Previous Next

Degei’s Descendants: Spirits, Place and People in Pre-Cession Fiji

6. The Yavusa: The Ideal and the Reality

The basic Fijian polity or yavusa is ideally taken to be a group of people (both male and female) claiming descent usually in the male line from a common mythological ancestral spirit, and associated with commonly held land. To the extent that a person can ideally belong to only one yavusa, the yavusa is exclusive and members are referred to as itaukei in contrast to non-members who are referred to as vulagi.

The yavusa ideally comprises sub-descent groups, referred to currently as mataqali in all recognised communalects of which I am aware. These have a recognised order of seniority based ideally on the mythical birth order of sons of the common ancestor. Each mataqali comprises a number of sub-divisions having an order of seniority again based ideally on the mythical birth order of grandsons of the common ancestor and usually referred to as itokatoka. To this extent, the yavusa is socio-politically hierarchical. This summary of the structure of Fijian polities is based on what has become an idealised model for Fijian socio-political structure, as described by Roth (1953:58), France (1969:166), Nayacakalou (1975:16) and Ravuvu (1987:16).

Identity of yavusa

To maintain the characteristic of exclusiveness, each yavusa has a number of features of identity which distinguish it from other yavusa. They reflect the three crucial and interacting elements of the ideological concept of traditional Fijian society (vanua 1). These elements are spirits, people and land, and they form the basis for investigating, understanding and analysing Fijian socio-political groups.


Original ancestral spirits

Each yavusa recognises a common original ancestral spirit (kalou vu in the East; nitu in the West), as well as other spirits. Such an ancestral spirit usually has the following features of identification, although some features may now be forgotten:

  • • name;
  • • place of origin and place(s) of settlement;
  • • place where presentations are made to the spirit;
  • • place of communication between the spirit and the living;
  • • waqawaqa or object (usually animal) of manifestation (not embodiment);
  • • name and place of origin of spouse;

In Rakiraki, Drilo Dadavanua was the ancestral spirit of the Malake yavusa. He came on the instructions of the culture hero Degei from the Nakauvadra Mountains to Navatu where he married Likumasei, a woman from the offshore Navatu islet of Cubu. They moved to Malake Island in order to supply seka or crabs to Degei. Later the Malake people became the turtle fishers for Degei. Dadavanua and Likumasei settled on the hilltop at Uluidrilo (the site is marked by a ring of stones) before moving to the coast to the swampy area of Lomalake. Here they settled at Navuniviavia, marked by a rectangular earth mound for the house where the bete would communicate with Dadavanua. Near the mound is a flat area where presentations were made to Dadavanua. Dadavanua's waqawaqa is a gata or snake living in a werewere tree at Navuniviavia.

Other spirits associated with a yavusa

  • • Spirits of war or kalou ni valu (nitu ni valu in the West)

Limasa is the nitu ni valu of the Kovacaki people of Nadi. His manifestation is an owl. His place is at Naviqwa, marked by an earth mound in a grove of trees. He has a holiday place nearby, identified by two earth mounds. Presentations were made to him at Nasavusavu, where a dolmen-like stone structure comprising a rectangular capstone and supporting uprights is surrounded by low earth mounds in a grove. The site, currently fenced and kept meticulously clean, is protected by the spirit of a brown dog. The present bete lived in a modest Fijian-style house, the only one in the village of Navoci, near Namotomoto, beside the main road into Nadi town. During my visit in 2003, this house was being dismantled after a fire, and a decision had not yet been taken about its replacement. The building where the bete used to communicate with Limasa was destroyed when the main road was widened.

  • • Spirits associated with prosperity

The Vunatoto people of Nawaka respect a series of spirits known as digiwai. These include the Digiwai ni Marawa or spirit of the gardens, and the Digiwai ni Ika and the Digiwai ni Manumanu responsible respectively for success in fishing or hunting. Qeti or presentations were made to the digiwai through the bete in the beto. If they were not made correctly, the gardens would suffer, or when people went fishing or hunting, a fisherman was liable to be bitten by an eel or shark; or a hunter might either fail to catch a pig, or be bitten by one.

  • • Spirit defenders (ba or sasabai) with associated places and manifestations

Bituwewe was the ba, the tuwawa or defending champion of the Leweiwavuwavu, the first yavusa to settle in the Sabeto valley south of the Tualeita range. He was a tall man whose head was at Unuineivua, his stomach at Neisauniwaqa, his legs at Nasaqa and his feet at Conua where they may still be seen as rock formations on top of the Tualeita. His manifestation was a ga ni vatu or falcon named Voili.

  • • Spirits ensuring good behaviour of members of a polity

Losausauega was a female spirit (her name includes the female honorary prefix lo-) whose place was on the right bank of the river flowing past Vunitogoloa, village of the Naqilaqila yavusa. If any one living there misbehaves, she appears in the form of a spider (suasauega means ‘spider’ in the Rakiraki communalect). A female spirit, LoRubaroba, associated with the Nairara people of Nakorokula, has a place at Vunitawa. Her face and body are swollen up. If anyone misbehaves, she will appear and spit at the person, whose face and body then become covered in spots.

  • • Cult spirits, such as rere (land spirits) and luve ni wai (water spirits)

The young men of the inland village of Nawaka saw a spirit in places around the village and set up a cult based on this spirit which was known to be a rere (with two long vowels). Because they ministered to this rere, the people of Nawaka had successes in war. A man of Ba living on the island of Nananu i cake off Rakiraki told me how he was approached by luve ni wai from a nearby islet with offers of assistance if required. He already had his own spirit from Ba, called Lewatumomo, who would come to his assistance if required. He rejected the offer from the luve ni wai, because he feared that if he were to accept their offer, he would then be in a position of obligation to the spirits. This he did not want. It is difficult to obtain detailed information nowadays about the rere or the luve ni wai.


Each yavusa has a name. Each mataqali has a name; and each mataqali has a socio-political status identified by such a term as turaga or chief; mata ni vanua or ceremonial official; or bete or communicator with the ancestral spirit. Each itokatoka has a name.

