The origin narratives which at the same time tell both about the migration of the original ancestors from the mythical homeland to the present day islands and give their genealogies, create the qualitatively separate island populations. The role of the resulting genealogical accounts consequently is not only to provide a genealogical link to the ancestor, or even to his divine forefathers, but to convey the qualities connected to the ancestor in question. This in turn can ensue in several ways, and accordingly there has to be several kinds of genealogies.

The genealogical representation of the present day population on the islands is a complex one. We do not have a simple distinction between historical narratives and uniform genealogical accounts, and even the genealogies are of varied kinds. My argument is that we have to make a clear distinction between genealogical information on the basis by which we as anthropologists are able to construct extremely encompassing genealogies and different genres of genealogies as they are used and recited on the islands. The politically significant ones are these genres, and not the possible sum of information in our notebooks. To draw conclusions based on comprehensive reconstructions in the way of the early anthropologists and colonial administrators is, of course, a legitimate pursuit. What is not tenable, however, is to claim that these kinds of conclusions are based on local genealogical representations of the society. For this purpose one has to look for the genealogies as they are really used.

Visually and logically, the simplest kind of genealogy is a straight list. It begins with the name of one of the founding ancestors, lists all the intermediating ancestors, and ends with the individual whose ascent is being represented. The term used for such a genealogy is ara, the road or the path. To give an example from a Ma‘ukean familybook, one ara reads:

Te ara is Koumu (The road of Koumu)
Tekeunui
Tekeumaina
Upokonui
Teenui
Tekatake
Atiuru Upoko
Utatakienua
Oe
Utaavarau
Tangaavarau
Koumu

Koumu’s road to an ancestor called Tekeunui is thus represented with the help of a list of names which is claimed to be a genealogy. Te Rangi Hiroa describes this kind of genealogy in Tongareva: “a particular line of descent from a specified ancestor is called ara (a path) … A person who is descended from more than one member of that family has more than one path (ara) to [a specified ancestor] and in a recital runs them down in order of seniority” (Buck 1932). In the Cook Islands, especially in the minute books of the land court, numerous examples are to be found in which an individual’s “path” to a common ancestor is given through a number of different “paths”. This is especially important in the case of candidates for a chiefly title. A combination of several paths can be claimed to provide a candidate with a stronger case, and the production of these combinations naturally requires more comprehensive genealogical knowledge, too. Rongomatane Ngakaara presented five versions of his genealogy as “different paths” in an argument over his chiefly position to the Land and Titles Court in 1903 and all of them were recorded as “genealogies of Rongomatane through different lines of ancestry”.

The interesting point in these chiefly paths is the way in which they combine the ascent to different original ancestors. So we have a picture in which the three chiefs of the islands construct their paths to the two ancestors of their own island.

The paths give a clear image of the independent origin of ‘Atiu and Ma‘uke. Only Mitiaro seems to be dependent. The paths lead, however, also to the ancestors of the neighbouring islands. These paths can be used as supporting evidence for a chiefly candidate’s claims. Alone, the connections to “wrong” ancestors are not sufficient; on the contrary, vague links to the ancestor of one’s own island and strong links to the ancestor of the neighbouring island can be used as counter argument. In the case of ‘Atiu and Ma‘uke the autochthonous origin of chiefly titles is a rule. As Tura is the hierarchical descendant of the ‘Atiu chiefs, so Uke is his Ma‘uke counterpart and they are the end points of the paths of the genealogies of their islands.

The ara form of genealogy seems to correspond closely to the western concept of descent. The focus of the paths is not, however, unilineal descent, but ascent, and the path can, in fact, not be directly translated into inheritance rights, although this legalistic reinterpretation of paths of ascent as lines of descent is now the norm, according to which the people in the Cook Islands have adapted their own interpretation of their genealogical data. The connotations of ara have begun to deviate even from the Tikopian ones described by Raymond Firth. During the ritual of the sacred canoes in generally patrilineal Tikopia, ara is a term used to refer to the relationship between mother’s father and daughter’s son, between Kafika and Taumako clans (Firth 1967:135, see also Hooper 1981:19). In eastern Polynesian languages ara alludes to progeny and birth in general, without any patrilineal connotations (see Koskinen 1963:68-69).

The path is not, however, sufficient even at the level of a formalized representation to account for the continuation of life on an island. As important as the path is, birth and marriage ties begin to play havoc in the minimalism of name list genealogies (see also McKinnon 1990). The single path multiplies itself, and the possibilities of choice and the political pragmatism connected with the preferences become central. At the same time all the connotations of the path become clear, adding further arguments against the legalistic descent-line interpretation of ara.

