Volume 42 1933 > Memoirs > No. 12 The migrations of a Pandanus people as traced from a preliminary study of food, food-traditions, and food-rituals in the Gilbert Islands. Part 2 by Arthur Grimble, p 51-84
                                                                                             Previous | Next   

- i
Memoir No. 12.
Supplement to the Journal of the Polynesian Society.
The Migrations of a Pandanus People As traced from a preliminary study of Food, Food-traditions, and Food-rituals in the Gilbert Islands

INSTALMENT No. 2.

Pages 51-84

- ii Page is blank

- 51
PART 2. TRADITIONS CONCERNING FOOD.

11. Foods of the Dead: (a) The well, the fish, and the tree; (b) The red food called te renga—12. Cannibalism and head-hunting: (a) Modern cases of cannibalism; (b) The Little Makin headhunting tradition; (c) The Tabiteuea Text; (d) The man-eating tropic-bird of Keaki; (e) The Matang tradition of Karumaetoa.

Appendices—1. Series of traditions from Little Makin, forming the prelude to the genealogy of the High Chiefs of the Northern Gilberts—2. Series of traditions from Tabiteuea, beginning with the first Ancestral Tree and leading up to the migrations of the children of Batuku the Skull—3. Text and interlinear translation of the tradition of the Keaki clan, concerning the immigration of the tropic-bird totem-groups from Samoa into the Northern Gilberts—4. Tradition of the Karumaetoa clan, concerning the immigration into Beru, Southern Gilberts, of a man-eating ancestor named Tewatu-of-Matang.

11. FOODs OF THE DEAD.

(a) The well, the fish, and the tree.

It is generally believed by the Gilbertese that when a departed ghost has safely passed the Bird-headed Woman 1 and the Old Man of the Cats-cradle 2 on his way to the - 52 western bourne of the dead, called Bouru, 3 he is caught in the netting strand of Nakaa, guardian of the entrance of that land, and entertained for three days upon the “food of spirits,” before being allowed to join the great company of his ancestors. During that period the ghost is fed—or, rather, feeds himself—upon the fruit of an inexhaustible tree, the fish of an inexhaustible lake, and the water of an inexhaustible well, owned by Nakaa.

The well has no name, being simply called Te Maniba, the term for the ordinary seepage-well of the Gilbert Group. Upon arrival the ghost is sent to draw water from the maniba, wherewith to lave his feet and slake his thirst.

The lake (nei) is sometimes called by the name Neineaba; it is believed to be of great expanse, and fringed with immense shoals not more than ankle deep. Neineaba is situated “in the middle of Bouru,” due north of the coastal place called Manra where Nakaa lives; it contains a single fish, the mon-n-taai (mon-of-the-sun), which is of a brilliant red-gold colour. The mon-n-taai, once caught in the net of a ghost, is immediately replaced in the lake by another. The belief is that, if the ghost can abstain for three days from eating either the flesh of this fish or the fruit of Nakaa's tree, and from drinking the water of the well, he will be free to return to his body in the land of the living; but being hungry and thirsty after his long journey to Bouru, he cannot resist the temptation of food and drink, and so forever binds himself to Nakaa.

- 53

The inexhaustible tree is called Tara-kai-maiu—an interesting name. Setting aside for a moment the first component, tara-, the meaning of the second and third is alternatively tree-vigorous, or tree-life. Either rendering would be compatible with the tree's character, the one referring to its unceasing fruitfulness, the other to its association with eternal life. A Baanaban myth of the origin of death not only settles the matter in favour of the second meaning, but also adds considerably to our special knowledge of Nakaa's tree.

It is related in the Baanaban myth that Nakaa lived formerly upon Boura with the first men and the first women, who as yet knew nothing of the sexual act. Nakaa said on a day to his people, “I am about to leave you for a while. When I am gone, you shall live separate—the men under this tree in the north, the women under that tree in the south. The men and women shall not play together when I am gone.” After he had departed on his voyage, a south wind carried the scent of tabaa (young pandanusbloom) from the women's tree to the men. So the men went over to pluck the bloom of the women's tree. They played with the women, and it was then that they first learned te buno (the sexual act). On Nakaa's return, his first work was to examine the hair of every man's head, and he found that grey hairs had come to all of them. From this he knew that they had disobeyed his word and played with the women. He turned to them all in anger, saying, “Now you shall leave this land, for you could not abide my word.” He pointed to the two trees, “Take your choice,” said he, “for one tree shall remain, and one shall go with you.” They chose the women's tree, and he said, “That is Tara-kai-mate, and its companion is death. If you had chosen Tara-kai-maiu you would never have known death. But Tara-kai-maiu shall remain with me, and you shall take death with you.” They took Tara-kai-mate and prepared to depart. While they were passing before him Nakaa plucked leaves from Tara-kai-mate, and in these he rolled a host of little insects. As the people were leaving, he pelted the backs of their heads with his leaf-bundles, saying, “These insects are grey hairs, and toothache, and - 54 stomach-ache, and all things that bring death; and as for the leaves, you shall use them as shrouds when you are dead.” Thus it is that, to this day, the dead are wrapped for burial in mats of pandanus-leaf.

