Do kula canoes of the Massim region of Papua New Guinea have a bow, a stern, and prowboards?

Harry Beran, independent researcher
Two types of outrigger canoes are used in the kula. This exchange system of shell valuables in the Massim region (Milne Bay Province) of Papua New Guinea is described by Bronislaw Malinowski in Argonauts of the Western Pacific (1932) and by Leach and Leach in The Kula (1983).
Western writers on kula canoes tend to write of the bow and stern of these canoes and to use the term prowboards for their wavesplitters and even their washboards, irrespective of whether they are at the canoe’s so-called bow or stern. This terminology obscures the structure of the canoes, which is quite different from most Western boats and ships. Below, I describe the structure of kula canoes and argue for abandoning the bow/stern/prowboard terminology in favour of a terminology based on how the Massim think about the canoes.
The issue may be of interest to writers on Massim art and to the curators of the many museums that have kula canoe boards. The issue also arises regarding Oceanic canoes from beyond the Massim region with a single outrigger and designed to have either end pointing ahead.
The larger of the two types of kula canoes is called nagega or anageg in the languages of the northern part of the Massim region and made primarily in Gawa and Kwaiwata islands, from where they are traded to Woodlark and in turn to the southern Massim region (Fig. 1).
(Fig. 1. Gawa-built nagega canoe sitting on a beach, which shows its highly curved dug-out keel and three planks. Photograph taken by Clare Harding in 1983. Reproduced courtesy of Clare Harding).
The smaller of the two types of kula canoes is called masawa in the Trobriands and Iwa Island, epoi in Dobu Island, and tadobu in some places because the type is believed to have originated in Dobu (Fig. 2). This type is made mainly in the places just mentioned.
(Fig. 2. A canoe of the masawa/epoi type, sitting on land, which shows its curved dug-out keel and two planks. Photograph by Templeton Crocker, taken in 1930, probably in the Trobriand Islands or Dobu. C. Templeton Crocker papers, California Academy of Sciences Archive. Courtesy of the Fine Arts Museums of San Francisco, de Young).
The structure of the main components of the two types of kula canoes is the same. They have a single outrigger float, which must always be windwards, as the wind in the sail cannot easily lift the float out of the water but can easily submerge it and capsize the boat (Malinowski 1932: 110; Campbell 2002: 73). The canoe is, therefore, symmetrical, being designed to sail equally well whichever end points forward. The basic shape of the two ends of the canoe’s dug-out keel, its two wavesplitters (tabuya), and two washboards (lagim) is the same. It has a huge steering paddle, which is carried to, and used at, whichever end is at the back.
Nevertheless, the Massim distinguish between the branchend and root-end of the dug-out keel, cut from a single tree, to which two or three planks are added to make a canoe capable of carrying ten and more men on open-ocean voyages. Trobrianders call the branch-end dogina and the root end uuna or uula (Campbell 2002: 73-4; Lawton 2004) (1). Woodlark Islanders call the former dabwen, the latter wowun (Fred Damon, pers. comm., Nov. 23rd, 2010) (2). Kitava Islanders have separate names for the two wavesplitters (tabuya) of the canoe, calling that for the branch-end tabudogina and that for the root-end tabuvaura (Scoditti 1990: 71, 136, 140) or tubuuula (Hallinan 1972).
Although the basic shape of the wavesplitters and washboards for the canoe’s two ends is the same, there are small differences between them.
According to Shirley Campbell (2002: 74) and the late Chief Narubutau (1979, Items 7 and 10), the root-end wavesplitters of masawa are perforated (have openwork carving) along the bottom, the branch-end ones are not (3). Campbell (2002: 73-4) notes a second difference: the shape of the ‘nose’ of the wavesplitters. These differences are shown in Figs 3a and 3b.
(Fig. 3, left. Dogina (branch-end) tabuya. Fig. 4, right. Uuna (root-end) tabuya. After Campbell (2002: 73). The arrows and the word ‘nose’ have been added).