A yavusa usually had features of identification such as:

  • • the title of the paramount ‘chief’, such as Tu Navitilevu in Rakiraki, or Momo Levu in Vuda;
  • • a housemound (yavu) for the ‘chief’, with a title such as Erenavula for the Momo of Sabeto, or Supani for the Ratu ni Natokea (Rakiraki);
  • • a mound for the building for communication with the ancestral spirit (bure kalou or beto/bito) Such a mound has a name. A permanent building on a low mound known as Nukuwasiga in Namotomoto village, Nadi, is still used by the bete of the Kovacaki people for communication with the ancestral spirit, Tutuvanua. The bete lives nearby in a fine modern-style house. He comes from the bete group known as the Naobekwa. A well-known mound (outside my main study areas, but where I have carried out excavations: see Parke n.d.- b, 1993, 1998) is Navatanitawake for the Kubuna people on the island of Bau. This had been the bure kalou of the original ancestral spirit, Ratu mai Bulu, whose place was on the mainland. After a military coup in the early part of the 19th century, the war chief or Vunivalu deposed and expelled the traditional paramount chief or Roko Tui Bau. Navatanitawake was then considerably heightened and developed and became the bure kalou of Cagawalu, the war spirit of the Kubuna people and particularly associated with the Vunivalu;
  • • ‘Totemic’ features, usually three, referred to as vutiyaca in the West or icavuti in the East. One such feature often represents the male genital organs; and another, the female organs. For instance, among the Navatu people near Vaileka, the deiro or sandfish represents the male genitals, and the boro or kind of mussel represents the female genitals. One may represent both. For instance, in the village of Korobebe up the Sabeto valley, three socio-politically associated yavusa have vutiyaca as follows:




kea ilava





hard wood

kind of yam









volo (two long vowels)

Malay apple

wild yam

kind of fish

They are all very tabu, and the mere mention of them can be disastrous unless ceremonies of apology are made immediately. I could not discover any convincing explanation for their origins or for the association between a particular set of such features and a particular yavusa. The only explanation suggested to me was that the original ancestral spirit brought them down from the Nakauvadra Mountains. In the West, such features usually include a type of tree, a root vegetable, and an edible creature of the land or sea, referred to in the western dialects as kai (tree), magiti (vegetable) and kea ilava or food concomitant with the vegetable.

In Rakiraki the classification of icavuti is less straightforward. In the Namotutu yavusa, this chiefly yavusa has four such ‘totems’—a kai (tree), being the vesi; a manumanu (animal), being the kula or parakeet; an ika (fish), being the vonu or turtle; and a vivili (shell fish), being the kusau or small cockle which represents the genitals of both males and females. I was told that the icavuti also included the vara (a shooting coconut) and the kuka (a small red land crab). To confuse the issue even more, the Rakiraki bete told me that each mataqali of the Namotutu yavusa had its own icavuti as follows:








a sugarcane


cooking banana





a tree

sea fish







Tuicakau and Vuninokonoko






It will be noted that the vudi is a common icavuti of all the mataqali in the Namotutu yavusa. This could be the unifying feature of a yavusa comprised of groups from other yavusa.

On enquiry, a greater number of icavuti may be given for any particular yavusa. This may be due to ignorance or confusion on the part of informants, or it may show that the original yavusa or group claiming descent from a single ancestral spirit now includes part or parts of another or other yavusa which brought with them one or more of its or their own icavuti. Further enquiries may reveal the origins of any such external groups. Thus, the reality of the composition of a particular current yavusa may be different from that of the ideal yavusa representing common descent from a common mythical ancestral spirit.

  • • A jumping-off place (na icibaciba in Rakiraki; ne ivilavila ni yalo in Western Viti Levu and the islands) for the yalo or spirits of the newly dead

Each place, often a cliff, would have a name and usually an associated pool in the sea or river into which spirits would jump. Some places have a rock in the pool or sea, which represents a vessel (vatu waqa or stone canoe) in which the spirit would sail away to join the ancestors. The spirits of the dead of the Malake yavusa go to the little broken cliff at Namuremure at the western end of the island of Malake. They jump into the sea. To the west of the cliff is a rock called Muadua. It is shaped like a vessel, and if someone sleeps on the rock they dream of it as a large vessel. The spirit boards the vessel and sails round the island to see the villages for the last time. Then the vessel heads to mangrove swamps at Rukuvakadua where the spirit disembarks, and the vessel returns to Namuremure.

A rock may have some other part to play in the journey of the spirit. The spirits of Namotutu, Rakiraki, jump from a beach at Navolivoli into the sea where there is an isolated coral rock on which the spirit is injured as it jumps. The spirit then limps ashore with the aid of a walking stick until it reaches another rock a short distance inland. Having recovered from its injuries, the spirit plunges its walking stick into the rock which may itself testify to this because it is covered with small round holes. The spirit then departs to join its ancestors. The spirits of Navunatoto, Nawaka which is beside a river, jump from the bank into a river pool at Nawakalevu and return to the land to a big baka or fig tree. A baka tree is generally associated with spirits.

The ivilavila ni yalo of the Yaukuve people on the island of Waya is at Nacilau near Nalauwaki. The spirit of the dead person flies around the island and then goes off to Nacilau with its iyaya or belongings. It throws its iyaya ahead of it down the cliff at Nacilau and then it jumps or vila into a pool beside a large flat rock known as Waqa ni Senio. The rock represents a vessel (waqa). The people of Waya hear two thuds representing the sound of the iyaya and of the spirit falling into the pool. The spirit goes aboard the vessel, and sails off into the West together with its iyaya.

  • • Other vanua tabu

Apart from the original ancestral spirit and vanua tabu (‘sacred’ places) associated with that spirit, other spirits connected generally with a yavusa have vanua tabu associated with them. The headland above the cave at Sawailau off Yasawa Island is regarded as the vanua tabu of the famous spirit Lewaqoroqoro associated with the Natubasa people of Nabukeru village, who appeared as a female person. She was carried off to Malolo Island by Tutusilo, the main spirit on the island, at the end of his penis which had erected when he smelt her urinating on Sawailau.