It is clear from this myth that Tara-kai-maiu is not merely the tree-vigorous but, in a strong symbolic sense, the Tree-of-Life, and equally clear that, at some time in the history of the race, it was held to be a pandanus. This second fact accords very well with the first component of the tree's name, for tara- is evidently built up of te and ara, whereof the first is the definite article, and the second the generic term invariably prefixed to the name of any species of pandanus.4

Nevertheless, according to the popular Gilbertese belief of to-day, Tara-kai-maiu is not a pandanus at all, but a coconut-palm. This is not so surprising as it seems. As stated in the final paragraph of Part 1 (p. 50), all the vital tradition and ritual surrounding the pandanus has been for many generations the sacred (kamaraia) mystery of three social groups only, 5 and, even within those groups, the secret knowledge was confined to a very narrow circle of clan-elders, who made it their deliberate business to edit myth and tradition for public circulation in such a form as to blind all but the initiated to its inner meaning. It is entirely natural, in the circumstances, to find that popular belief as to the identity of Tara-kai-maiu is wide of the mark. The secret lore of the pandanus being regarded as kamaraia, 6 its transmission to any but a select few of the authorised social groups would have been held - 55 to endanger the life both of the giver and the receiver, and this is doubtless the basic reason for the deliberate corruption of tradition that I have indicated. The over-jealous hoarding of the authentic story, allied to the swift decay of custom during the last half-century—especially of religious custom—has almost secured its obliteration to-day; but there are still a few old men alive in the Northern Gilberts who know enough to deny that Tara-kai-maiu was a coconut-palm, and one or two bold enough to volunteer that it was a pandanus.

(b) The red food called TE RENGA.

According to the old man Taakeuta, of Marakei Island, the substance traditionally known as te renga was the food of ancestors (bakatibu) in “the line of lands in the west” called by the inclusive name of Te Bu-kiroro or Te Bongiroro.7 Quoting the same authority, who is backed by other old men of Marakei, Abiang and Tarawa—all in the Northern Gilberts—“te renga was a food which made the mouth red when it was eaten.” There is a tradition in Taakeuta's social group, 8 and also extant upon the island of Abaiang, that this substance was not taken alone, being chewed (kantaki) with the leaf of a certain tree.

Taam, of Marakei, who is descended through nine generations from a Beru ancestor named Kaabwibwi, remembers a story of his clan relating how Kaabwibwi used to “visit the west in dreams,” there to chew te renga in company with his ancestral deity Tabu-ariki. Kaabwibwi is believed to have gone, after death, to live in Bouru with all his ancestors and to feast with them upon the red food.

Supporting this individualized account is found a general tradition in the Northern Gilberts that te renga is the food of all departed ghosts, when they have accomplished their three days' sojourn with Nakaa, and joined the company of their ancestors in Bouru.

- 56

A belief which finds acceptance on most islands of the Group is that the red food is the diet of the fair or red-skinned ancestral deities of the race—Auriaria (the spirit of the pandanus), Nei Tevenei (his wife, the meteor), Nei Tituaabine (his sister-paramour), Riiki the eel (whose belly is the Milky Way), Tabu-ariki (the thunder-god), and Taburimai—in their western home called Matang. The red lightning that flashes in the storm-clouds of the westerly winds is sometimes called in old songs “the renga of Matang” and “the renga of Nei Tituaabine.” The redness of the sky at sunset is held to be a memorial of the food's colour, and of the western lands (sometimes Bouru, sometimes Matang) where it originated.

Clearly, all the above accounts of te renga have reference to a single family of ideas: in Taakeuta's story, it is the food of ancestors; in Taam's, the food of a specified human ascendant in company with his ancestral deity; in one generalized account, it is the diet of the great deities from whom descent is traced; and in the other, it is eaten by all departed ghosts when joined with the shades of their forefathers. The land of shades and the ancestral fatherland of at least one branch of the Gilbertese race are thus compactly identified, the one with the other, so that the red food of ghosts and gods may be regarded as an article once used by the human antecedents of the race in their western homes of Bouru and Matang.

