Degei’s Descendants: Spirits, Place and People in Pre-Cession Fiji
7. The Diversity of Fijian Polities
An overall discussion of the general diversity of traditional pan-Fijian polities immediately prior to Cession in 1874 will set a wider perspective for the results of my own explorations into the origins, development, structure and leadership of the yavusa in my three field areas, and the identification of different patterning of interrelationships between yavusa. The diversity of pan-Fijian polities can be most usefully considered as a continuum from the simplest to the most complex levels of structured types of polity, with a tentative geographical distribution from simple in the west to extremely complex in the east.
Obviously I am by no means the first to draw attention to the diversity of these polities and to point out that there are considerable differences in the principles of structure, ranking and leadership between the yavusa and the vanua which I have investigated and those accepted by the government as the model. G.K. Roth (1953:58–61), not unexpectedly as Deputy Secretary of Fijian Affairs under Ratu Sir Lala Sukuna, tended to accord with the public views of his mentor; although he did point out that ‘The number of communities found in a federation was not the same in all parts of Fiji’. Rusiate Nayacakalou who carried out his research in 1954, observed (1978: xi) that ‘there are some variations in the traditional structure from one village to another’. Ratu Sukuna died in 1958, and it may be coincidental that thereafter researchers paid more emphasis to the differences I referred to. Isireli Lasaqa who was Secretary to Cabinet and Fijian Affairs was primarily concerned with political and other changes in the years before and after independence in 1970. He also discussed critically (Lasaqa 1984: 19 et seq.) the views of O.H.K. Spate (1959) about Fijian society, as implying ‘a thoroughly conservative social order’. He investigated the situation in the case of the vanua of Dawasama, Tailevu and determined that ‘the authoritarian and rigid traditional structure was not always adhered to, particularly by the younger generation to whom this traditional structure was fast becoming obscure, if not obsolete’. He did not specifically explore pre-Cession society, but he told me (personal communication) that he equally considered that the reality did not accord with the model. In the case of a matanitū, I am generally in accord with Sayes (1982) who entitled her thesis ‘Cakaudrove: Ideology and Reality in a Fijian Confederation’ about a major complex polity with which I am familiar as former District Commissioner for the area. Nicholas Thomas, no respecter of graven images, claimed (1986:6, 65) that ‘From the late 1840s, various major social and cultural changes became apparent’ and that ‘Investigations of the nature of factional conflict involved posing the question of the extent to which Fijian leadership and succession was ascribed or achieved’. David Routledge (1985:27–30) presented what he described as a schematic representation of a theoretical abstraction of the social structure of Fijian society and said that ‘Quarrels…particularly over ceremonial matters, resulted in the break-up of yavusa and the fortunes of war caused them to be conquered and dismembered’. Asesela Ravuvu, former Director of the Institute of Pacific Studies at the University of the South Pacific, was to my knowledge the first (Ravuvu 1983:14) to differentiate the two meanings on vanua on the lines on which I have based my research: vanua 1 covering traditional society, with the three elements of spirits, places and people; and contrasting with vanua 2 having the more familiar meaning of a socio-political federation of yavusa. Christina Toren was, however, the first to discuss (1995:163–183) the concepts of spirits, places and people in the context of the village of Sawaieke, Gau where she stayed for twenty months from 1981–1983. I believe that I am the first to pursue these two meanings of vanua, and the three elements of vanua 1 in exploring the realities of traditional Fijian society.
With simple polities tending to be found towards the west and the most highly complex polities generally found towards the east, Marshall Sahlins (1963, 1968) discussed political types in Melanesia and Polynesia with a cultural intergrading in and around Fiji. Melanesian society is described as typically segmental, consisting of many autonomous kinship-residential groups, each being economically self-governing and the equal of the others in political status. Leadership is based on power, and the so-called big-men do not succeed to office but attain status by a series of acts which attract the loyalty of lesser men. Such local independent groups appear in Polynesia as subdivisions of an inclusive political body which is described as ‘pyramidal’, and ‘an enclaving chiefdom-at-large’ (Sahlins 1968:31, 65). As Sahlins put it: ‘Polynesian chiefs did not make their positions in society—they were installed in societal positions’ (1968:90). Sahlins did not see a dichotomy between Melanesian and Polynesian societies. Rather he saw the overall situation as a continuum. Linguists in the past tended to interpret the linguistic scene in Fiji as a dichotomy between east and west but have now determined that there is an element of continuum within and between the two divisions. I have gone one step further, and I believe that I am the first to interpret the overall socio-political situation as a continuum as well. This follows synchronic and diachronic as well as geographical studies of the yavusa and vanua and (where relevant) matanitū.
Synchronically, Fijian polities can be classified socio-organisationally, not absolutely but relatively in relation to a continuum based on levels of hierarchical complexity of the individual polities. At one end of the continuum would be the simple, independent descent group. At the other would be the highly complex socio-political confederation, involving lesser federations and descent groups bonded in several hierarchical levels of authority, culminating in a top level of independent paramountcy. In between would be many varieties of lesser socio-political federations and of associations of descent groups.
Diachronically, Fijian polities can be classified in accordance with the same criteria. The simple, independent descent group would have been the oldest form of polity, and the stable and highly complex socio-political confederation of federations and associations of descent groups would have been the most recent form.
Geographically, Fijian polities in the west tended to be single yavusa or simple groupings of descent groups and minor socio-political federations with limited spheres of influence. In the east, Fijian polities tended to be most highly complex confederations with wide spheres of influence. This was especially the case, first, where the chiefs were able and ambitious but perhaps more importantly had the most powerful military and naval facilities at their disposal; and, secondly, where the ideological and military influences of eventually-monarchical Tonga were strongest.
This chapter discusses the general patterning of pan-Fijian polities of differing degrees of socio-political complexity of structure, bearing in mind the synchronic, diachronic and geographical implications of such patterning. I have based the degree of complexity of an individual polity not so much on some absolute scale of criteria but rather on a tendency towards a continuum of complexity of hierarchical levels within a polity. At one extreme of the continuum would be the yavusa as a simple independent descent group. At the other extreme would be the matanitū or highly complex multi-level confederation, comprised of a number of dependent vanua or groups of yavusa, each at various levels of hierarchy and complexity with their own chiefly authorities but culminating at the highest level in the person of a paramount chief to whom all the lesser chiefs acknowledge authority. In between would be vanua or less complex federations of individual yavusa or groups of yavusa each acknowledging their own chiefs but accepting the authority of a paramount chief.