  • • Personal names

Certain personal names may be especially connected with a particular yavusa, such as ‘Bolobolo’ with the Namotutu yavusa of Rakiraki; and ‘Dawai’ with the Navatulevu yavusa of Nadi.


Nowadays, land is generally communally owned by a sub-division of the yavusa. In some areas, the land-owning unit was traditionally the mataqali, in others it was the itokatoka or equivalent. Andrew Pawley (personal communication) told me that on Waya, the unit was the yavu or housemound. This situation can occur sometimes on Rotuma in the case of hanua ne fuag ri or land associated with a particular housemound. I know of no example in my research areas where the yavusa owned land. The traditional iyalayala or boundaries of such areas were recognised, although the precise boundaries may have been vague. Nowadays land boundaries are fixed by legislation. Each area of land has a name. Land (qele) may be divided into three categories: village sites, iteitei or gardens, and veikau or bushland. On such land there would be a number of vanua tabu or ‘sacred’ sites associated with an ancestral spirit or other spirit.

Village sites

Each yavusa occupies one or more villages (koro, or rārā in the West), which it may share with one or more neighbouring or related polities or parts of polities. Usually it would be situated on land owned by one or more of the occupying mataqali or itokatoka whichever was the land-owning unit. In the district of Rakiraki, the yavusa of AiSokula is the only yavusa based on the village of Nakorokula. To the west lie the immediately adjacent villages of Navuavua and Navutulevu, commonly referred to together as the village of Rakiraki. The original owners were the yavusa of Namotutu who were later joined by four stranger yavusa, the Navuavua, Natiliva, Cakova and Wailevu, who originated from the other side of the Nakauvadra Mountains. The inhabitants of the Nadi village of Sikituru belong to the yavusa of Saunivalu. Following what they regarded as the wrongful distribution of a ceremonially presented pig, they had split from the Saunivalu yavusa who live in the present village of Keiyasi, far up the Sigatoka River, and they now form an independent unit. They assisted the Nadi people against the Sabeto people in battle, including the great battle of Vauroka. Out of gratitude, the Saunivalu were given land by the two yavusa of Kai Loa and Vucunisai, who agreed that the leader of the Saunivalu should be recognised as Momo or paramount of the three yavusa. In 1993, they were finally reconciled with the Saunivalu of Keiyasi in the course of a tearful ceremony, with pigs being presented in the correct manner.

Features of a village

Each village would have a name, and would be divided socio-politically, ideally round a central unoccupied ceremonial area (rara, or darata in the West). Each mataqali or itokatoka would have a recognised area in the village. At one end of the ceremonial area might be the chief’s house referred to by such terms as Vale Levu, Sue Levu or Were Levu. At the other end might be the building where the bete communicates with the ancestral spirit or where socio-political ceremonies take place. Nowadays this may have been replaced by the church or the ceremonial meeting house. The village plan of Bulia, Kadavu, follows exactly this model. The chief’s house is at one end of the rara, and the church at the other. At Lalati, Beqa, the chief’s house is at one end, the community hall at the other, and the church is halfway between, on the edge of the rara.

Gardens and bush

Some garden sites may be close to the village, particularly those used by the women. Other sites may be some distance from the village, but seldom so far that a gardener cannot get home at night though he may have a garden house. The bush is mainly useful for gathering firewood, house posts, wild yams, and for hunting wild pigs. Some areas are cut down for garden sites in the process of shifting agriculture.

The spirit world and associated geographical features

The areas of Rakiraki and of Nadi/Vuda/Nawaka in my study areas have mountainous backgrounds behind the coastal plains. Most of the Yasawa islands have a mountainous centre, except for the isolated sandy cay of Viwa, lying far westward of the other islands. The mainland areas have hilly piedmonts with rocky outcrops and isolated errant boulders. The coastal areas are relatively flat but otherwise characterised by cliffs, streams, rivers and deltas often associated with mangrove swamps. Where the rivers flow from the mountainous interior, waterfalls often occur. With such a varied and dramatic terrain, it is not surprising that many features, natural, humanly made or natural but humanly modified, abound in association with the spirit world of the polities in these areas.

Such features generally fall into the following categories:

Subterranean or sub-surface features

  • • Pools in the sea

Waicacavori is a spirit associated with the AiSokula people of Nakorokula, Rakiraki. His place is at Sogoi, a pool in the sea to the west of Ellington wharf. His waqawaqa is a big vai or ray often seen jumping out of the water in the bay. If someone in the bay is in trouble, a vai will appear and push a log or something else in order to save him.

  • • Pools in the mangrove swamps

Namo (meaning a pool full of fish) is a pool on the edge of mangroves to the west of the village of Namuaimada, Rakiraki. It is the place of a female spirit called Vai'eka, associated with the local Natokea people. Passers-by are very frightened of her, and she would ill-treat those who fail to pay her due respect. Natokea people wishing to fish can, with suitable ceremony and presentations, seek Vai'eka's permission to fish at Namo.

  • • Pools created by subterranean water seepage

Naqali is a small pool of fresh water seeping from the hillside near the village of Korokula. It is the vanua tabu of Varoi, a male spirit associated with the Naisokula yavusa. His waqawaqa is a duna or eel. If someone comes with a container to get water from the pool, an eel's tail may be seen to wave and splash water into the container.

  • • Pools in rivers or streams

A Tongan woman called Bilovesi while visiting the Nakauvadra Mountains became pregnant by Degei who sent her to give birth at Navatu. Degei's waqawaqa is a gata or snake, and he is known as Gata i vanua (snake on land). When he heard of his son's birth, Degei decided that the son should be called Gata i wai (snake in the water). Gata i wai's place is a pool in the stream near the present village of Narewa, beside a rock called Vatudamu. His waqawaqa is also a gata, and his descendants are the people of Naivuvuni who are regarded by their neighbours as having a Tongan look.