Confirming the Marakei and Abaiang evidence that the food was not a simple but a composite substance is its name te renga, which means the mixture. 9 The invaluable details (a) that one of the elements of the mixture was the leaf of a tree, (b) that the whole was chewed, and (c) that it stained the mouth red, read together with the information that the food originated in the far West, enable the immediate identification of this substance with the betel-mixture (areca-nut, betel-leaf, lime), which is, of - 57 course, still commonly chewed in the far western Pacific and Indonesia. 10

The question which naturally arises, if the Gilbertese forefathers had the betel-chewing habit, is why their descendants have not persisted in the practice until to-day. This is fairly answered by the physical conditions of the Gilbert Group, whereof the almost purely coralline soil will support only two food-trees—the pandanus and the coconut-palm.11 If the areca-palm ever was introduced into these atolls, it could not well have outlasted the first generation of settlement. As the betel-chewing habit must thus have been involuntarily abandoned at an early epoch of the race-history, the memory preserved of the ancestral practice is remarkably precise.

Indonesia being the focus of the betel-chewing habit, it is natural to look first in that area for the far western lands named in the Gilbertese renga traditions; and, as the Moluccan portion of Indonesia stands at the gates of the Pacific, it seems prima facie more likely that the culture-stream which brought the Boura-Matang-renga beliefs to the Gilbert Group emanated from that area of the Asiatic archipelago. Such a supposition is encouraged by the conclusion of Rivers that the betel-culture was brought to the western Pacific by immigrants from Indonesia, and receives further support from Haddon's finding, on quite different evidence, that the Moluccas were the most probable starting-point of the various race-movements into the Pacific Ocean.12

Bouru is highly reminiscent of Buru, a large island in the centre of the Moluccan area, and a pair of topographical coincidences strengthens the suggestion that Buru may have been the ancestral land of the renga tradition. - 58 First, there is the story of Nakaa's great lake on Bouru, wherein the ghosts of the dead do their fishing.13 A glance at any good chart will show that the centre of Buru is occupied by a lake of quite exceptional size, no such considerable expanse of water occurring on any other island to westward of Celebes.14 And second, there is the tradition, also recorded in the previous section, that a place or area called Manra lies either on the south coast or else to southward of Bouru. A further reference to the chart will disclose the Banda Islands and Banda Sea immediately to southward of Buru.

Some further coincidences of nomenclature demand record in conjunction with these facts. Matang, the name of the other renga paradise already mentioned, is a widespread Indonesian place-name between Mattang of Sarawak and Medang of New Guinea; and Mwaiku, yet another Gilbertese paradise, recalls Waigiou by the Macassar Straits. Gilolo, facing Waigiou on the other side of the Straits, and immediately north of Buru, has already claimed notice in connection with the Kiroro cooking-oven, and Unauna in the northern bight of Celebes in connection with the Onouna of the katura oven. Between Unauna and Buru on the chart are seen the islands of Bangaai and Taliabu Bangai and Taribo are common place-names in the Gilbert Group. To the West of Buru lie Manipa and Serang (Ceram): there are many Manibas and several Terangs in the present home of the Gilbertese.

The cumulative value of these coincidences is enhanced by the diverse nature of the traditions which make them apparent. Two similarities of nomenclature have appeared in connection with cooking-ovens; four from an examination of Gilbertese place-names; one in a paradise story; and three in the paradise-renga traditions. It is certainly remarkable that whenever, in this diffuse material, the name of an original land is mentioned, it finds its counterpart in a single small area of Indonesia. The effect is that of - 59 a series of sign-posts set up at different points in Gilbertese culture and tradition, every one of them pointing to a common centre. Adding to this the commonly admitted likelihood that from this very centre—the Moluccan area—both the betel-people and other migrant swarms emerged into the Pacific, there seems to be very reasonable ground for the belief that Buru, the Banda Islands, Serang, Gilolo, and the places grouped around them were once the homes of those Gilbertese ancestors who chewed the red food called te renga.

Taakeuta of Marakei told me that the tree whose leaves were taken to chew with the renga mixture was Tara-kai-maiu, thus suggesting that the Tree of Life discussed in the previous section was neither a coconut (as believed by the masses) nor a pandanus (as held by the initiated), but a betel-palm.