A continuum of socio-politically structured polities
The basis of the idea of a continuum is the premise that the complexity of a polity depends on the number of levels of hierarchy in which the leaders of each sub-polity at one level vakarorogo or heed the authority of the leader of a higher level of sub-polity or to the paramount leader of the polity. The paramount who does vakarorogo to no one is independent of any higher authority (tu vakai koya). The continuum is illustrated by the following examples:
(a) The simplest structure: the independent yavusa
At one end of the socio-organisational continuum of late Fijian social polities is the independent, single social group or yavusa. An independent yavusa, with its internal structure of mataqali and itokatoka had its own overall chief (na kena turaga) who tu vakaikoya and vakarorogo to no other person. The sub-groups may all claim descent from a common original ancestor or ancestral spirit (vu or kalou vu). On the other hand, some sub-groups may not claim descent from this ancestor but may have become included in the yavusa for reasons already discussed.
Na kena turaga may be either a descendant from the original ancestor or a vulagi or kai tani, a stranger who has been invited to become leader. In some cases, a female may be appointed (as in cases I have been familiar with in Nadroga, Vuda, Ba and Rewa). The person to be leader may or may not be formally installed at a ceremony of veibuli or installation at which he will be given a yaca buli or title. If a stranger who does not claim descent from the original ancestral spirit of the yavusa is invited to be leader, it is likely that he or she will be formally installed and given a title (yaca buli), and at ceremonies be vagunuvi (presented with the first ceremonial bowl of yaqona to drink). The essential part of the veibuli is the yaqona ni veibuli or yaqona ceremony for installation, at which yaqona is prepared in a tanoa or yaqona mixing-bowl. A bilo or half coconut shell is filled with yaqona and is presented first to the person to be installed, by the tu yaqona on behalf of the yavusa to the accompaniment of cobo or ceremonial clapping. This procedure is regarded as the essence of the transfer of mana or sau (spiritual power) from the kalou vu to the person being installed as the spiritual leader of the yavusa. Katz (1993: 47, 57) described yaqona presented in ceremonial circumstances as ‘a spiritual messenger’ and ‘a channel for mana’.
If the new leader of a yavusa is a descendent of the original ancestor, and if the yavusa is composed of sub-groups, all of which claim descent from the same original ancestor, the eldest member of the senior division of the senior sub-group of the yavusa would ideally be recognised as the leader. However, even if the appointment of the new leader of a yavusa is made in accordance with this ideal basis of veitarataravi vakaveitacini or descent from elder brother to younger brother and then from eldest son of eldest brother to younger brother, the appointment could well be disputed, especially if the candidate is weak or unsuitable or if the yavusa includes members not descended from the original ancestor. It is plainly politic, in practice, to make quite clear who the candidate is to be.
The overt presentation of the first bilo to the person to be installed clearly indicated the choice of the new chief to the assembled participants. Similarly, when the yaqona is about to be presented to the person to be installed, the wā or coir rope attached to the tanoa is stretched out in the direction of that person. This aspect of the yaqona ni veibuli presumably derives from similar Tongan ceremonial with which such Fijian visitors to Tonga would have become familiar. The end of the wā facing the person is often adorned with one or two buli tabua or white chiefly cowry shells which have been described to me as the eyes of the snake—perhaps associating the mana of the ceremony with Degei, and thereby combining the validating traditions of Tonga with those associated with the myths of the Nakauvadra Mountains. The ceremony in the context of leadership of a political federation is discussed later.
The new head of a yavusa may be recognised but not formally installed, at least not initially. In this case he may hold a title relating to the chiefly yavu of the yavusa. In the case of the independent yavusa of Naua in the tikina of Nadi, the recognised but not yet installed leader held the title of iTaukei Sawaieke (owner of Sawaieke, the chiefly yavu). After a considerable period of time he was formally installed with the title of Momo i Naua (Chief of Naua) in the presence of a number of dignitaries from other connected polities.
The following two examples illustrating the characteristics of independent yavusa are both in the Yasawa group. Ketekete is a simple descent group associated with the village of Vuake on the island of Matacawalevu, being part of the Old tikina of Nacula in the New tikina of Yasawa. Kai Kese is a simple political group or construct associated with the village of Kese on the island of Naviti, being part of the Old tikina of Naviti in the New tikina of Naviti.
(i) Ketekete: a simple descent group
The Ketekete at Vuake are nowadays recognised as a yavusa, divided into two related mataqali; and each mataqali is divided into several related tokatoka, as follows:
A member of the Nasivitu is installed as the Momo or iTeiteimata (chief or leader) of the Ketekete yavusa, with the title of Rātū. He is independent of any higher authority. The original ancestral spirit (anitu, in the local communalect) is Ravokavoka. He came from the Nakauvadra Mountains to the island of Matacawalevu where he settled at Ketekete at the place regarded and respected as the original spiritual settlement (yavutū) of the Ketekete. His patrilineal descendants comprise collectively the Ketekete yavusa. He had two children, Baba and Loaloa, whose descendants stayed together in the original settlement. The descendants of Baba, the elder son, became known collectively as the Nasivitu mataqali, and the descendants of his sons became known as the Nasivitu (the chiefly itokatoka), Taganikula and Bau respectively. The descendants of Loa, the younger son, comprise collectively the Ketekete mataqali, and the descendants of his sons became the itokatoka of Ketekete, Sava, Davekadra and Tabale. When the water supply at Ketekete became inadequate, the descendants of Ravokavoka moved and settled together at Vuake where they have remained ever since.
I recorded from present members of the Ketekete this account of the yavusa, and their current understandings accord with the investigations of the Native Lands Commission (NLC). Membership, including leadership, is based on patrilineal descent from a common ancestral spirit. The iteiteimata of the yavusa is appointed from the tokatoka comprised of descendents of the eldest son of the eldest son of the original ancestral spirit. Ketekete therefore exemplifies the model of a simple independent descent group having a hierarchical structure at three levels, being at one extreme of the posited continuum of structural complexity of socio-political groups and constructs spanning the overall structure of later Fijian traditional society.
(ii) Kai Kese: a simple political group or construct
The Kai Kese at Kese are nowadays recognised as a single yavusa, divided into two mataqali; and each mataqali is divided into several tokatoka, as follows:
Lasawa Naqeleravu Nadonia
Vite (or Vile?)
A member of the Nasukamoce is installed as the leader of the Kai Kese yavusa, with the title of Rokotakala. He is independent of any higher authority.
The Kai Kese are constructed of three socio-political elements. The first, the majority of the Kai Luvuka, are the descendants of Botabota, the original ancestral spirit, who settled at Kese. The second, the Kai Nasukamoce, came from Marou, Naviti, but originally from Votua, Ba. The third, the Leweivawa, originally from Vitogo, Vuda, are the descendants of a number of persons who split from of the Leweivawa yavusa of Muaira on the island of Naviti and joined the Kai Luvuka.