  • • Caves

In the Rakiraki area, Salusalumatana is a spirit associated with the Natokea people. His place is at Naqaraituruturu, near the village of Navolau. It is a shallow cave in a big rock, which had a coconut palm on top in my time. The place of Degei is described in legend as a cave at the peak of Uluda, on top of the Nakauvadra Mountains. I have been shown a rock with a modest crevice, said to be the cave, where I made suitable presentations. This reality contrasts with legends which refer to a deep cave, where people are said to have entered and seen marvelous things, including Degei, as a snake or as a half man/half snake. Williams (1858:217) referred to a ‘gloomy cavern—the hollow of an inland rock’. Joske (later Brewster) in 1885 actually visited and described the cave. He is cited by Kaplan (1988:152) as saying that ‘Degei the Kalouvu or Fijian Creator lived in a cavern at Na Kauvadra. He was a gigantic serpent but, at such times as he willed, assumed the human form’. I was told in Rakiraki that within living memory two men from the village went into the cave and were allowed by Degei to take some buli tabua (white cowry) back to Rakiraki where they were placed at the end of the balabala (tree fern) ridge poles of Tu Navitilevu's house. They were shown to me on his house in 1952.

At the northern base of the Tualeita Range about 3 km from the western end at Edromu there is an impressive rock formation, with a cave and a pool. This is the place of Sadidi, female ancestral spirit of people living at the present village of Lomolomo, and the pool is her bathing place. Her manifestation is a duluduluwai or dragon fly. Nearby is an earth mound where presentations are made to her. To the west, at Edronu at the base of the end of the Tualeita, there are two caves. At the western cave, known as Qwara ni Masumasu, supplications for support were made to the local spirit, Balawakula, whose place is inside the cave. The eastern cave, known as Qwara ni iYau, is the place of the spirit Sagavulunavuda, son of Lutunasobasoba. Persons wanting sau or supernatural power for war clubs for use against the enemy, or for a meke to be performed by the supplicants would place the clubs inside the cave for four nights. The bete would communicate with Sagavulunavuda and request the help being sought. Such presentations were made when the Fijian Military Force’s troops departed for the Solomons Campaign in 1943.

The impressive and well-known cave on the island of Sawailau, south of Yasawa island, has a pool where two fishes are said to be the waqawaqa or manifestations of spirits associated with the people of the village of Nabukeru. One wall has a series of petroglyphs associated, according to some I met, with the spirit Lewaqoroqoro. However, most people denied any such association. Steps cut in the wall of the cave lead up to an alcove described as the seat for the bete when communicating with the spirits.

Superficial features, natural and man-made

  • • Rocks, naturally static or errant

These may be found on land, in a river or in the sea. Some rocks may show no obvious feature to merit choice among otherwise similar rocks as being spirit-associated. At the edge of the village of Namuaimada beside the main road, a large but otherwise unremarkable rock called Tadili was said to be the place of Qelo, the club-footed ancestral spirit of the Navolau yavusa. When I called there in 2001, the road had been widened and the surrounds of Tadili had been disturbed. Qelo is a violent spirit if disturbed, and I asked the people living beside the rock if anything unfortunate had happened. They said that Qelo had simply gone away from Tadili.

In the area known generally as Korotabu near the village of Vatutu, Nadi, is the place of Raituvulaki, ancestral spirit of the Vunataqwa yavusa. A feature of this site is an otherwise unremarkable rock identified as the seat of Raituvulaki. On one occasion, he sat here with a local female spirit, Lewatu. He had an erection and entered her, and his penis so extended that it carried her as far as the black sand beach at Wailoaloa. It was suggested to me that this was the origin of the name Utiloaloa (black penis) which refers collectively to Vunataqwa and other associated yavusa.

  • • Other rocks with a natural feature

A rock on the north coast island of Malake has a cone-shaped hollow, described to me as resembling a vagina, and associated with the female spirit Bilovesi. A rock I saw at Edronu with markings described as resembling male genitals was associated with a male spirit, Sagavulunavuda, being either Balawakula referred to above, or Lutunasobasoba, a culture hero who came with Degei aboard the Kaunitoni.

  • • Rocks modified by human activity

Some rocks may be inscribed with petroglyphs, such as those near to the caves at Edronu; or at Natuvi, the rocky hilltop of the ancestral spirit, Raituvulaki. Others may be shaped to resemble a shark, such as the rock known as Nadakuvatu sited beside the yavu of Sayake in the village of Tavakubu, near to Lautoka. Others may be ground to form a either a round hollow (for pounding yaqona) as on several boulders beside the burial mound Kawalevu, to the north side of the Navatu acropolis in Rakiraki; or to more elongated hollows suitable for sharpening adze blades, of which there are many on the rocks near Edronu referred to earlier.

  • • Rocks associated with a pool

These may represent the place of a spirit or where it bathes. To the northern side of the Tualeita up not far up the valley from Lomolomo is the pool shaded by trees beside rocks, identified as the place of Sadidi, the ancestral spirit referred to earlier. Twin spirits, known as the Ciri, associated with legends of Degei, have a place at Nukunitabua to the north side of the Nakauvadra Mountains. It is a mound with a megalithic complex on top. Below it a stream runs between rocks, in which is the bathing pool for the Ciri.

  • • Rocks carried from an earlier site

These may be erected in a new settlement, perhaps on the ceremonial site for communication with the spirit. In the naga-like ceremonial site at Wasavulu, Labasa, on the north side of the central mountain range on Vanua Levu, there now stands a massive monolith, possibly the tallest in Fiji. I was told that the Labasa people carried it over the range from their antecedent site at Nukubola when they came over from the southern side of the range to settle on the north side. With it came their ancestral spirit.

  • • Rocks hallowed in connection with the installation (veibuli) of a new paramount chief

For the installation of the Momo Levu of Sabeto, two round stones are used. Momo Levu sits on one and holds the other. They are kept in the house on the ceremonial mound of Erenavula in Koroiaca village. For the installation of Tu Navitilevu of Rakiraki, the chief leans against a monolith, which at present stands beside the church. It was brought from the yavu tabu of Yavunuku in the original settlement of the Namotutu people, where it stood on top of the mound and was associated with the ancestral spirit of the Namotutu yavusa.

  • • Rocks used in connection with cannibal activities (vatu ni bokola) usually found at vanua tabu

Standing beside Navatanitawake, the main bure kalou on the island of Bau, was a monolith used for the killing of people to be presented as bokola or bodies to the spirit of the ‘temple’. The stone was later moved to the newly built church, and was used as the font when I saw it.