Though Taakeuta's other evidence about te renga is backed by a good deal of outside testimony, he is the only witness known to me who associates the Tree of Life with the red food, and he is unable to state whence he obtained this exceptional information. On the other side, the identification of Tara-kai-maiu with the pandanus is based upon the direct testimony of the Karongoa clan—the preeminent authority—being also supported by the Baanaban myth and by the etymology of the tree's name. The likelihood therefore is that Taakeuta (who is not a Karongoa man) has not the authentic story. Nevertheless, I have found him, despite his great age, a reliable witness in many directions, and I cannot avoid feeling that his association of te renga with the Tree of Life amounts to something more than a mere confusion. Upon this matter further research may throw a light which I have been unable to obtain, perhaps necessitating a modification of my conclusion that a single tree only—the pandanus—is bound up in the tradition of Tara-kai-maiu.

- 60
12. CANNIBALISM AND HEAD-HUNTING.

(a) Modern cases of cannibalism.

There can be no doubt that sporadic cases of cannibalism have occurred throughout the Gilbert Islands until very recent times. A man was pointed out to me on Butaritari in 1922, whose father, just deceased at the age of about 80, was known to have strangled one of his wives a short while before the establishment of the British Protectorate (1892), and eaten raw her thumbs, great toes and breasts. It seems that he committed this atrocity whilst drunk with sour toddy, under the goad of sexual jealousy. His object was not to procure food, but to load the dead woman with the last imaginable indignity. He is reported to have said, while eating her flesh, “Ai beka-u mamma-m aei” (my excrement withal this thy breast).

Individual cases of cannibalism from two to five generations old collected from eight islands (including both northern and southern units, and also Baanaba) indicate that by far the most common motive of cannibalism, in later times, was that which appeared in the above example—the ultimate abasement of the dead.

A common practice during war-time in the Northern Gilberts was to pluck out the eyes of enemies slain in battle, and crush them between the teeth. The mere biting in two appears, as a rule, to have sufficed, but I have obtained from several old men of Tarawa and Marakei the admission that they actually swallowed the eyes thus enucleated. An idiom still in common use at moments of extreme anger is “I bia orai mata-m” (Would that I might eat uncooked thy eyes). The operation was usually performed in the heat of battle, standing over the newly-fallen enemy; but there is a tale of a certain High Chief of the Northern Gilberts, not very long dead, to the effect that he would occasionally cause his suspected rivals to be murdered in cold blood and brought to him, in order that he might bite their eyeballs with due deliberation.

An interesting story from Baanaba relates that, four or five generations ago, a Tabiteuean canoe containing five starving occupants drifted ashore there. The castaways - 61 were kindly treated, one of them, named Tebuke, being adopted into a household of the village of Buakonikai. After several years, Tebuke was suddenly missed from the village and, after vain search, was given up for dead. From that time onward many other people of the same village district began to disappear mysteriously, and it was believed that they had become victims of the same evil power that had spirited away Tebuke. After a good many years, Tebuke reappeared, sick and on the point of death. Just before dying, he confessed that he had lain hidden all the time in a hollow rock (now known as Tebuke's rock), which stood near one of the paths taken by fishermen to reach the eastern shore of the island. Whenever a man or a woman passed the rock alone, Tebuke had followed and killed the victim; he then dragged the corpse back to his hiding place, to eat it at his leisure. There seems to be no reason for doubting this story, which shows that, in some cases at least, there was a tendency to revert to cannibalism for purely gastronomic reasons.

The word “revert” is used advisedly, because tradition seems to leave no doubt that the eating of human flesh was commonly practiced, in conjunction with a form of organized head-hunting, by the race-ancestors who came to the Gilbert Group from Samoa, about 22-25 generations ago. This, however, is one of the most carefully-hidden secrets of the Karongoa clan; it was not until my ninth year among the Gilbertese that an authentic account of the facts was given to me.

[NOTE—A preliminary reading of all the traditions exhibited in Appendices 1-4 is advised before the perusal of the following pages is undertaken.]

(b) The Little Makin head-hunting tradition.

In 1923, three old men of the high chiefly group of Little Makin15 allowed me to take down at their dictation the text of which a translation appears in Appendix 1. Though some parts only of the narrative are pertinent to - 62 cannibalism and head-hunting, the text is given in full, as it contains much that will be of use hereafter.

Section 2 of the story opens with an account of the place in Samoa where (according to Section 4) human heads were laid in sacrifice, and of the spiritual powers to whom they were offered. The locality was Maunga-tabu (the sacred mountain) whereof the summit “smoked, and sometimes burned fiercely”; the deities were, first of all, Auriaria, who dwelt in the top of an ancestral tree upon the mountain-slope, and second, Batuku, a skull (which was also an ancestor) believed to have sprung from the smoking summit.