Interestingly, the two designs can easily be distinguished among the masawa/epoi wavesplitters George Brown collected between 1890 and 1905; three of them are recorded as having been collected in Dobu, one in Kiriwina, and three in ‘New Guinea’. Two of them correspond to Campbell’s dogina tabuya (H137332 and H137334), and five to her uuna tabuya (H137330, H137331, H137333, H137336, and H138762) (4). Two tabuya carved in Kitava and illustrated in Scoditti (1990: 389, Plates 58 and 61) also conform to the distinction Campbell makes.
The branch-end and root-end wavesplitters of nagega also differ slightly in the motifs carved on them (pers. obs.) The central part of the upper section of the former is usually plain (Fig. 5) or, if covered in motifs, they do not include a prominent vertical hook-shaped scroll; root-end wavesplitters do display this scroll (Fig. 7). The carvings attached to the top of the wavesplitters (called sakusaku in Kilivila, the language of the Trobriands, Gawa, and Kwaiawata, and maan in Muyuw, the language of Woodlark) also differ; that for the branch-end wavesplitter is symmetrical (Fig. 6), that for the root-end asymmetrical (Fig. 8) (5).
(Fig. 5. Wavesplitter (tabuya) for the branch-end of a nagega. John and Marcia Friede (Jolika) Collection, ex author’s collection (HB 856). Carved by Urisaku of Gawa and collected in Gawa, Distinguished from wavesplitters for the root-end by having a comparatively plain upper section, below the peg to which the sakusaku is attached. Photograph by John Bigelow Taylor, courtesy of John Friede).
(Fig. 6. Sakusaku of symmetrical design for the branchend wavesplitter of a nagega. John and Marcia Friede (Jolika) Collection, ex author’s collection (HB 257). Collected in Egum Atoll but carved in Gawa. Photograph by John Bigelow Taylor, courtesy of John Friede).
(Fig. 7. Wavesplitter (tabuya) for the root-end of a nagega. John and Marcia Friede (Jolika) Collection, ex author’s collection (HB 443). Collected in Boagis Village, Woodlark but probably carved in Kwaiawata. Distinguished from the wavesplitter for the branch-end of the canoe by having an upper section (below the peg for the sakusaku) fully covered in motifs, always including a large vertical hook-shaped scroll. Photograph by John Bigelow Taylor, courtesy of John Friede).
(Fig. 8. Asymmetrical sakusaku for the root-end of a nagega. John and Marcia Friede (Jolika) Collection, ex author’s collection (HB 258). Collected in Egum Atoll but carved in Gawa. (The sakusaku is placed on the tabuya with the section trailing down from its main part toward the washboard; that is, rotated horizontally 180 degrees from the way it has been photographed; cf. Munn 1976: 146.) Photograph by John Bigelow Taylor, courtesy of John Friede).
The washboards for the masawa’s two ends differ in that the wing on the outrigger-float side is normally bigger than that on the other side (Campbell 2002: 73). In his essay on fourteen washboards from masawa in the collection of the Papua New Guinea National Museum and Art Gallery, Narubutau (1975) identifies some as ‘prow’ boards (Figs 9, 10, and 11) and others as ‘stern’ boards (Figs 1, 4, 6, 7, 8, 12, 13, and 14). He was able to do this because kula canoes are built so that the outrigger float is always on the left of the hull when the canoe’s branch-end is pointing forward (6). Therefore, if the washboard’s right wing is the larger one, it is from the branchend (‘prow’) of the canoe, if the left wing is the larger one, it is from the canoe’s root-end (‘stern’).
The washboards of nagega also tend to have the wing on the float side larger than the other, but on many boards the difference is small, on others even absent.
Western writers render dogina and dabwen (top of the tree) as front or bow of the canoe and uuna and wowun (base of the tree) as back or stern (Scoditti 1990 and Damon 2008: 131, note 10). However, in contradiction to this terminology, some writers use the term prowboard for the bow and stern wavesplitter, and sometimes also for both washboards (Malinowski 1932: 110, 112, 113 and caption for Plate XXIV; Munn 1977: 47, note 22; Campbell 2002: 73; and Damon 2008: 132). Douglas Newton (1975, figs 49, 50, 52, 53, 55, 60, and 61) calls all the masawa and nagega wavesplitters he illustrates ‘prow carvings’ without considering at which end of the canoe they were used.