  • • Earthen mounds, which may have a monolith on top

Such mounds may be places where the spirit stays, or where presentations are made to the spirit, or where the bete communicates with the spirit. They are usually man-made. They may be rectangular or round, terraced, and flanked by natural or carved stones; and are generally higher than other mounds in the village, or in their the vicinity. On Bau, the bure kalou of Navatanitawake is rectangular, terraced, flanked by carved and natural stones, and is higher than any other mound on the island. In Rakiraki, the yavu tabu of Yavunuku at Namotutu is almost round, unterraced, with no flanking features. It had a monolith on top until the Namotutu people moved to the present combined village of Rakiraki, and took the monolith with them and re-erected it beside the church. In the old village of Saravi, Nadi, one high rectangular mound had been surmounted by a monolith. This stone had been taken down, deliberately broken up and the pieces had been buried in the mound. Nearby, a simple circular mound was pointed out to me as being associated with the presentation of gifts to the spirit of the place.

Height above ground is an important element in the symbolism of Fijian socio-politics and the spirit world; and in this connection, comparative height is an important feature in the construction of man-made mounds. The higher the housemound, the higher is the social status of the owner, and the closer is the owner connected with the original ancestral spirit. The highest mounds in a village are usually those connected with the paramount chief and the ancestral spirit. Vunisei, the house mound of the Momo Levu of Vuda, is high and impressive in comparison with the other mounds in Viseisei village. Excellent examples of high and impressive spirit mounds are Navatanitawake (Bau), Natavasara (Taveuni) and at Tokatoka (Rewa).

Features rising significantly above the land or sea

  • • Cliffs

Cliffs are often associated with jumping-off places for the spirits of the newly-dead. The cliffs at Nacilau, Waya Island, and Namuremure on the island of Malake are places from which spirits jump into the sea where there were rocks. A cliff at Ucunivanua, Nakelo, is the end of a spirit path connecting Nakelo with the spirit centre at Uluda on the Nakauvadra Mountains. This path is symbolised by the roots of a tree which grew on the Mountains. It is referred to as the wakanivugayali or the roots of a large tree of the myrtle variety. I have been shown what are purported to be the roots sticking out of the cliff at Ucunivanua.

  • • Waterfalls

Vaikitu is a bara ni sava or waterfall above the present village of Nayaulevu on the south side of the Nakauvadra. This is the settlement of Kitu, the original ancestral spirit of the Nacolo people now living at Vatukacevaceva on the north side of Nakauvadra. Kitu's waqawaqa is an ula, a kind of frog, which stays in a qila or crevice in the waterfall. If it appears in his house, the bete will speak to it vakaula or in frog's talk which no one else understands. If there is trouble, the bete prepares yaqona and presents it to Kitu at Vaikitu.

  • • Islets, with or without man-made features, caves or pools

The islet of Cubu, offshore from Navatu, has a small cave associated with the female spirit Bilovesi. The islet of Narokorokoyawa in the Northern Mamanucas has a number of features associated with the ancestral spirit Tuirevurevu. These include caves, a pool, man-made structures and monoliths; and a more recent stone complex where visitors now make presentations to Tuirevurevu. Tuirevurevu's place called Rukurukulevu is regarded as very tabu, greatly feared and respected.

  • • Mountain tops, on which may be natural features such as rocks and caves

The Nakauvadra Mountains dramatically overshadow the Rakiraki area. Features include, first, the main spirit centre at Uluda symbolised by the cave of Degei in a rock; secondly, an impressive rock known as Rogorogo i Vuda, representing a lali or drum for passing messages to the other main spirit centre at Vuda; and thirdly, the site of the vugayali or myrtle tree, the roots of which symbolise the spirit paths (see under ‘trees’). Further research may determine whether these elements represent more than one original cult even if present tradition weaves them into one. On the top of the Tualeita mountain range above the spirit centre of Edronu with associated caves and petroglyph-covered boulders, a tree-covered rocky eminence represents the ivilavila ni yalo were spirits who have followed the Tualeita spirit path jump into the sea and head west. On the island of Waya, the southern peak of Ulu i Tunaiau (about 550 m above sea-level) is the place of the autochthonous ancestral spirit Tunaiau. He is regarded as the ivakatawa or guardian, and the itaukei du or true owner of Waya. He went to Samoa and brought back his wife and other Samoans who settled on Waya. The Wayans still recognise what they regard as their Samoan connections.

  • • Natural hillocks, on which may be natural features such as rocks or caves, or man-made features such as mounds

These include the two hill-top sites of Navatu (Rakiraki), and Korovatu (Vuda). The former was first settled by Waqabalabala, the original ancestral spirit of the Navatu people, and then by his two sons, all three of whom have a particular rocky peak. Among other spirits associated with Navatu is Coci, who has a hare­lip, and occupies a cave on the south side. Korovatu was first occupied by Lutunasobasoba who came to Fiji from the west in the vessel Kaunitoni. When he left to travel overland towards the Nakauvadra, his eldest son, Sagavulunavuda remained on Korovatu. A number of yavu or housemounds can be seen on top of the hill, and one is regarded as that of Sagavulunavuda, and others those of his bete and household. Even in my time, people still approached him on top of Korovatu, for assistance. There is supposed to be a cave beneath the hill where lives a spirit who is part of the defence apparatus against those who try to attack Korovatu.

Trees (being partly subterranean, partly surficial and partly rising towards the sky)

  • • Certain species of tree such as baka or fig trees, being either the places of the spirits, or closely associated with them

These trees are regarded as tabu; and if they are cut down or damaged on purpose some disaster will befall the perpetrator unless appropriate ceremonies of apology are performed. After the 1952 hurricane destroyed much of Rakiraki, a baka tree near the mound of Leka, the ancestral spirit of the Cakova yavusa, was badly damaged and several of the branches had to be cut off. Leka took great offence and appeared in the form of a short man (leka means short). Ceremonies of apology were performed on his mound, and all was well.