Section 3 describes the building of the canoe wherein the children of Batuku the Skull “fared forth to seek the food of their father from the west.” It is mentioned in this context how the canoe was launched over the bodies of dead men. Section 4 proceeds to give a clear account of how organized head-hunting raids were conducted against one island to westward of Samoa, called Butuna, and two to the south, called Tonga and Nuku-maroro. Butuna is clearly Futuna, or Horne Island, about 250 miles due west of Savaii; Tonga, correctly placed to southward, needs no explanation; and Nuku-maroro, given the alternative name of Nieue by the old men of Butaritari, is easily identified as Savage Island, a little to eastward of Tonga.

The victims selected by the raiders were “men who were the first-born, and bearded, and bald ”; their heads were cut off and hung in the rigging during the homeward voyage, while their trunks were heaped in the raiders' canoe. On arrival in Samoa, the heads were immediately taken as an offering to Batuku and Auriaria, being laid for this purpose on the lower slopes of the sacred mountain, “because the treading of that place was feared.” After this ritual had been observed, the bodies of the slain were divided as food among “the people of Samoa.”

The salient features of the Little Makin account are corroborated by a somewhat less detailed version collected from the Karongoa sib of Beru in the Southern Gilberts: discounting marvels, I see no reason for doubting the - 63 general accuracy of the facts related. Further supporting evidence is supplied by the traditions connected with the canoe-crest of Karongoa. This crest consists of various arrangements of tufts and pennants of pandanus-leaf, which I have described elsewhere.16 Almost any Karongoa man in the Group knows that the tufts are representations of human heads, “in memory of the food of the kings of Samoa in olden times.” The account given in the Little Makin texts of how the heads of the slain were hoisted in the rigging of the raiders' canoe interlocks very well with this widespread tradition.

As I have stated, the head-hunting and cannibalism of the Gilbertese ancestors in Samoa and elsewhere is the dark secret of an inner circle of Karongoa. Members of the outer circle, and of other social groups, possess versions of the Little Makin story, told, not in terms of fact, but in curious cryptic form, which they relate without in the least understanding their hidden significance, and which, set side by side with the authentic story, form a most interesting study. They are mythopoeic renderings of the truth which the initiates of Karongoa have put into currency, in order the more completely to conceal the real facts of history.17

It is related in the cryptic class of traditions that stranded porpoise formed the favourite food of the people of Samoa, and that the heads of the porpoise were the portion (tiba) of the kings of Karongoa. To a bitter quarrel arising out of the unfair division of certain porpoise is attributed the scattering of the people from Samoa, and their migration to the Gilbert Group. This, it is seen, agrees in general outline very well with the Little Makin account; only, porpoise-flesh replaces human flesh, and all details concerning the practice of head-hunting, the rituals - 64 surrounding it, and the deities with whom it was associated, are suppressed. In the expurgated versions, no mention is made of Auriaria or the skull named Batuku, these two personages being replaced by a king of Samoa called Namakaina (the moon); and though the ancestral tree figures in the story as the “abode of the kings of Samoa,” nothing is said of the “sacred mountain that smoked” whereon the authentic account places the tree.

Assembling the details with which the two classes of tradition, read side by side, furnish us, we have the following information:

  • 1. Cannibalism among the Gilbertese ancestors in Samoa was secondary to the offering of human heads in sacrifice to certain deities.
  • 2. A form of organized head-hunting was practised to supply the deities with their “food.” The heads of those who were the “first-born” and “bearded and bald” were preferred for ritual reasons.
  • 3. The spiritual powers to whom sacrifice was made were Auriaria, a god believed to dwell in the top of an ancestral tree, and Batuku, who was associated with an enormous ancestral skull. There seems to be a connection between these two beings and the moon.
  • 4. The ritual of sacrifice was connected with a sacred volcano called Maunga-tabu, the home of the tree-god and the skull-god.
  • 5. The victims of sacrifice were not, in later times at least, inhabitants of Samoa, having been fetched from the islands of Nieue, Tonga, and Futuna, all about 250 miles distant from Savaii.
  • 6. The euphemism—perhaps the ritual word—used to designate a corpse to be eaten by the people was te kua—a porpoise.
  • 7. The partition of dead bodies among the various social groups was a ceremonious occasion. It was some failure to observe the rights of a social group or groups in the course of such a ceremonial that caused the break-up of the race in Samoa.
- 65
  • 8. As the carriage of corpses by canoe from the neighbouring islands named to Samoa could not have occupied under two or three days, the flesh must have been putrid before arrival. This suggests that the form of cannibalism practiced was theoretical or ritual rather than actual or gastronomic.