As for the Kai Luvuka, the origin myth tells how Botabota came from the Nakauvadra Mountains to the island of Naviti, and settled at Vanua in the middle (luvuka, in the local communalect) of the island. Vanua is regarded and respected as the yavutū or original spiritual settlement of the Luvuka mataqali. Botabota’s patrilineal descendants comprise collectively the Luvuka mataqali. He had six children, the eldest of whom was Matatini, whose yavu or housemound was called Luvuka, and whose descendants comprise respectively six of the seven tokatoka of the mataqali of Luvuka—one of them is now extinct. Each was named after the respective yavu of the six children.
The Nasukamoce, the senior itokatoka of the Luvuka mataqali are descended not from Botabota, but from persons who had split from the Kai Koro of Marou, Naviti. The latter had come to the island of Naviti following a quarrel at Votua, Ba. They were settled at Marou by the leading Naviti group, the Suelevu, who decided that, because of their strength and abilities, one of the Kai Koro chiefs should be installed as paramount for the adjacent villages of Marou and Malevu. The person chosen was not the first-born but a younger brother. This upset the Kai Koro, who split up, and many left Marou among whom were the group who went to Kese and were invited to settle with the Luvuka original landowners. The Luvuka were so impressed with the qualities of these vulagi that they decided to make them a part of the Luvuka and to install the Kai Koro leader as leader of the Luvuka, with the title of Rokotakala. Those of the Kai Koro who came to Kese were known as the Nasukamoce, after the name of the land given to them by the Luvuka landowners.
As for the origins and development of the third element, the Leweivawa, oral tradition states that a group of people came from Vitogo, just east of Lautoka, to the island of Naviti and was settled by the Kai Suelevu at Muaira on land called Vawa. Hence they were known as the Leweivawa or People of Vawa. With the permission of their hosts, the Leweivawa accepted an invitation (veilakovi) from the Luvuka to go and join them at Kese. So the Leweivawa moved from Vawa and settled on Luvuka land at Nabouvatu.
The Leweivawa heeding the authority of the Luvuka as their landowners had not been long at Nabouvatu when they decided to install the elder of two brothers formally as leader and to give him the title of Ratu. They may have taken this formal step because of a split brought about through ill feelings by the younger brother against his elder brother, spurred on by his ambitions to take over the position of leader himself.
The resulting political construct was comprised of the Luvuka (as original landowners) and the Nasukamoce (as invited leaders, originally from Votua) forming one mataqali, the Luvuka; and the Leweivawa (originally from Vitogo) forming another mataqali, the Leweivawa.
The yavusa Kai Kese, because of its composition, is not in accord with the official model. The construction of the composite mataqali and itokatoka of this yavusa is based on principles different from those of descent from a common original spiritual ancestor. Similarly its leadership is noteworthy because over the course of time the leader has not been the senior patrilineal descendant of the original ancestor whose position would normally be legitimised on the basis of ‘divine right’. Indeed, the leader does not claim any form of descent from the original ancestor of the first settlers. He is a ‘stranger king’, after the Sahlins (1981) model, selected for leadership on the basis of personal qualities by the itaukei (literally, the persons owned by the land) and installed by them. Leadership is legitimised first by his being presented with yaqona which will pass the mana of the original spirit of the landowners to the stranger, and secondly by his being given the title of Rokotakala which would normally be given to a Luvuka person. The original landowners retain their prestige by becoming masters of ceremonial (mata ni vanua) and by retaining their right to install the stranger as the leader of their choice.
Further consideration as to why the Luvuka voluntarily handed over the leadership to the Kai Koro can be productive, bearing in mind that an important element of Fijian traditional society is the ideology of vanua with its three dimensions of people, lands and spirits. A clue may lie in the position of bete or priest for the yavusa curiously resting with the Lasawa division of the Leweivawa. Although the Luvuka, the Kai Koro and the Leweivawa each recognised their own anitu, the anitu which the Leweivawa brought over with them from Vitogo in the powerful Ba mainland territory may have been accepted as much more powerful than the local spirits. To expand this theme, mainland coastal Ba may have been considered to be a place of both spiritual and secular prestige and power, and the Yasawa islanders may have welcomed the mainlanders and their spirits as useful and prestigious additions to their local cosmology and polities.
(b) Intermediate structures: formative and expansionist federations
I have so far discussed the characteristics of the independent single yavusa as the simplest form of polity which tu vakaikoya and vakarorogo to no higher authority. It lies at one extreme of the structural continuum of polities of differing degrees of complexity which comprise the totality of traditional Fijian society. At the other extreme lies the matanitū or confederation that will be discussed later. Between these extremes investigations reveal a bewildering variety of ephemeral or opportunist associations of yavusa, parts of yavusa or groups of yavusa, as well as formative, expansionist and more stable federations. These intermediate structures represent vanua or federations at different levels of political complexity, and studies show how the process of federation is subject to varying degrees of fusion and fission, development and decline.
A vanua is a socio-political federation comprising polities at two or more socio-political levels of hierarchy. Such a hierarchy is one of social and political status, as demonstrated on ritual and ceremonial occasions and in making decisions on the extent of participation of the vanua in activities, be they social, political or military. These may be co-operative efforts such as fish-drives or building a house for the paramount chief; matters of mutual interest with other vanua such as alliances or the exchange of goods (perhaps salt and fish from a maritime vanua, and yaqona and root-crops from an inland vanua); or actual military assistance in times of war. The leader of the most senior yavusa among yavusa, by virtue of power in war or as a protector of weaker yavusa, or through prestige both political and spiritual, would be recognised as the paramount chief of the vanua. Such a leader would be independent and heed no higher authority, except when a vanua forms part of a matanitū, in which case the head of the vanua would heed the authority of the paramount of the matanitū. Studies show that there is a variety of ways that seniority and ranking are recognised in the development of social structure including leadership of a vanua. I will now discuss how a yavusa comes to be the senior yavusa, and an individual person becomes the leader of the vanua.
In a more complex polity, the component yavusa or part-yavusa at any level of hierarchy within the complex may be descent groups or socio-political constructs or both. The groups of yavusa may be bonded by myth, blood or marriage, or they may be socio-politically bonded. This bond or ivau may have arisen or been created in various ways described in more detail later. Perhaps the strongest is the Veitauvū or spiritual relationship between original ancestral spirits. More mundane but equally binding are bonds created typically by common agreement for mutual protection and assistance; or on the basis of obligations of conquered or subservient groups (qali) to heed the demands by the paramount for food, goods and services (vakatadumata); or on the basis of military assistance provided, when the occasion demands, by sympathetic neighbours at the request of the paramount; or by groups in the more formal relationship of bati or military alliance with the paramount. Such demands or requests would usually be conveyed formally by the presentation of a tabua or whale tooth.