  • • Particular mythical trees

Certain trees may have special connections with the passage of the spirits of the dead. The mythical vugayali on the Nakauvadra above Rakiraki has roots (the wakanivugayali) branching out to a number of places in Fiji including Edronu (Vuda), Ucunivanua (Nakelo), Waya and Ucunivanua (Bua). These roots symbolise spirit paths between the spirit centre on Nakauvadra and these other places. A particular balawa or pandanus lies on a spirit path running along the North coastal region of Vanua Levu. It is referred to as na balawa viriki (the pandanus tree which has something thrown at it). When the spirit of a recently dead person reaches this tree, it must throw at it a stone known as a vatu ni balawa or stone for the pandanus. If it hits, the spirit proceeds on its journey. If it misses, it remains there. I have been shown the place near Udu Point.

  • • Living trees

Some living trees, either particular individual trees or trees of a particular species, are regarded as having spirit-derived properties. There is a particular tree in the old village of Nawaka Makawa, near the present village of Nawaka. If a mother is short of milk, she may go to this tree, and slash it. As the resin runs out, so will her milk supply miraculously increase.

  • • Trees as symbols of identification

A most important aspect of the spiritual and socio-political aspect of trees is the almost general reference to a particular type of kau or kai tree among the icavuti or vutiyaca (‘totems’) of a particular yavusa. As such, the tree is one of the symbols of identification of the yavusa. Its exact significance is nowadays not at all clear, though it is often associated with the male genital organs. Any insult to it such as a derogatory reference can still lead to most unfortunate results for the offender unless ceremonial apologies are hastily performed.

Orientation, horizontal and vertical

Myths abound about spirits from the east travelling westward along the appropriate spirit path to the central spirit place on the Nakauvadra. Others tell how spirits go either directly from the icibaciba or jumping-place or via the Nakauvadra and the western spirit-path, to Edronu and thence head west. I am aware of little evidence for the significance of horizontal orientation in such tasks as the siting and design of a village.

Similarly, I have scant evidence for vertical orientation in the spirit world. Some myths refer to a spiritual settlement known variously as Burotu Kula, Vanua Kula or Purotu, (see Geraghty 1983a:343–384), described to me as floating in the sky, as it was recently seen in Rakiraki. Other myths refer to the underworld of Bulu, and the original ancestor of the Bauan paramount polity is known as Ratu mai Bulu. His place is on the Viti Levu mainland whereas his present mound of communication is Navatanitawake on the island of Bau (see Parke n.d.- b, 1993, 1998). Similarly there are myths regarding Lagi, the sky, and of spirits or spirit places floating in Lagi. There is an expression ‘ki Bulu, ki Lagi’ or ‘to Bulu, to Lagi’ used in ceremonial speeches, which connect the underworld with the spirit world. These questions of horizontal or vertical orientation in the spirit or the socio-political worlds of Fijian society need further consideration. There is little evidence of which I am aware in the current literature. It is therefore important to carry out further investigations in the field before relevant myths are forgotten.

The centre, the outside and the threshold

A yavusa is described earlier as ideally being socio-politically exclusive and hierarchical, but these ideal characteristics can be looked at from another angle too.

The paramount chief is the socio-political centre of the yavusa just as the original ancestral spirit is the centre of the yavusa’s spirit world. The two worlds merge in the person of the paramount chief when his installation has been validated by installation ceremonies. These include invoking the ancestral spirit, and may involve the use of installation stones associated with the spirit. Nowadays installation is formalised by the drinking of yaqona. I was told that in the west tovu or sugar cane was once used in the place of yaqona. Yaqona or tovu may be the medium for transferring mana or spiritual power from the spirit to the paramount.

Because of this association between spirit and installed chief, the installed chief becomes tabu or dredre (difficult to approach). Similarly, the paramount's housemound is tabu, just as the sites associated with the ancestral spirit such as the spirit house are tabu. It is regarded as dredre sara, or very difficult, to approach the house mound or enter the house of the paramount chief, just as it is dredre sara to approach a tabu site associated with the ancestral or indeed any other spirit. To put this in another way, the chief’s house may be regarded as the socio-political, physical centre of the village; and it is dredre sara to go from the outside (tuba) of the house across the threshold (tabailago) to the inside (loma). Similarly at the village level, it is difficult for a stranger casually to cross the boundary (iyalayald) of a village from outside (tuba) of the village into the inside (loma ni koro). Once inside the village, it is difficult for the stranger casually to enter the inside of a house, with the household of which he or she has no socio-political or spiritual connection. It is certainly very difficult indeed to approach the chief, or to enter his house or the ceremonial house. To cross such thresholds or boundaries, ceremonies of request and appeasement of the spirits must be conducted in the appropriate manner, to the appropriate person or persons, and in the appropriate order, depending on the nature of the centre to be reached, be it socio-political or spiritual. In theory and practice this comes down to the same thing.

It is similarly extremely disrespectful for any one casually to cross or wander about on the ceremonial central area (rara, or darata in the west) in the middle of the village. Nowadays it may be used for a cricket pitch, and this may be an indication of the attitude of the traditional chief of the village to cricket.

The yavusa in reality: fusion, fission and movement

The ideal yavusa appears to be exclusive, hierarchical, static, independent (tu vakaikoya) and socio-separate from other yavusa. In reality, socio-political internal fusion and fission as well as geographical movement and interaction with other yavusa and with external forces such as Tongan and European visitors, settlers and missionaries, were features in the development of most yavusa, but this raises the issue of how a social unit becomes a yavusa in the first place. Ideally again, a yavusa is the descent group of a particular ancestral spirit. In reality if a group from an established yavusa arrived in a new place they would probably not regard themselves as of separate yavusa status unless invited by the host yavusa to do so, or until they become sufficiently numerous to justify doing so without appearing to the itaukei to be viavialevu. This means having ideas above their station without physical or spiritual justification to maintain their new status—a deadly socio-political sin.

Fusion and fission

One of the most interesting features in a yavusa development is the degree of socio-political and socio-economic connections with other yavusa through fusion and fission. Fusion and fission within a yavusa may have come about through:

  • • disagreements and insults between one mataqali or itokatoka and another.