The connection of Batuku the skull-god with the moon is arresting, because the association of cannibalism with beings who dwelt in the sky is of common occurrence in the Pacific.18 I refer especially to the famous tales of the heroic personage called Tawhaki by the Maori, wherein Tawhaki himself is seen to be descended from a cannibal grandmother, Whatitiri, who was not only a sky-dweller, but the variant of whose name, Whaitiri, means “thunder.” The astronomical associations of cannibalism are even better defined in some versions of the story quoted by Dixon;19 for these place Tawhaki eventually, with his grandmother, in heaven as a deity of lightning.

In the next few sections it will appear how closely the tree-god Auriaria, who shared with Batuku the sacrifice of human heads upon the sacred mountain of Samoa, was associated (in agricultural rituals) with the sun and moon; and later still he will be shown as the sun-god in very person. This being once apparent, the connection of human sacrifices with the moon by the Gilbertese ancestors in Samoa seems to be but one aspect of a sun-moon cult which embraced a wide range of religious activities. I thus anticipate my subject only to point out in passing that the astronomical associations of cannibalism throughout Polynesia—and especially in the Tawhaki traditions—may owe their origin to an ancient cult of the sun and moon, wherein the sacrifice of human heads and the subsequent - 66 eating of human flesh played a part. The likelihood of such a hypothesis will become more apparent when the deeper strata of Gilbertese myth and religion have been examined.

(c) The Tabiteuea text.

Much light is thrown upon the antecedents of the Auriaria-Batuku-Tree people of Samoa by a series of traditions from Tabiteuea (Southern Gilberts), which is presented at full length in Appendix 2, and will now be examined side by side with the Little Makin series of Appendix 1, which I have just reviewed.

Opening with a version of the Creation Myth, the Tabiteuea tale passes in its second, third, and fourth sections through a series of exploits of the well-known trickster type, wherein Na Areau the creator is the malicious hero;20 but the fourth section ends with a good deal of genealogical information, up to which climax the whole preceding series of narratives, in the manner common to such annals in the Gilbert Islands, is intended to lead.

The opening sentence of Section 1 of the tale relates how, before the creation era—which is to say, in the very distant past—two western lands named Aba-the-little (Abaiti) and Aba-the-great (Abatoa) were the home of the first ancestral tree (the ancestress sun), a pandanus, whereof the presiding spirit was none other than the deity - 67 whom we have already seen in possession of the tree of Samoa, Auriaria. There can hardly be in the history of the Gilbertese race two different and unrelated ancestral pandanus-trees, both owned by the single god Auriaria. The conclusion is that the first tree of Abaiti and Abatoa was the prototype of that later pandanus-tree called Kai-n-tikuaaba which (according to the Little Makin story)21 was planted on the sacred mountain of Samoa by Ariaria when, in the significant phrase of the text, “he trod the south.” Between Abatoa-Abaiti of the Tabiteuean tale and Samoa of the Little Makin account we are given, in fact, the first and the last milestones in the migration track of a tree-descended, head-hunting people, with their god Auriaria, out of a far western land or group of lands into the southern Pacific. Any doubt as to whether the two texts do indeed refer to the same people, god, and tree will disappear as the comparison proceeds.

On Abaiti and Abatoa, according to Section 1 of the Tabiteuea text, grew Na Areau the creator in the primeval darkness; then came the separation of heaven from earth; and finally “grew the lands.” 22 The land called “Kai-n-tikuaaba in the west” is stated to have grown first. Kai-n-tikuaaba, as noted in the preceding paragraph, is the name found attached in the Little Makin text to Auriaria's tree of Samoa: whether it be by origin the name of a land or a tree (and its meaning—tree of Tikuaaba—seems to include both connotations), its earliest associations were plainly with the west, whence it was later transferred to Samoa, doubtless when Auriaria “trod the south.” That Samoa was indeed the last of a series of lands settled by the tree-people is evident from the context under reference, for there it is carefully stated that, after Kai-n-tikuaaba in the west, grew “Tarawa in the east”; after Tarawa, Beru; then Tabiteuea; and last of all “Samoa in the south.” This is the key, land for land, of the order in which the tree-people occupied their successive homes on their way out of the west into nuclear Polynesia. The - 68 final stage of the migration, i.e., the movement from the southern Gilberts into Samoa, is described in Section 4 of the Tabiteuea story, which will now be analyzed.

The section opens by describing how Na Areau the creator, whilst at Tabiteuea, stole from a being named Taranga his wife Kobine, and made her the progenitress of ancestors. We have already seen this tale, under a rather different guise, in the Little Makin series (Appendix 1), for it is there related (Section 1) how Auriaria stole from Taranga in the underworld not a wife indeed, but a tree which nevertheless became an ancestress. The essential myth-fabric—the victim, his name Taranga, the theft, its result the birth of ancestors—is the same in both cases; only externals vary,23 and the story has been localized in the Tabiteuea account: that which remains common to both recensions represents what was evidently one of the fundamental beliefs entertained by the people of Auriaria concerning the first growth of their race.