In addition to the simple “path” form of genealogy, there is another formalized mode of genealogical recitation and, since the coming of missionaries at the beginning of the last century, also writing. This more complex form is not a single path, but a combination of generations and marriages. It is produced patrilineally around the expressions “so and so no‘o ki te vaine so and so, anau tana …” (“so and so lived with the woman so and so and so and so was born to them”) and the result is a genealogy called papa‘anga. The term refers to layers or growth as does its synonym tupu‘anga. The addition is a minor one, but it brings a totally new dimension to genealogy: marriage ties. In the simple path mode the sex of the generations is insignificant. In my calculations I have found that about 30 per cent of the names mentioned in the “paths” of the chiefs are female, but the sex is never mentioned. First the no‘o ki te vaine … anau tana mode of genealogical representation gives some patrilineal emphasis to the genealogies (if it is not expressed in the form no‘o ki te tane, lived with a man …).

In the political context of Ngaputoru marriage ties and thus reproductive ability play an important role. The origin narratives do give an asymmetric image of these abilities to the first generations. Uke finds a wife on his journey to the new island of Ma‘uke. In a significant manner this wife is not a virgin; on the contrary, she is a wife of a warrior whom Uke kills in a fight. Uke’s wife also has a son from the killed warrior, and this son becomes the husband for Uke’s first born daughter. Uke thus brings to his island all the prerequisites for society: edible plants, rituals, wife, and husband for his future daughter. On ‘Atiu in turn, Tura comes alone and has no wife. Accordingly, ‘Atiu has to rely on Ma‘uke for female reproductive powers, and Tura has to marry the second born daughter of Uke. Thus Ma‘uke’s position is an ambivalent one: on the one hand, the island is more complete, and thus hierarchically above ‘Atiu; however, in the context of Ngaputoru politics it acquires the status of a tributary wife-giver, and thus becomes hierarchically below ‘Atiu. The genealogical situation between the islands begins to look like this:

According to this genealogical construction Moenau seems to be the representative of the senior line on Ma‘uke. He also appears on all the genealogies of the Ma‘uke chiefs in a prominent position. The two marriage ties around the daughters of the younger Ma‘ukean line represent the decisive alliances, which finally determined the power on the island. The intermarriage of the chiefly lines of both islands, however, creates a situation in which all three chiefs of both islands have paths to both of the mythical ancestors, both to Uke and to Tura.

Marriage, although clear on the level of genealogical representation, is not so systematic in the Ngaputoru context after all. The genealogies give reason to wife-giver/wife-taker analysis and to the consequent emergence of hierarchical levels. In fact, marriages are not transactions connected with social prestations, gift-exchange, etc. The significant way of “living with a woman” (no‘o ki te vaine) in the historical narratives is to visit a woman secretly, possibly a wife of somebody else, and have offspring. The recurring feature of the narrative accounts about the deeds of the heroic chiefs in the Cook Islands is that the chiefs have heard about a beautiful wife on another island. This competition for women is the leading force behind the raids and interisland wars. A paradigmatic example in the Ngaputoru political context — an example which still creates hectic disputes wherever Ma‘ukeans and ‘Atiuans meet — is the case of Akaina. He was Parua Ariki of ‘Atiu, and had heard about a beautiful wife of a Ma‘ukean chief. So he decided to sail over to Ma‘uke to fulfil his desire. The narrative tells us how Akaina arrived on the shore and met there the husband of the woman. In typical ‘Atiuan fashion, their version tells how Akaina replied to the inquiry about the purpose of his visit: “I heard about a beautiful woman named Eturere. I have come to sleep with her.” To this the husband of Eturere replied: “Welcome, Akaina, we will go to the house”. Without opposition he gave his wife to Akaina, who slept the night with her. The men of Ma‘uke, however, were not quite satisfied with the situation, and organized a war party to kill Akaina. In this they succeeded, and the killer was the paramount Ma‘uke chief Mana’s warrior. Mana’s younger brother in turn helped the younger brothers of Akaina to escape and return unharmed to ‘Atiu. In consequence of this love affair and ensuing fight the ‘Atiuans revenged, and transferred the chiefly title on Ma‘uke from the tuakana to the teina, from the elder brother Mana to his younger brother Tava, the helper of Akaina’s younger brothers.

In the case of Akaina the result was not birth, but death. Both can have effects on the chiefly succession and the emergence of hierarchy. It is notable that successful genealogical use of the secret deeds of a travelling chief combines path and birth, and gives path its literal meaning, i.e., the way the ancestor has been travelling. In the history of Tangiia, written down by Te Rei Tamuera and Stephen Savage, we have an example of the way the combination is made. In this narrative Tangiia arrives from the sea to the coast of Huahine. There he is sighted from the land, and an attempt is made to prevent his landing. According to the narrative, Tangiia composes an aka-enua song:

(Savage 1907:13)

Travelling Tangiia is really “on the path” of his ancestors. The aka-enua is a song composed to demonstrate his kinship ties to the island where he is trying to land. The decisive proof is one’s position on the same path, the ability to trace the way the ancestors have travelled — even secretly. The path, the way the ancestor has travelled, and the descendant’s attempt to trace that path are actually a Polynesian type tale. The typical tale “tells of a high chief who weds away from home and departs, leaving tokens with the mother for the child about to be born by which the child’s paternity may be recognized” (Beckwith 1970:478). So the ara is not the descent, but rather the path, of the ancestor. Through their travels the warriors and chiefs create not only genealogical links between different islands, but at the same time a differentiated political space, which is hierarchically organized; and this hierarchy is expressed in genealogical terms. In a dispute about the seniority of rival islands claiming genealogical connections to the same heroic ancestor, the order of his visit to the islands is decisive. The ara of the islands is dependent on the route or path of the hero.