These could result in the disaffected group leaving the rest of the yavusa unless ceremonies of reconciliation were performed successfully.

  • • an increase in numbers in the yavusa.

This could lead to insufficient space in the village site, or inadequate planting land close at hand, or an inadequate water supply. This could result in an amicable polity-wide decision to split and form another village with adequate facilities, either on land owned by the splitting group or on land requested by the group from a neighbouring yavusa.

  • • a request from outsiders to join the yavusa, perhaps after an unreconciled quarrel.

Members of a mataqali or itokatoka of another yavusa could approach the yavusa in question with a request to join them and settle on their land. Similarly, a yavusa could invite members of another yavusa to join them, perhaps for reasons of prestige. In either case the approach would be made formally through the ceremony of ilakovi. The vulagi would then be accepted as itaukei for all intents and purposes, although if they tried to override the real itaukei in important traditional discussions they could be reminded of their origins.

Change in social structure of a yavusa

The social structure of the hierarchy recognised by the status of the different mataqali comprising a yavusa or the ranking of individuals within a mataqali might change because of:

  • • internal quarrels, including fighting, within a yavusa

This could lead to a mataqali lower in the hierarchy forcefully replacing a more senior or the most senior mataqali.

  • • refusal to accept responsibility

The most senior member of the chiefly mataqali may not wish to accept the responsibilities of installation to paramountcy.

  • • unsuitability for office

Members of the various mataqali, especially the senior mataqali, may not consider the most senior member to be worthy of installation and a person whom they regard as more suitable may be installed instead. Alternatively, no-one may be installed as paramount. This may be a face-saving temporary alternative for the most senior member or for the yavusa as a whole. However, sooner or later, the socio-political situation of the yavusa may result in internal instability, or in political machinations or misunderstandings on the part of other yavusa in the course of their relations with the yavusa concerned.

Geographical movement

The first settlement

Parallel with the realities of socio-political fusion and fission within a yavusa, my investigations show or confirm that, in reality, the yavusa’s village site was not static but usually changed on several occasions from the traditional first settlement (yavutū) of the yavusa. According to local myth, the site for such a settlement may have been the same place as where the original ancestral spirit finally settled, having wandered about looking for a suitable and vacant site. Many myths have it that the original ancestral spirit may have come down either from the spirit centre on Nakauvadra, or from the mythological first landing of the first craft in the Vuda area. Alternatively, the son of the original ancestor may have been born a spirit at the final settlement of the original ancestor and turned into a human who remained in the spirit settlement of his father and grandfather. Finally, the first human settlement may have been near that of the final settlement of the original ancestor.

The descendants increase in number

Whatever the nature of the first human settlement, it was here according to the model that the human descendants of the ancestral spirit increased in numbers. In course of time these descendants divided up and comprised recognised groups which are now referred to as mataqali, the hereditary status of which was based on the mythological order of birth of the sons of the original ancestor. Eventually these mataqali divided into what are now referred to as itokatoka or some equivalent term depending on the dialect and geographical area, the seniority of which within the mataqali was based mythologically on the order of birth of the grandsons of the original ancestor. Whatever the reality of the situation in prehistoric times, this was the socio-political situation allegedly derived from mythological origins as accepted by the Native Lands Commission (NLC) and the Colonial Government for practical purposes and for the ease of Fijian land administration. The question may well be raised as to whether the first split was not rather into itokatoka which later united with related itokatoka and created a mataqali. It is probably too late to try to determine this matter and beyond the main purposes of this project. In course of time, whatever its precise pattern of development, the yavusa may have moved from the first settlement in toto or may have split up and settled in various settlements, or it may have been joined by members of other yavusa.

Later settlements

In any case, parts or the whole of the yavusa moved to new village sites, settling either by themselves or together as a yavusa, or joining other yavusa and sharing a village site. The new site(s) may have been chosen for better access to gardening land, water supplies, or fishing or hunting areas, or as being more easily defended in the course of inter- or intra-polity quarrels or warfare. Sometimes a new village was sited to be near a place of spiritual significance. Sometimes it was situated deliberately far away from the earlier site, especially if the new settlement had arisen from a split in the yavusa as the result of internal quarrels and insults. Sometimes the change in village site resulted from an invitation from another yavusa to settle with it for the purposes of mutual protection. In other cases the new settlement may have resulted from a powerful yavusa wanting to settle trusted people in a newly conquered area. A new village site may have taken with it the name of a previous settlement or it may have been given a name based on a local feature, geographical or spiritual.

Symbolic connections with older sites

If a polity or part of a polity moved to a new site, the new site would retain some features from the previous site, especially those connected with the spirits, and it would do so in spite of any realities of socio-political fusion and fission. The polity might take with it a monolith representing the original ancestor, or the installation stone for the paramount and associated with the spirits. It would construct special mounds for the house of the paramount or for the building where the bete communicates with the ancestral spirit, perhaps taking the names of the similar mounds at the previous village site. The names of some of the other housemounds might also be taken from those in the old village site.

Networks and inclusiveness

A feature of the ideal yavusa is its exclusiveness, but my investigations indicate that, in reality, some of the most significant factors in the development of Fijian society were the socio-political and socio-spiritual connections existing or created between one yavusa and one or more other yavusa. Such connections may have arisen through myth or by custom; they may have developed voluntarily between yavusa; they may have been forced by factors outside a yavusa’s control, but brought about by other yavusa; or a grouping of yavusa may have developed because of external factors outside Fijian society. Sometimes the formally acknowledged membership of a yavusa, based on direct descent from the original ancestral spirit usually along the male line, may have been extended by the inclusion through custom and due ceremony of persons who would not otherwise have been recognised as members of the ideally constituted yavusa.

To this extent, the realities of fusion and fission indicate that a yavusa may be characterised as being potentially inclusive. An analysis of the development of a particular yavusa should examine the extent of exclusiveness and inclusiveness in its composition both synchronically and diachronically. I will now consider some recognised mechanisms for establishing networks between yavusa.