In the case of a more complex polity in which the member yavusa are bonded socio-politically, political dominance determines the ranking and principles of leadership. So long as a yavusa remains an independent polity (tu vakaikoya), leadership is ideally based on inheritance in accordance with the principles of veitarataravi vakaveitacini. The assumption of these principles is that the person recognised to be the most senior in descent from the original spiritual ancestor is the person most closely associated with the spirit and thereby having the greatest amount of mana or spiritual power, and is the person most qualified to be leader.
Once a polity becomes more complex and members are bonded according to socio-political principles other than seniority of descent from a common ancestor, leadership becomes based schematically on a hierarchy of leaders, graded in relation to each other according to their relative secular power (kaukauwa). The leader of the dominant group would become the leader of the complex polity by virtue of his superior kaukauwa. In effect, succession to the leadership of a political complex generally depends on a person's ability to demonstrate and maintain secular power over rivals, by political or military dominance. Leadership is no longer a divine right but an institution. Installation (veibuli) is a ceremony for the general recognition of the new occupier of the office of leader, after his selection by groups within the polity. Such ceremonies involve especially the presentation of yaqona for the leader to drink and thereby acquire the mana which provides a spiritual legitimisation to his secular power. The overt aspects of these ceremonies emphasise the institutional nature of leadership. The yaqona is prepared publicly in a wooden tanoa, from which a rope of sinnet is attached. This rope is stretched to point to the leader. The first bowl of yaqona is offered to the leader. Once the leader and his mata ni vanua or ceremonial officer have drunk, the tanoa is declared empty. No one else partakes of this first round of drinking. In such ways the difference between the new leader and his possible rivals is emphasised. The people participate in these ceremonies by clapping and chanting together. This manifests a feeling of unity between the people and the leader and confirms their support and allegiance as well as their acknowledgement of his paramount authority. The appointment is also regarded as having been legitimatised spiritually through the yaqona ceremony.
In some cases, the paramountcy of a socio-political hierarchy may take account of what is recognised as the relative power of the spirits of those in a complex polity. For instance, Limasa, the war spirit of the Nakovacaki people of Nadi, was regarded as very strong. When the Navatulevu people came down from the hills and settled in Nadi, they were regarded as very powerful militarily and took over the leadership of Nadi from the Nakovacaki. To validate their superior position spiritually as well as militarily and politically, they asked the Nakovacaki for Limasa to be their war spirit. Later the neighbouring Navo people asked if Namama, Limasa's son, could be provided as their war spirit. This created a bond between the polities, and was one of the elements on the basis of which the Navo felt respect towards the Navatulevu.
Once the position of secular leader becomes more institutionalised and succession depends on military and political dominance and mutual acceptability rather than on the closeness of descent to an original ancestral spirit and the amount of mana thus acquired, the accepted principles of leadership become more flexible. Not only is the choice of leader potentially much wider, but also a selected leader can be the more readily toppled either by ambitious or jealous rivals who consider that they have the greater political and military support, or because the paramount has shown undue arrogance or failed to meet his reciprocal obligations to those qali or bati who have provided him with food, goods and military assistance. To topple a spiritual leader is a matter of grave concern because it would almost certainly be followed by spiritual retribution through the ancestral spirit, probably involving the death or serious sickness of the perpetrator and perhaps his descendants, until such time as ceremonies of apology and contrition (isoro) are performed to the offended ancestral spirit. To topple a secular leader may not be a matter of such concern to the perpetrator unless he has misjudged the extent of his support, in which case apologies would usually have little effect and physical retribution would normally follow. Even so, the new leader would normally still not only demonstrate overtly his position socio-politically but would also legitimise it spiritually through the yaqona ceremony.
So far I have discussed the first level of hierarchy of a vanua, and have shown how a particular yavusa comes to be ranked as the senior yavusa, and how a paramount achieves or acquires that position. The second level of hierarchy would comprise one or more individual yavusa or groups of yavusa which in turn may be termed vanua, and be referred to in this description as subsidiary vanua. In the case of one or more individual second-level yavusa, members of each yavusa would heed the authority of the head of their own yavusa, and the senior member of each yavusa would heed the authority of the paramount of the first-level yavusa. In the case of a group of yavusa which includes one or more subsidiary third-level yavusa and comprises a subsidiary vanua, the head of each subsidiary yavusa at the third level would acknowledge the head of the senior yavusa of the group at the second level. The head of the senior yavusa of the group would, on behalf of the other yavusa in the group, acknowledge the overall authority of the paramount of the vanua. The heads of these third-level subsidiary yavusa, together with the head of any other yavusa that may itself be the head of a group forming a third-level subsidiary vanua would acknowledge the authority of the head of the second-level subsidiary vanua. The same principles apply if there is a lower socio-political level of hierarchy.
These principles are illustrated in the following diagram:
YAVUSA A(i) YAVUSA A(ii)
YAVUSA B(i) YAVUSA B(ii)
Yavusa A is the senior yavusa of the vanua, and their head is the paramount. Yavusa A(i) and Yavusa A(ii) are individual single yavusa, the heads of which heed the authority of the paramount. Yavusa B is the most senior of a group of yavusa, the subsidiaries of which are Yavusa B(i) and Yavusa B(ii). The heads of these subsidiary yavusa heed the authority of the head of Yavusa B, who in turn heeds the authority of the paramount.
In reality, in groupings and federations of yavusa it may well occur that not all the component mataqali of a yavusa heed the authority of the leader of that yavusa or the overall authority of the leader of a group of yavusa at the same level of hierarchy, or indeed the authority of the paramount of the federation. Characteristic of the development of polities is the fission of a part of a yavusa from the parent yavusa and consequent fusion with a different yavusa. The case of the Kai Kese yavusa already discussed was that of a typical simple political construct, with which there might or might not have been some previous connection, by blood, marriage or veitauvū (relationship between original ancestral spirits).
The following examples from the Navua area of the southern coast of Viti Levu illustrate the working of the principles and mechanics of fusion and fission which may be regarded as characteristic of the creation and development of formative socio-political federations or vanua. I choose this area which comprises the yasana of Namosi and Serua partly because it is some distance east of the major socio-political complexes of Bau and Rewa and is away from the traditional areas of influence of these two matanitū; and partly because, although these yasana are not included in my specific field areas, I am familiar with the polities in the area and visited all their villages at least once at the time I was District Officer Suva/Navua.