The names of Na Areau's progeny by Taranga's wife on Tabiteuea are given as Au-te-rarangaki, Au-te-venevene, and Au-te-tabanou, which signify respectively: Au-the-continually-overturned, Au-the-continually-reclining, and Au-the-skull. These, according to the text, “were the first ancestors of Karongoa on Tabiteuea,” which is to say, their names stand as symbols representing a whole group (or perhaps three separate sub-groups) of Karongoa folk who had immigrated into that island. The same context links them together in a single religious category by stating that their anti was Auriaria. Their names obviously belong to the same family as that of Au-riaria, which signifies Au-continually-rising-over-the-horizon,24 and it seems pretty - 69 clear that in the god and his three eponyms (who are certainly not human ancestors) we have but four different personifications or attitudes of a single central identity named Au, who was the object of the Karongoa cult.

Having thus defined his groups by reference to their socio-religious indices, the native historian proceeds to describe the migration from Tabiteuea to Samoa, making Au-the-skull now the inclusive index of the movement: “(21) …. The day of voyaging came. Au-the-skull with his people voyaged to Samoa.” Three of the canoes which carried the migrants are named, and their names are found to have significant reference, in every case, to the heads of human beings offered in sacrifice.25 Thus obliquely does the narrator refer to the head-hunting habit which the Karongoa people carried with them to Samoa.

In the next paragraph (22), dealing with the settlement of the immigrants in Samoa, the Tabiteuea history links itself directly with the Little Makin tradition through the name of Batuku. The only differences are that, in this version, Batuku is presented not as a god or a skull, but as the king of the tree of Samoa and the progeny of Au-the-skull; while Koururu the brow appears as his brother, not his offspring as the Little Makin story [Appendix 1, Section 2 (9)] makes him. As far as the practice of human sacrifice is concerned, the text from Tabiteuea tersely confirms the more detailed Little Makin account by recording that the food of the kings of the tree was human heads.

Such, upon the evidence of the Tabiteuea tradition, is the tale of migrations implied in those few opening words of Section 2 in the Little Makin text: “Then was planted on Samoa the tree named Kai-n-tikuaaba, for there Auriaria planted it when he trod the south.” The last movement of all was the reflux of the head-hunting people from Samoa, back along their ancient migration track, into the Gilbert Islands once again. Section 6 of the Tabiteuea version deals with that event in semi-mythical language, stating - 70 that the progeny of Au-the-skull were flung by their anti Auriaria northward from Samoa to Tabiteuea Island. The Little Makin version interlocks perfectly with this account, in that it also brings a child of the skull, one Rairaueana, from Samoa to Tabiteuea; after which, in its closing paragraph (53) it shows how the line of Rairaueana migrated still farther northward, up to Butaritari, there to produce the ancestors of three high-chiefly dynasties in the Gilbert Group, and of another in Mille of the Marshalls.

Traced backwards into history, therefore, upon the evidence of the texts examined, the lineage of, say, the high-chiefly dynasty of Butaritari takes us first southward to Tabiteuea, and thence southward again into Samoa. Looping back northward, the line passes once more through Tabiteuea; thence, up to Tarawa; thence, westward to the land called Kai-n-tikuaaba; and finally, back to the earliest fatherlands called Abaiti and Abatoa. According to the Tabiteuea account, it was in those very ancient homes of the race that Auriaria, the god of the head-hunting rituals of Samoa, first dwelt with his tree, a pandanus, whose name was The Ancestress-Sun.

Another, and singularly valuable, link with the west is the description of the kings of the tree of Samoa, in paragraph 22 of the Tabiteuea text, as “the breed of Matang, the breed of red men.” The text is supported at this point by a mass of authoritative Karongoa tradition, as an example of which I select a second account of Batuku and his kin obtained from the island of Beru, Southern Gilberts. This account states that, after the work of creation had been completed, the being Na Areau-Tekikinto26 undertook a voyage from Tarawa, Northern Gilberts, to Samoa and, having arrived in that southern land, there begot “the eldest ancestor, whose name was Te-i-matang (The-man-of-Matang).” This is as much as to say that the migration of the Matang people into Samoa was by way of the Gilbert Islands, which constitutes valuable confirmation of our conclusions from the Tabiteuea text. - 71 Tei-i-matang, according to the Beru authority, was the ancestor of Batuku the Skull. To quote now from the Beru text:

“Batuku the Skull had a brother whose name was Kanii. Batuku and Kanii are said to have been kings beneath the tree of Samoa, and their food was the heads of the first-born, the eldest. The heads of the first-born children of the people of Nikumaroro were taken to be the food of those kings. And in the men Kanii and Batuku appeared the breed of Samoa, the breed of red men, who were called the people of Matang, the people of the tree Kai-n-tikuaaba: Tabu-ariki, Taburimai, Nei Tituaabine, Koura, Riiki, Nei Tevenei.”