Haddon and Hornell (1975: 240-81), who provided a survey of Massim canoes, were not aware of the branch-end/root-end distinction the Massim make. They write of the ‘bow and . . . stern’ of Massim canoes (ibid.: 276), and of their fore and aft end-erections (wavesplitters) and breakwaters (washboards). However, they avoid the terms prow and prowboard, except once (ibid.: 264) when paraphrasing the caption of a published wavesplitter they refer to.
Two of the writers mentioned seem to express unease about the bow/stern terminology. Giancarlo Scoditti (1990: 71) writes that the translation of the Kitava terms for the two ends of the canoe as bow and stern is ‘rather conventional’ as the canoe’s two ends ‘are equal and symmetrical’. And Campbell (2002: 73) observes that because the float of kula canoes must always be windwards: “the bow and stern of these boats change according to the direction of the wind. This means that the two ends of the outrigger canoe must function as both bow and stern”.
The bow/stern terminology was created for Western ships designed with a permanent front, more or less pointed, and a permanent back, more or less rounded or blunt, and with a rudder fixed to the back. Such ships can reverse but are not designed to move forward with the blunt end and rudder in front.
Further research is required to establish whether the Massim conceptualise the branch-end of the kula canoe as its front and the root-end as its back. There are three hints that this may be so. First, Narubutau (1975) speaks of the ‘prow’ and ‘stern’ of canoes. However, he spoke little English and these terms almost certainly came from his translator, perhaps his nephew John Kasaipwalova, in consultation with Ulli Beier, the editor of the journal in which the essay appeared. Second, Campbell (2002: 73, 130) records that Vakutans consider the dogina (branch) end of the canoe more important than the uuna (root) end. For example, the dogina wavesplitter ‘is the first board placed on the masawa . . . [and it] lands first on the beach’ when the canoe arrives at the place of the kula partners. Thirdly, photographs indicate that kula canoes are normally beached and placed in sheds with the branch-end towards the sea.
If the Massim do regard the canoe’s branch-end as its front, and its root-end as its back, there may be a case for calling the branch-end, including wavesplitter and washboard, its bow or prow and the root-end its stern. However, this terminology would still be misleading as it would divert attention from the fact that these canoes are designed to sail equally efficiently with either end at the front. In my future writings on kula canoes I will, therefore, avoid the terms bow, stern, and prowboard and speak instead of the branch-end and root-end of canoes and of branch-end and rootend wavesplitters and washboards (7).
– Campbell, Shirley F. 2002. The Art of Kula. Oxford: Berg.
– Damon, Frederick H. 2008. ‘On the Ideas of a Boat: From Forest Patches to Cybernatic Structures in the Outrigger Sailing Craft of the Eastern Kula Ring, Papua New Guinea’. In Beyond the Horizon: Essays on Myth, History, Travel and Society, edited by Clifford Sather and Timo Kaartinen. Helsinki: Finnish Literature Society.
– Haddon, A.C. and James Hornell. 1975 (1936-8). Canoes of Oceania. Honolulu: Bishop Museum Press.
– Hallinan, J.P. 1972. (ed.) Kula Canoe ‘Sopikarin’: Ethnic and Historical Notes. Typescript to accompany this canoe bought by the South Australian Museum.
– Lawton, Ralph. 2002. Dictionary Kiriwina to English. Canberra: published privately in six copies.
– Lawton, Ralph (editor, but not named in the typescript). 2004. Masawa: A Canoe for Open Ocean Sailing from Kiriwina, Trobriand Islands, PNG. Typescript to accompany a model masawa canoe given to the Australian National University by Nelson Toposona, Daniel Fellows, and Lepani Ahab.
– Leach, Jerry W. and Edmund Leach (eds). 1983. The Kula: New Perspectives on Massim Exchange. Cambridge: Univ. Press.
– Malinowski, Bronislaw. 1932 (first impression 1922). Argonauts of the Western Pacific. London: George Routledge & Sons.
– Malinowski, Bronislaw. 1966 (first edition 1935). Coral Gardens and Their Magic, 2 vols. 2nd edition with an Introduction by Edmund R. Leach. London: George Allen & Unwin.