In the Navua area there are three vanua with associated spheres of influence. The vanua of Namosi is based on the chiefly yavusa of Nabukebuke, and its sphere of influence lies generally to the east of Navua and up the Navua River to the mountainous main village known as Namosi. The vanua of Serua is based on the chiefly yavusa of Korolevu, and its sphere of influence lies to the west of Navua along the coast of Viti Levu. It includes the island of Yanuca and the main village on the offshore island of Serua, and inland towards the mountain range peaked by Tikituru. The vanua of Dravuni is geographically between those of Namosi and Serua, but confined to the coastal area around the mouth of the Navua River.
The vanua of Dravuni comprises two yavusa. One yavusa, the Deuba, includes the original people at the mouth of the river who were fishermen of unknown origin or interrelationship. The other, the Dravuni, is a construct of various groups of people who came down from the mountainous interior of Viti Levu at different times. As these groups arrived, they each in turn agreed to join together as one yavusa under the leadership of the first group to arrive. Through marriage, a group of the Dravuni joined the Korolevu at Serua. Meanwhile the fisherfolk had settled on land and formed the yavusa Deuba, and had federated with the remaining groups of the Dravuni.
The vanua of Namosi was based on a descent group with nine sub-groups, living in the north-eastern interior of Viti Levu. Following attack by a neighbour, they split up. One group went south and ended up in Namosi. Here the son of the leader quarrelled with his father about some coconuts, and ceremonially burned Namosi village, after telling his father to evacuate. This is known as buka vakaturaga or chiefly burning without bloodshed. Some of the father's followers were recalled to build Namosi, while others went and joined the Dravuni people at Navua by the sea. There were continual family quarrels about the leadership, and another group from Namosi settled near Navua, but inland from the coast. Another group joined the Korolevu people at Serua where they were given land.
The vanua of Serua was based on the Korolevu yavusa, formed of Noi Koro people who had come down from the interior of Viti Levu from the upper reaches of the Sigatoka River. They settled on the coast, basing themselves on the island of Serua. They were attacked continually by their neighbours, and were allowed by the Dravuni people to go to Navua and settle there in peace, provided that they acted as labourers for the Dravuni. The Korolevu became so dominant towards their hosts that the Dravuni people left and went to the Namosi people to ask for help against the Korolevu. The Namosi drove the Korolevu back to Serua, and the Dravuni returned to Navua. The Dravuni people thanked the Namosi people ceremonially for their services, but the Namosi people proceeded to demand that the Dravuni paid them regular tribute, eventually including women. At this stage the Dravuni became resentful and sought the help of the Korolevu people to drive the Namosi away. By now the Namosi had become so powerful that they threatened to dominate not only the Dravuni but also the Korolevu. Fighting then broke out and the Namosi attacked not only the Dravuni but also the Korolevu at Serua. The war lasted for some years until the Namosi were finally driven back to their own territory (for details see Appendix B).
This account illustrates the procedures for the creation and dissolution of alliances in a situation such as when two relatively small but powerful polities, Namosi and Serua, have spheres of influence bordering on a weaker polity, the Dravuni, living between them. Each powerful polity in turn gives the appearance of helping the weaker polity against the excessive demands of the other powerful polity. In doing so, each powerful polity takes advantage of the situation to extend its own sphere of influence over the weaker in the first instance but eventually over the other powerful polity as well. It also raises the question of the real reason for family quarrels leading to splits in polities, a situation frequently met in those accounts of polities which I have recorded. Excuses such as a quarrel over coconuts or, as I have noted, receiving the wrong piece of a pig or a fish, probably conceals an ongoing and smouldering political ambition and jealousy on the part of a son or a half-brother against a father or more favoured half-brother.
Even at a high level of development, the basic concept of vanua with its elements of persons, place and spirits remains crucial. Important considerations in respect of these higher levels of complexity of development and structure concern how the principles of unification and integration of a polity were developed and maintained in practice. These considerations concern not only the extent to which the principles of hierarchy of sub-groups within a descent group and of seniority of birth within a sub-group were adhered to in practice; but also to the degree of flexibility within these principles. In particular, such considerations concern, first, those cases of divergence from such principles which occurred by general consensus. For instance, an elder brother might voluntarily pass the position of leadership to a younger brother, or a respected or related outsider might be preferred by all. Secondly, and perhaps even more importantly in relation to the circumstances of warfare and socio-political expansion prevailing before Cession, these considerations concern cases of divergence from these principles which occurred on the basis of the realities of political expediency or of superior secular and perhaps spiritual force.
(c) The most complex structure: the matanitū
In contrast to the independent, simple social group or yavusa, and at the other end of the socio-organisational continuum of late Fijian social polities stands the highly complex socio-political confederation or matanitū.
The most complex form of socio-organisational confederation is found in the eastern parts of Fiji, especially in areas which have come into close contact with influences and ideologies originating from outside Fiji, particularly those of monarchical or near-monarchical Tonga and perhaps also the highly stratified and status-differentiating Samoa. Such a confederation may be expected to remain in power, provided that it can resist the ambitions of a more powerful rival matanitū or of an external power such as Tonga, or can enlist the assistance of friendly matanitū, or turn to its own advantage the forces of an external power such as Tonga.
Tabular outlines of the main elements making up the highly-complex socio-political confederations or matanitū of Bau and Rewa will serve to illustrate the general character of this apical institution in comparison with the other side of Viti Levu. The crucial problem for the paramount of each of these polities was how to control a wide-flung sphere of influence by means of diplomatic and administrative arrangements backed up by easily transportable military might, and by judicious use, when appropriate, of spiritual legitimisation of leadership and decision-making (for details see Appendix A(i) and (ii)). The terms bati and qali used in these tables and elsewhere will be discussed in some detail later in this chapter.
Tables of structure of Bau and Rewa confederations
(i) The Polity of Bau
The Chiefly Core (Kubuna) and Officials
Roko Tui Bau (spiritual chief)
Masau (Paramount's personal herald)
Tunitoga (Kubuna herald)
Takala (law and order)
Vusaradave (bodyguard of Vunivalu)
Bouta (personal servants of Vunivalu)
Kai Kaba (maintain canoe fleet)
Viwa (eastern island)
Navuloa (base of Kaba peninsula)
Each element cited above vakarorogo or heeds the authority directly or indirectly of the Vunivalu or war chief of Kubuna, being at Cession the paramount of the polity of Bau. It will be recalled that the Vunivalu had earlier deposed the Roko Tui Bau or spiritual chief as paramount. Most of these elements, especially the core and peripheral craftsmen, are themselves complex polities with a leading chief, a leading yavusa and several other yavusa, the leaders of which in turn heed the authority of the overall leader of the complex polity. I have not included a number of elements of the Rewa polity, which, after the 19th century wars between Bau and Rewa came, as the result of Rewa's defeat, to be part of the Bau polity. This included some of the bati such as the Nakelo, the Tokatoka and the Waikete who changed sides when they saw how the wars were going to favour Bau.
