old lady and they both lived in a tree, at Woyar (Goodenough). When he grew up, he was so handsome that he attracted the attention of all women who saw him. Motivated by jealousy, their husbands banded together to chop down the tree in which he lived. Dararuk responded by distributing among the men all the things that distinguished the clans. Dararuk and his grandmother bored two holes in the tree where they hid as the tree was chopped and then burned. The two branches that protected them flew away. Eventually Dararuk emerged in the river and lived among the people in secret until they had forgotten the earlier attack.

Asor interprets the story of Dararuk’s distribution of distinctive customs to the clans as their dispersal from Goodenough Bay to Collingwood Bay and identifies two movements: one outwards from a site in Collingwood Bay and another suggesting a movement back in from Goodenough. In the 1960s Margaret Stephens recorded that the Ubir migrated from the Cape Vogel area, the Oyan from Uwe and the Onjob from Mt Victoria (1974:33).

During my ethnographic present, migration stories were not discussed publicly and tension surrounded the question of who among the clans arrived in Collingwood Bay first (Bonshek 2008). Perhaps this was also the case during Money’s time.

Some accounts (Bonshek 2005) say that an argument between the clans caused the residents of the stockaded village to disband. Perhaps the distribution of the mission’s buildings between the villages of Rainu and Oreresan was an attempt to resolve rivalry between the two groups. Chignell (1911:19–20) recounts that he and the South Sea Islander staff settled in Oreresan (an Oyan village), and Money and schoolchildren (and the goats) resided in Rainu (an Ubir village). At that time Old Komabun (an Ubir village) was already established half a mile north across the river, while the Onjob lived one mile inland at Aieram and the Aisor-speaking people lived at Murin and Naukwate.

In hindsight, what can be made of the information that the mission’s new site was the prior residence of a known clan – a site Money concluded had been long abandoned?

The story of the extinction of a descent line frames Money’s interpretation of disconnection. But it could be argued that the elderly woman’s memory (and that of the elderly man corroborating her story) constituted a connection with the people who used to live on the new mission site. And what had been the relationship, or alliance, that must have existed between the two clans that had lived in close proximity? If, by 1904, only one clan remained, had an alliance broken down? Had the clans been brothers? Had one chosen to depart? Had one been made to depart, perhaps through sorcery? By 1904 the Tribal Wars had not yet ceased: fighting in the region ended with the ‘Breaking of the Spear’ held on 12 March 1905 (Bonshek 2005:82). It could be that at the time Money made his inquiries, he also unwittingly recorded tensions between clans, encountering both memory and active forgetting (Bonshek 2008:93). It seems unlikely that clans who have genealogies of six to seven generations would not know something of the former residents, especially with one still living among them.

Who were the prior residents of the new mission site? How long ago had the old woman’s family left the site? And why? How long had they lived there? Did they engrave the Conus shells or bring them in through exchange networks, or did they bring them with them as they migrated into Collingwood Bay? Do the dates established for the Wanigela Conus shells place them beyond the reach of social memory in 1904?

Objects highlighted in this chapter were on display at the British Museum from March 2020.

References

Ambrose, W., F. Petchey, P. Swadling, H. Beran, E. Bonshek, K. Szabo, S. Bickler and G. Summerhayes 2012 Engraved prehistoric Conus shell valuables from southeastern Papua New Guinea: Their antiquity, motifs and distribution. Archaeology in Oceania 47:113–132. doi.org/10.1002/j.1834-4453.2012.tb00124.x.

Asor, M. 1974 An oral tradition from Wanigela in the Northern District. Oral History 3:2–11.

Australian Museum Archives. AMS9, Letter Received M:1/1906, P. Money to R. Etheridge, 18 November 1905.

Australian Museum Archives. AMS9, Letter Received M:2/1906, P. Money to R. Etheridge, 18 November 1905.

Australian Museum Archives. AMS9, Letter Received M:85/1904, P. Money to R. Etheridge 24 September 1904.

Barker, J. 2008 Ancestral lines: The Maisin of Papua New Guinea and the fate of the rainforest. Toronto: University of Toronto Press.

Beran, H. and E. Aguirre 2009 The art of Oro Province: A preliminary typology. Sydney: The Oceanic Art Society.

Bonshek, E. 1989 Money, pots and patterns: The Percy Money collection of bark cloth and pottery held at the Australian Museum. Unpublished Master of Arts thesis, Department of Anthropology and Sociology, University of Queensland, St Lucia.

Bonshek, E. 2005 The struggle for Wanigela: Representing social space in a rural community in Collingwood Bay, Oro Province, Papua New Guinea. Unpublished PhD thesis, The Australian National University, Canberra. doi.org/10.25911/5d5141c73a1e1.

Bonshek, E. 2008 When speaking is a risky business: Understanding silence and interpreting the power of the past in Wanigela, Oro Province, Papua New Guinea. Journal of Material Culture 13:85–105. doi.org/10.1177/1359183507086220.

Chignell, A. 1911 An outpost in Papua. London: Smith Elder.

Chignell, A. 1913 Twenty-one years in Papua: A history of the English Church Mission in New Guinea, 1891–1912. London: A.R. Mowbray & Co.

Hermkens, A. 2013 Engendering objects: Dynamics of barkcloth and gender among the Maisin of Papua New Guinea. Leiden: Sidestone Press.

Hurley, F. 1924 Pearls and savages: Adventures in the air, on land and sea in New Guinea. New York: G. P. Putnam’s Sons.

Lutton, N. 1978 C. A. W. Monckton: Reprobate magistrate, Papua New Guinea. In J. Griffin (ed.), Papua New Guinea portraits: The expatriate experience, pp. 48–74. Canberra: Australian National University Press.

Lutton, N. 1986 Monckton, Charles Arthur Whitmore (1873–1936). Australian Dictionary of Biography, Retrieved 6 March 2019 from adb.anu.edu.au/biography/monckton-charles-arthur-whitmore-7619/text13315. (First published in Australian Dictionary of Biography, Volume 10, 1986.)

Monckton, C.A.W. 1905 Annual report on British New Guinea for the year ending 30 June 1904. Victoria: Government of the Commonwealth.

Myers, C.S 1941 Charles Gabriel Seligman 1873–1940. Biographical Memoirs of Fellows of the Royal Society 3(10):627–646. doi.org/10.1098/rsbm.1941.0026.

Seligmann, C.G. and T.A. Joyce 1907 On prehistoric objects in British New Guinea. In H. Balfour, W.H.R. Rivers, R.R. Marett and N.W. Thomas (eds), Anthropological essays presented to Edward Burnett Tylor in honour of his 75th birthday Oct. 2 1907, pp. 325–341. Oxford: Clarendon Press.

Spriggs, M. 2013 Three carved stone shells from Collingwood Bay. In N. Thomas, L. Bolton, E. Bonshek, J. Adams and B. Burt (eds), Melanesia: Art and encounter, pp. 7–10. London: British Museum Press.

Stephens, M. 1974 With bar sinister on his chicken feathers: A study of the integration of kin terminology with social structure in Wanigela, Northern District, Papua New Guinea. Unpublished PhD thesis, University of North Carolina at Chapel Hill, Ann Arbor.

---