– Malnic, Jutta with John Kasaipwalova. 1998. Kula: Myth and Magic in the Trobriand Islands. Wahroonga, NSW: Cowrie Books.
– Mosuwadoga, Geoffrey. 2006. ‘Art and Artistic Training in Boyowa Island, Trobriand Archipelago, Papua New Guinea.’ Pacific Arts N. S., vol. 1, pp. 12-25.
– Munn, Nancy. 1977. ‘The Spatiotemporal transformations of Gawa canoes.’ Journal de la Societe des Oceanistes, vol. 54-55, pp. 39-52.
– Munn, Nancy D. 1986. The Fame of Gawa: A Symbolic Study of Value Transformation in a Massim (Papua New Guinea) Society. Cambridge: Univ. Press.
– Chief Narubutal [that is, Narubutau]. 1975. ‘Trobriand Canoe Prows: Fourteen Pieces from the National Collection in the Papua New Guinea Museum.’ Gigibori, vol. 2, nr 1, pp. 1-14.
– Chief Narubutau. 1979. ‘Eleven Canoe Prows from the Trobriand Islands.’ Gigibori, vol. 4, pp. 40-6.
– Newton, Douglas. 1975. Massim: Art of the Massim Area, New Guinea. New York: The Museum of Primitive Art.
– Scoditti, Giancarlo M.G. 1990. Kitawa: A Linguistic and Aesthetic Analysis of Visual Art in Melanesia. Berlin: Mouton de Gruyter.
1. Ralph Lawton (2002) offers a number of meanings of ‘dogina’, including ‘top of the tree’ and ‘end part of magic formula’ (but not ‘front’), and a number of meanings of ‘uula’, including ‘base
of the tree’ and ‘the beginning (of process)’ (but not ‘back’); cf. Malinowski (1966, vol. II: 92, 96-7).
2. The contrast between the Kilivila and Muyuw terminology is not as great as it appears. Malinowski (1966: 92) records ‘dogina’ as ‘tip’ [of plant], ‘dabwana’ as ‘’top of . . . plant’, and ‘u’ula’ as ‘basis of trunk’. The ‘dabwana’ and ‘uula’/‘uuna’ of Kilivila seem to be cognates of the ‘dabwen’ and ‘wowun’ of Muyuw.
3. The openwork carving takes many forms: it can be a row of circles, as in the drawing in Fig. 2, a row of scrolls, or a mere horizontal slit, as on the wavesplitter in Campbell’s book (2002,
plate 9).
4. The catalogue of the George Brown Collection in the National Museum of Ethnology, Osaka, is online with illustrations.
5. There are a few photographs of canoes where the wavesplitters do not conform to the usual different designs for the two canoeends. A masawa example, built by Narubutau a few decades ago and called Toilamlaguyau, was photographed a number of times by Jutta Malnic. Her photograph of the whole canoe is shown in Malnic with Kasaipwalova (1998: 42-3), that of what appears to be the canoe’s branch-end in Mosuwadoga (2006: 14 and on the cover of the journal the essay appeared in), and that of what seems to be the canoe’s root-end in Malnic with Kasaipwalova (op. cit.: 12). Both wavesplitters seem to be of the design normally used at the masawa’s root-end.
6. This is not stated explicitly in the published literature on kula canoes. However, it is recorded in an unpublished note on masawa terminology, edited by Ralph Lawton (2004), which
states that the person sitting at the ‘front end, bows’ of the canoe, presumably looking ahead, has the outrigger on his left. It is also presupposed by Campbell (2002: 73) when she writes that
because of the prevailing wind direction, when Vakutans sail to Dobu their canoes have the dogina (branch) end pointing ahead and the float on the left, but when they sail to Kitava their canoes have the uuna (root) end in front and the float on the right. The two masawa kula canoes held by the South Australian Museum, Adelaide, and the Pigorini Museum, Rome, have the structure mentioned.
7. I am grateful to Barry Craig, Fred Damon, and Ralph Lawton for comments on the topic of this essay.

Leave a Reply