(ii) The Polity of Rewa
The Core (Burebasaga)
Roko Tui Dreketi (spiritual chief of Narusa)
Vunivalu (war chief of Nukunitabua)
Sauturaga (6 groups of landowners, who led the army and acted as heralds)
The Qase or ‘old men’ (special supporters of chiefs)
The Kaso (descendants of young chiefs), including the Kai Nalea, who were priests)
Qalitabu (no duties to chiefs), responsible for installing the paramount. They may have been earlier settler chiefs or a senior branch of present chiefs but had lost paramount position during internal upheavals.
Nukui and Naselai (fishermen)
Nabua, Sigatoka (Tongan sailors)
Nakelo (12 villages), with Kuku and others subservient to Nakelo
Noco, with Tui Noco and several villages
Tokatoka (7 villages)
Beqa Island, including Raviravi (4 yavusa) under Tui Raviravi;
Sawau (6 yavusa) under Tui Sawau: and Rukua (1 yavusa).
Kadavu Island (parts)
Kai Batikeri (comprising Kai Nadoi and villagers of 3 villages, Nakuru, Drekena and Veiniu)
Qalivakawai (comprising 3 groups of Kai Narocivo, Kai Tavuya and Kai Nateni, living in the mangrove swamps)
Each element cited above vakarorogo or heeds the authority directly or indirectly of the Roko Tui Dreketi of Narusa, Burebasaga, being the paramount of the polity of Rewa. Most of these elements, especially the core and peripheral craftsmen, are themselves complex polities with a leading chief, a leading yavusa and several other yavusa, the leaders of which in turn heed the authority of the overall leader of the complex polity. As mentioned above, a number of elements of the Rewa polity came to be part of the Bau polity as the result of Rewa's defeat at the end of the 19th century wars between Bau and Rewa. This included some of the bati such as the Nakelo, the Tokatoka and the Waikete which changed sides when they saw how the wars were going to favour Bau. The inclusion of Tongan sailors is noteworthy with regard to the question of the extent of the importance of Tongan influence in eastern Fiji through ideology, military power or, in the case of Rewa, through naval assistance. Bau had the final advantage over Rewa because of the size and strength of its naval fleet, but Rewa made good use of its fleet (including its Tongan sailors) when keeping its island qali of Beqa and part of Kadavu under control.
The main conclusions to be drawn from these two accounts are that in reality:
- • simple local polities in the east developed into major polities which became widespread geographically, and highly complex socio-politically, with varying levels of groups associated with the paramount group through an system of alliances and tributaries, resulting from conquest and from voluntary association based on judicious marriage, common origins both ancestral and mythical, mutual aid and protection;
- • able and ambitious leaders, through judicious marriages, military (including naval) power backed by spiritual power, and a competent system of socio-political communication, could expand and maintain their sphere of influence geographically and politically;
- • the personal position of the paramount could become dangerously precarious either if they were too arrogant and tyrannical, or if there was internal treachery among relations, especially half-brothers, who considered that they should hold the position of paramount. Continuing potential tension between loyalty and expediency on the part of allies, and between loyalty and rivalry on the part of relations, could affect the stability of the polities; and
- • the indirect and direct influence of Tongan political ideology and military and technical assistance could be of particular importance in the development of major polities. Ambitious paramounts, through visits to Tonga or through visits to them by Tongans to whom they might be related by blood, had become aware of socio-political developments and ideologies in Tonga, and might well seek to emulate the semi-monarchical position of Tonga. They could also seek to obtain assistance from Tongans, either as craftsmen or sailors or warriors.
Waxing and waning within a complex polity: Na iVau kei na kena iSereki
The fortunes of the complex polities in Fiji which I have described would have waxed and waned from time to time as I have indicated. Nevertheless each of these polities and especially the matanitū managed to maintain, at least for a while, a certain level of stability even in times of trouble. This level of stability may have enabled a polity on the wane to survive and re-assert itself as a power or at least to command respect from its neighbours, respect based perhaps on memories of its former glories and wider powers. Stability would depend on the bond between the component polities and between their respective leaders. Emphasis was and is still placed on the importance of the bond (na ivau) in the course of ceremonies involving the presentation of tabua and yaqona. Mata ni vanua while accepting the presentation would often use the expression ‘Me tu dei tiko ga na kena ivau’ or ‘May the bond between the donors and the recipients be always maintained’. In smaller polities the bond may be that of blood (dra). In descent groups, the theoretical bond is by definition that of common blood, and the bond is referred to as ‘dra vata’ or common blood. Members who are dra vata claim to be descended from a common original ancestor, usually along the male line. In some parts of Vanua Levu, the basis of common descent is along the female line. An extension of the concept of a bond based on dra vata is the concept of a bond of relationship based on common spiritual blood. Such a bond is based on a spiritual kinship relationship between two original ancestral spirits. The basis of the bond may have been two brothers who came to be regarded as the original ancestors of two separate descent groups of the nature of yavusa; or a brother and sister, in which case one may have been an original ancestral spirit and the other the spouse of another original ancestral spirit. A relationship based on descent from two such related ancestral spirits is referred to as veitauvū, and people in such a relationship address each other as ‘Tau’. A bond created by the relationship of veitauvū is considered to be very strong.
An interesting question concerns the nature of bonding when the basis of the bond is less straightforward than that of descent from a common original ancestor or related original ancestral spirits. In such case, the bond must become more a socio-political matter than a family issue. The creation and maintenance of the bond was especially important in such highly complex polities as Bau, Rewa, Cakaudrove and Verata (see Appendix A(iii) and (iv) for the latter two) which had not confined their spheres of influence to areas that were geographically contiguous.
For example, according to Sayes (1982), Cakaudrove's first power base was at Wainimosoi on the banks of the Waikavu River on south central Vanua Levu, and here the first Tui Cakau was installed as paramount chief of a motley construct of locals and visitors. Following a split, the leading chiefs went to Taveuni and the others followed in about 1820. The influence of the matanitū of Cakaudrove then extended from its new power base on the island of Taveuni along the south coast of Vanua Levu and to the islands to the east of Taveuni. Similarly, Rewa spread to the island of Beqa and part of the island of Kadavu, while Bau spread to Lomaiviti and to Vanua Levu.
Any discussion of the development and structure of these great socio-political complexes of the east must inevitably involve some consideration of devices developed and introduced in order to maintain their stability or, as the above-quoted ceremonial expression goes, ‘me tu dei tiko ga na kena ivau’.