The narrative about Akaina is a paradigmatic one because it reveals the constantly recurring pattern, according to which the marriage ties between Ma‘uke and ‘Atiu affect the chiefly succession. It is not only the hierarchical relationship, but also the gendered values of the different islands which are the determinants of the political power. ‘Atiuans as males and lonely warriors of the original genealogical situation reproduce their position by constantly travelling to Ma‘uke “to prey for food and women” as Ron Crocombe put it (1967). At the same time the Ma‘ukeans proliferate their position as senior and female.

The marriage alliance in this situation is a strange one: it is an alliance without marriage. There is the travelling male and the beautiful woman of the land whom the male visits. He does not stay, but continues his travel, and the paths of these travellers form a web of relationships around the archipelago. The hierarchical position of the island societies in the total polity is determined by the combination of the paths and the births, travels and sexual values of the islands.

In the social situation of the origin genealogies of Ngaputoru, only Ma‘uke is complete and able to reproduce itself. ‘Atiu, in contrast, is given as a lonely warrior who has to conquer a wife for himself on some other island. This leads to the original marriage between Ma‘uke and ‘Atiu, and the conceptualization of the island communities continues to give them the values of a husband and a wife, a man and a woman. ‘Atiu is claimed to be the island of warriors and Ma‘uke the island of va‘ine purotu, the island of beautiful women, whose beauty is praised in numerous songs.

The conquering of the Ma‘ukean women by ‘Atiu warriors has not, however, occurred against the will of the Ma‘ukeans. In the internal classification of younger and elder lines the connections to the warriorlike ‘Atiuans play a decisive role. The internal chiefly rivalry constantly seeks a means of legitimation outside its borders. The interplay of the values of female and wife-giver, male and wife-taker result in the intervening fame of the women’s beauty or the event of outside wife-conquering. The alliances created on this basis are unstable ones, and can be understood only as valued by the precedence of a whole set of categorical operators (Fox 1994). When wife-giver is defined as female, it is seen as inferior to the conquering male wife-taker. When the chiefly family is not seen as wife-giver but a senior beautiful woman whose beauty the travelling chiefs can not resist, the external wife-conqueror is utilized to support the claims of the internal chiefly line in question. The set of operators thus includes the distinctions between elder/younger, female beauty/male conqueror, autochthonous/foreign. The complementary combination and consequent reversals of the values of social units on the basis of the application of these categories forms the dynamics of chiefly politics.

In the case of Akaina, the killing of the love-hungry visitor led to the transfer of the title from one line to another. Similarly, the giving of daughters as wives to ‘Atiu chiefs gives the candidates for a title a decisive advantage: the support of the island of warriors.

The continuation of this situation has not led to a complete and genealogically legitimated ‘Atiu dominance in the archipelago, despite the early European impressions and some later interpretations. The path of the Ma‘uke chiefs guarantees them their seniority because they can trace their path through the seniors, and the path of the ‘Atiuans always passes through the juniors. But this is so only if the path is traced to Uke, the original ancestor of Ma‘uke. The place and the ancestor are tightly connected, and the path metaphor highlights this connection. The path is not only the way of counting genealogies, it is also the path of the ancestor and his qualities from beyond to the present island. This ties together the place and the genealogy in a mutual legitimation. A genealogy can be used in one place, but not in another because both are actually expressions of the same process, the extension of the ancestral tumu in space and time. Correspondingly the seniority of one place can turn out to be juniority in the context of the polity of the whole archipelago. Precedence is determined in a recursive way, creating an overall social hierarchy.

The place-bound character of precedence is manifest also in the duality of power. Mana enua, power over land, can only be possessed by somebody who has a legitimate path to the original ancestor of the place. Conquest did not lead to the alienation of land in the Ngaputoru group. Even if it did, the conqueror had to create genealogical and even legitimate (i.e., through senior lines) connections to the ancestor of the place. Mana tangata, power over people, in turn, is able to emerge on the basis of marriage alliances abroad, and is thus able to transcend the horizon of a single island. The duality of power gives rise to a political situation of a relatively stable system of precedence in which a total reorganizing of the power structure in the Hawaiian style is impossible (cf. Valeri 1990).

The ‘Atiu chiefs, who held absolute political power in the Ngaputoru group, were tuakana, or elder brothers, but they only possessed mana tangata, power over people. This power did not enable them to deprive the people of other islands of their land. The Ma‘ukeans had mana enua on the basis of their path to the ancestor of their own land, and this path reversed the precedence in relation to the place. The overall authorities were said to be juniors, if they came to Ma‘uke, because the point of reference — the operator — changed.