(a) Myth or custom

Connections may have been established by myth or custom through:

  • • veitauvū

Veitauvū is a relationship between the ancestral spirit of a particular yavusa and one or more ancestral spirits of one or more other yavusa. Such relationship may have been based on sibling spirits, or on an ancestral spirit who married the sibling of the ancestral spirit of another yavusa.

  • • Vasu

Vasu is a relationship arising from marriage between members, especially high-ranking members, of different yavusa. The children of the brother of the female spouse had a special relationship with the members of the yavusa of which the uncle was a member. Requests by the former from the latter were difficult to refuse.

(b) Voluntary associations

Associations may have been voluntary and based on such factors as:

  • • mutual convenience

Hill-folk and sea-folk might agree to exchange root crops and yaqona for salt and fish.

  • • mutual assistance in times of trouble or war.
  • • an ineffective yavusa without adequate martial or natural facilities for survival in times of war or peace. Such a yavusa may voluntarily approach a powerful and rich yavusa seeking to come under its protection and authority. This dependence is called tiko vakararavi in Bauan.
  • • association between a politically powerful yavusa and a warrior yavusa,

A politically powerful yavusa may have established a relationship which enabled it to call upon a warrior yavusa for assistance in times of war. The relationship would continue until the powerful polity failed to provide the expected goods in exchange for the military services provided by the warriors, or until another powerful yavusa seduced the warrior yavusa by agreeing to provide more attractive goods in exchange for services. The warrior yavusa in such a relationship is known as bati (literally ‘tooth’ or ‘border’) and may break the relationship (known as veibati) under the circumstances described.

(c) Involuntary associations

Associations may have been involuntary on the part of one or more of the polities involved, due to such situations as when a powerful polity may compel, by force or threat, a weaker polity to come under its sway. This relationship is known as veiqali. The weaker yavusa is referred to qali (literally ‘twisted’ as a coil of rope). Rather than being available on request to provide assistance, as in the case of the bati, the qali is essentially subject to the demands of the more powerful yavusa. The qali may attempt to revolt from the relationship and may be severely punished, even by death. Alternatively it may successfully escape the relationship and become recognised as the qali of another more powerful yavusa. This may well lead to war between the two powerful yavusa.

(d) Influence of outside circumstances

As discussed in detail earlier, polities may combine for mutual convenience and protection due to factors outside Fijian society such as the Tongans’ monarchical ideology and their ambitions to expand and impose their authority over Fijian territory; the arrival of European settlers and traders; and the arrival of missionaries who wanted Christianity to supersede the traditional Fijian socio-spiritual world. In trying to impose Christianity in the place of the Fijian spirit world, and to stamp out practices such as those referred to earlier, the missionaries were striking against important elements at the very basis of traditional politico-spiritual Fijian society.

Spirit and socio-political networking

Networks were facilitated or symbolised by:

  • • spirit paths—sala ni yalo in the east, or calevu ni (y)anitu in the west.

Some spirit paths may be followed by the spirits of the newly dead travelling from an icibaciba or ivilavila (jumping off place of their particular yavusa) to the spirit centre of Uluda on the top of the Nakauvadra Mountains. Other spirits may go by one of the main spirit paths of the wakanivuniyali to the central spirit place of the vugayali on the Nakauvadra. In either case, Uluda and the vugayali and associated spirit paths to Nakauvadra represent communication systems of spiritual unity symbolic of the inclusiveness of Fijian society. Another such system is based on the main spirit centre at Edronu and the spirit path represented by the wakanivugali stretching from Nakauvadra along the mountain ridges of the Tualeita and ending at Edronu.

These paths to Nakauvadra lead not only to the vugayali but also to the mighty peak of Uluda where dwelt the major culture hero Degei who himself may be regarded as a spiritual symbol of unification of Fijian society. The myths about the vugayali and its roots, and about Degei at Uluda are nowadays usually interconnected by a single myth. As suggested earlier, it is possible that these two myths were diachronically or even synchronically separate at some earlier period. From Edronu, the spirits of the dead head off to the west. Other spirits may go off to the west direct from their i cibaciba which often itself faces west. The tendency for spirits of the recently dead is to go west towards the setting of the sun and the mythical area from which the original ancestors are said to have come in the Kaunitoni. This mythical connection with the west could be regarded as a spiritual symbol for the recognition by traditional Fijian society that it had some connection with the outside world—in other words, for the inclusiveness of Fijian society.

  • • paths of socio-political communication—sala in the east or calevu ni matamataraki in the west.

Such paths may be associated with the bringing of tribute from qali or yavusa subject to a paramount polity; or the sending of messages from the paramount to subject qali, demanding goods or services, or to allied bati, requesting assistance in war; or for sending messages of request (kerekere) or invitations (veisureti) from the paramount of one yavusa to that of another.

  • • the development of a system of representatives of a yavusa being sent to live with an associated yavusa.

This may be based on an exchange of ‘diplomatic postings’ between associated polities, especially when geographically far apart. Such a posting is referred to as a mata ki. It may also be based on a system of posting representatives of a powerful yavusa to another yavusa under its sway to make sure that such a polity, especially if recently defeated or if likely to revolt, toes the party line.

  • • a system of ceremonial officers (mata ni vanua) symbolising the threshold between one yavusa and another.

A ceremonial approach to a chief would be made by first approaching the chief’s mata ni vanua who is in a traditional position to open the door and escort the visitor across the threshold into the chief’s house. The arrival of such a visitor would be announced from the outside by calling out the tama. This varies from place. In Rakiraki, the announcement of a male visitor would be ‘Dua! Dua! Dua!’and of a female ‘Mai na vaka dua!’ In each case the reply from inside the house would be ‘Oi dua!’ followed by an expression such as ‘Mai’ (here), ‘Vano mai’ (come) or ‘Ruku mai’ (enter).

Emerging geographical patterns of complexity

Having identified various factors, both internal and external, which may have affected the development of yavusa both diachronically and synchronically, I now proceed to an introductory study of how yavusa variously developed at the period before Cession in 1874. This attempts to indicate whether geographical patterns of complexity in traditional Fijian society generally can be recognised to have emerged by that time.

Previous Next