As these complex polities expanded, the position of the mata ni vanua in his formal role as master of ceremony and mouthpiece of the paramount became increasingly important. In these complex polities, the mata ni vanua was installed at a ceremony and given a title: at Bau, Tu ni Toga, Naitaka, Mataisau; and in Cakaudrove, Ko Mai Kavula, Ko Mai Nanukurua and Ko Mai Naitala. Equally important were his more informal duties of guarding against breaches of protocol which might have led to offence, and of keeping himself and so his master aware of grumblings (kudrukudru i Ra Mo or grumblings by people dissatisfied with the orders of the chief, but who heeded them in silence in his presence) which might have led to plotting (vere) and instability within the polity. Similarly, the role of mata ki or ambassador was created to enable the paramount to maintain close contact with the far-away periphery of his sphere of influence. Such a person might have been a relation of the paramount and certainly would have been a person of his choice and worthy of his trust. He would act particularly as the representative of the paramount for the purposes of passing on demands for goods and services, and of ensuring that such demands were met. He was also expected to act as the ears of the paramount, and to anticipate any plotting or signs of disobedience or treachery. The bond was also maintained by the judicious use of political marriage. Either the paramount would provide a bride for the local chief, or the local chief would provide a bride for the polygamous paramount. The resulting vasu relationship has been discussed above. The paramount was also liable to go and pay visits to his periphery, sometimes to remind a wavering group of his military strength, sometimes to gather tribute from recalcitrant tributaries, or sometimes simply to ensure that good relations were maintained.
As socio-political complexes, both vanua and matanitū developed with increasing levels of hierarchy and, as they expanded geographically, more formal, more lasting and stronger bonds were developed for administrative necessity and for the purposes of maintaining stability. In this way certain recognised categories of relationship developed between the paramount and the associated polities. The Fijians describe the basic relationship neatly by referring to a paramount as one who tu vakaikoya (or stands by himself) and to the associated polities as those which vakarorogo or heed the authority of the paramount. In considering the polities associated particularly with Bau and Rewa, the relationship between these and the paramount is distinguished between those referred to as the bati and those referred to as the qali. The relationships are termed veibati and veiqali respectively.
The bati are referred to as the ‘borderers’ by Capell (1941:6), and this is probably as good a descriptive term for them as any. The term is used when alluding to a form of relatively stable relationship based on military alliance that developed between paramount and powerful neighbouring polities which were recognised as bati. Before the Bau/Rewa wars, the most powerful bati associated with Rewa were the Tokatoka and the Nakelo who controlled the major waterways in the Rewa delta. The Bau bati included the Namaka, Namuka, Dravuni and Navuloa who were scattered strategically around the Bau heartland. During the Bau/Rewa wars, Tokatoka and Nakelo were excellent examples of how bati were liable to change sides, according to what they saw to be to their best advantage, and depending especially on how they were treated by Bau and Rewa respectively. The bati lived by their weapons. When the Roko Tui Dreketi of Rewa sought their assistance in war, he would send a message to them and give them presents and request their help. He could not simply order them to fight. In return for what they considered to be adequate respect and presents of food or goods, they would offer their services to fight for him. If not so respected or properly presented with gifts, the bati were liable to change their loyalties. Nakelo was a bati of Rewa until Cakobau, the Vunivalu of Bau, induced the Tui Nakelo to change sides, by promising him his half-sister in marriage. Cakobau then found that he needed the girl to give to Gavidi, the chief of the Lasakau who lived on Bau, in order to keep him loyal and firmly in support. So Nakelo changed sides once again, and at one stage compelled the Bau army to retreat (Waterhouse 1866:149). Bati were not subservient to the paramount, and did not pay tribute. They were essentially warrior allies, and the bond between bati and paramount remained so long as the bati were well-treated (Williams 1858:20). The term bati is found in accounts of western polities, but does not refer to semi-independent allies but rather, as in the case of the Yasawa, to a specialised warrior sub-group within a descent group. Ratu Sukuna in his model of Fijian polities does include a mataqali bati in his model yavusa. It may be that in his model he acknowledged the general significance of the term bati by using it as that for a particular category of mataqali; whereas the reality in the east was that it was a term rather for the relationship between a paramount and his military allies.
The qali was described by Capell (1941:189) as ‘a province or town subject to another’. Williams (1858:20) who may well have been the authority for Capell's description said that the qali ‘represents an area or settlement which is subject or tributary to a chief town’. Koto (n.d.) added that a chief town was that to whose chief the qali vakarorogo (or heeds its authority). Jackson alias Diaper (1853:451) was more explicit when, presumably from his own observations, he stated that qali were obliged to pay tribute (isoro) periodically. The relationship of qali was created by conquest or by fear of conquest. In the case of fear of conquest, the qali may have sought this relationship either out of fear of attack by the paramount or out of fear of attack by another power. There are for instance certain areas in Lomaiviti which are described nowadays as qali vakaBau or subject to Bau. During the period before 1874, such people would have been obliged to provide goods and food on demand to the Vunivalu of Bau as paramount as recognition of his authority. Failure to do so would have resulted in punitive measures being taken against the disobedient qali, some of whom could well have ended up in the cannibal ovens of Bau. In the case of Rewa, the island of Beqa fell to the status of qali and was required to pay tribute to Rewa because on one occasion Beqa people ate some Rewan castaways and the island was besieged successfully by the Rewans and reduced to qali status out of revenge. In Cakaudrove, the Natewa people had been associated with Cakaudrove; but after transferring their association to Bau thanks to Bauan cunning and trickery (hence the expression vere vakaBau or trickery as in the Bauan style) surrendered themselves back to the ambitious Tui Cakau after a long drawn out defence and so fell to a qali relationship with Cakaudrove, paying tribute with canoes, barkcloth and coconut sinnet cord. Tribute was sometimes brought by the tributary to the paramount, as in the case of Tui Nayau of the Lau group who brought a double canoe, ten tabua, and fifteen rolls of sinnet to Bau in 1843 (Williams 1872:411). Sometimes, the paramount would collect the tribute when on tour of his sphere of influence. Tanoa of Bau spent three years in Somosomo, Cakaudrove, collecting tribute from dependants in Lau (Derrick 1946:80); and Jackson (1853:443) described a visit to Bouma, northern Taveuni, when he went there with the Tui Cakau in order to collect tribute. In the case of Verata and its wide-flung sphere of influence beyond the seas, tribute from Vanua Levu was assembled at collecting points on the island before being transferred to Verata. A situation might have arisen where a paramount considered that a polity was a qali to him but the other polity did not consider that it was so obligated. The Vunivalu of Bau thought that Macuata was a qali and obliged to provide tribute on demand, whereas Macuata was prepared to acknowledge the power of Bau by presenting goods (iyau) on a friendly basis (solevu, see below) when it thought appropriate. In 1841, Macuata refused the demands of Bau for tribute, and Bau sent its ally, Viwa, to attack. The Viwa chiefs prevailed on Macuata to surrender, and Ritova, future paramount of Macuata, paid frequent visits to Bau, bringing gifts. Nevertheless, the formal status of the relationship remained ambivalent.
The bond between two polities depended on the basis on which the lesser polity vakarorogo (heeded the authority) of the major polity. It may have been one of military assistance on request and in return for presentations (veibati), or of tribute on demand and punishment for failure to comply (veiqali). In a highly developed socio-political complex such as Bau, a paramount had to be in a position to obtain sufficient goods (iyau) and food from his tributaries for presentation to his bati. It was through this regular presentation of food and goods that groups publicly acknowledged the authority of the paramount, and the paramount gave an outward and visible sign of his authority over the tributaries by ceremonially accepting the isoro. In the same way, a paramount could give an outward and visible sign of his respect for, and appreciation of, the bati for their continuing loyal military assistance by public presentations of goods and food. For their part, the bati could make known to all their kaukaua or secular strength by performing before the paramount the ceremonial bolebole, a fearsome demonstration of their powers to challenge all comers. Alternatively, before a battle, they could assemble and perform the taqa or display of their military might. Ostensibly this was to assure the paramount of their loyalty and their ability to meet their military obligations to him in return for his presentations to them. It was also an opportunity for the bati subtly to remind the paramount that they were indeed a military power to be reckoned with and not to be underestimated in the event of the paramount's failing to fulfill his obligations towards them.
Groups also established socio-political relations with powerful polities by the ceremonial presentation of iyau and food, accompanied by requests for assistance. It is doubtful whether these groups could be described as qali or their presentations as isoro. For instance, in 1846, the chiefs of the Cakaudrove polity went to Bau to seek military aid. They presented vast numbers of yams, a high wall of yaqona and bales of barkcloth, and then a demonstration of several thousand warriors. Thus they planned to show the Bauans how helpful they could be in the wars with Rewa, and how much produce they could make available to Bau in return for a common bond of understanding and mutual assistance in times of war. This situation and relationship was not one of vakarorogo, of Cakaudrove heeding the authority of Bau. It was one of mutual assistance and socio-political cooperation in times of need. It was a relationship that would have been confirmed from time to time by the performance of solevu, the ceremonial exchange of goods and food recognised as a visible symbol of the maintenance of a bond created or existing between two polities. The presentation of goods and food as isoro or tribute by qali reflected a relationship based on demands by a polity of superior strength to which a subservient polity refused to accede at its peril. The exchange of food and goods at a solevu reflected a relationship of mutual assistance based on good will. All these relationships, be they based on bati, qali, or solevu exchanges, reflect an awareness of the need to stabilise socio-political relationships in highly complex polities.
Paramounts must have been only too aware of factors likely to affect the stability of their positions, such as the political ambitions and plots of rivals within the polity, the military powers and diplomatic abilities of rivals on their borders, and the increasing expansionist ambitions of their Tongan neighbours whom they both admired and feared. Those, especially in the west, who traditionally recognised the authority of neither Cakobau nor Christianity were faced with what they regarded as the insidious spread of Christianity, which they coupled with an extension of Cakobau's sphere of influence. They were forced to determine how to accommodate the mana or spiritual power of the Nakauvadra and the bete or priests who mediated with the ancestral spirits, with the power of the church militant (the so-called Lotu of Tonga) and its incomprehensible priesthood, and the fierce Tongan warriors who were forcefully spreading the Lotu around the Lau group in the east and thence northward to Bua and westward to Cakaudrove and Bau and neighbouring polities, and finally as far west as the island of Yasawa. Any one or any combination of these factors could cause developments in these polities whereby the bonds so forcefully or diplomatically forged were loosened or broken. As in the case of Verata (Appendix A(iv)), a highly developed socio-political complex can become unstable and virtually disintegrate to a much more modest form of polity, albeit still highly respected for its past glories.
The fascination of exploring the development of, and interrelationship between traditional Fijian polities prior to Cession (1874) lies in the great variety of socio-political structures at the time; the factors both internal and external leading to such structural variety; the situations involved in the hierarchical ranking of sub-groups in groups at all levels; and the principles and practices related to the acquisition or achievement of leadership. Equally fascinating is the investigation of the symbols of identity of a group; the nature of bonds created between groups of the same level of structure and between those of different levels; the factors leading to the breaking of such bonds; the ceremonies, both spiritual and socio-political, recognised and accepted as invalidating the installation of a leader at all levels; and the ideals and realities for procedures for a change of leader or for his disposal. I have discussed these points in relation to the three levels of polity described earlier—the yavusa, the vanua (that is, vanua 2) and the matanitū. I have taken into account the three elements of the concept of the term vanua 1 (discussed earlier) which run through traditional Fijian society at all levels—spirits, places and people. The three following Fijian expressions reflect the very heart of my discussions in this chapter, and summarise the nature of polities, relationships, ranking and leadership.
‘Me tu dei tiko ga na kena ivau’ or ‘May the bond between us last for ever’ is the customary culmination of a ceremonial speech for the presentation of yaqona by one polity to another, giving assurances of a continuing association of friendship or loyalty. The yaqona represents the spiritual element, and na ivau refers to the people involved.
‘Me cecere tu ga o Uluda se Vunisei’ or ‘May the chiefly yavu or housemound named Uluda or Vunisei remain for ever high’ is an example of a recognised culmination of a speech of presentation of yaqona, seeking spiritual assurance for the stability of the current paramountcy. Uluda is the name of the yavu of Tu Navitilevu, paramount of Rakiraki; and Vunisei is that of the yavu of Momo i Vuda, paramount of Vuda. The yaqona again represents the spiritual element. The chiefly yavu symbolises collectively the people of the polity, and the superior height of the yavu symbolises the paramountcy of the chief whose place it is by virtue of his paramountcy.
‘Na ivau kei na kena isereki’ or ‘The bond and its becoming untied’ is a perhaps wistful expression to signify an acceptance of the often harsh realities of fusion and particularly fission in polities, in contrast to an idealised stable traditional Fijian society with recognised principles of hierarchy, leadership and authority based on primogeniture and seniority of descent.
1 (Editors’ note) We have not been able to find this exact reference in Williams’ writings. It may represent a later edition of Williams (1858).