Matang, as will be remembered from the preceding section, is in popular belief the far-western paradise where the fair-skinned ancestral deities—Auriaria, Nei Tituaabine, Tabu-ariki, and others—forever feast upon the red food called te renga. The inference has been drawn that this land was one of the early fatherlands of the Gilbertese ancestors. It is almost unnecessary to point out how greatly such an inference is strengthened by the independent evidence from Beru and Tabiteuea that the race of Auriaria was still called, long after it had migrated out of the west into Samoa, the breed or people of Matang. The concrete nature of such evidence clearly sets the land of Matang within the category of material realities.27

As a piece of cultural information, the direct connection of Batuku and his breed with Matang is of first-rate importance, for it brings their practice of head-hunting, allied with the cult of an ancestral skull, into immediate concatenation with the chewing of the red food called te renga. The close original association of the betel-chewing habit with the practice of head-hunting and a highly developed cult connected with the skulls of relatives has been demonstrated by Rivers.28 The identification of te renga with the betel-mixture thus rests not only upon the - 72 very strong internal evidence of the paradise stories, but also upon its perfect consistency with other characteristic features of the betel-culture—head-hunting in particular—now made apparent in the practices of the tree-people.

Rivers has stated that there is no evidence of headhunting in Polynesia as an organized and habitual practice, having the social or religious importance which attaches to the habit among the head-hunting peoples of Melanesia.29 Chiefly upon this ground, the same authority doubts whether the betel-culture ever penetrated into Polynesia. The evidence of Gilbertese tradition just examined is therefore of a somewhat sensational nature, and if, as it seems to show, a numerous head-hunting folk with memories of the betel-chewing habit did indeed penetrate, by way of the Gilbert Islands, as far as Samoa, it will be necessary to explain why the vestigia of the betel-culture now traceable in Polynesia are—if any at all—so slight as to be almost unrecognizable. A sufficient explanation will, I think, disclose itself when certain other aspects of the culture of the tree-people shall have been examined.

(d) The man-eating tropic-bird of Keaki.

The remainder of this section will be devoted to a review of some further traditions of cannibalism in the Gilbert Group, obtained from social groups other than the Karongoa clan. The first of these, whereof the vernacular text and an interlinear translation appear in Appendix 3, emanates from the clan of Keaki, which claims the tropic-bird as one of its totems 30 and Nei Tituaabine as its ancestral deity. The free translation of the story here follows:

- 73
THE KEAKI TRADITION.
  • 1. After the breaking of the tree (of Samoa) Kai-n-tikuaaba by Te Uribaba,31 all the beings who lived in its top were scattered. The birds of Nei Tituaabine, the red-tailed and the yellow-billed tropic-birds, flew away. The yellow-billed tropic-bird flew westward, and settled upon the land of Beberiki; the red-tailed tropic-bird flew eastward to eat the redness of the sunrise, and after that, it came down-wind to the tip of Little Makin, where it settled upon the branch of the pandanus-tree called Te Ani-koura, or Te Aara-maunga-tabu, or Tara-kai-mate, above the bathing-pool called The-laughter-of-waves.32
  • 2. And behold! if any man (of Little Makin) went to bathe in that pool, the red-tailed tropic-bird leapt upon him and ate him.
  • 3. And behold! Nei Tituaabine arose (from Samoa) to follow and seek her bird: she came from the south, passing up to windward of the islands, and carrying with her one withered coconut and her divination set called te kirikiri.33 She arrived at Little Makin and went to the maneaba (meeting-house). When she met the people in the maneaba,
- 74
  • she said, “Have you seen my tropic-bird here?” (They answered) “We have seen it, and now there are hardly any people of our land alive, for the bird has well-nigh eaten them up. Canst thou, then, not save us from that bird of thine, for thou art indeed our mother also?” That woman said, “I can: plait two fans, and with them you shall (be able to) kill it.”
  • 4. So when those fans were ready, she told the women Bairuti and Batikoran to go and fan the bird. They did as they were told: the bird died, and they returned to tell Nei Tituaabine. Then she went to bury her bird: over its head she planted her withered coconut, and around that plant she set an enclosure of three hard stones. When she had finished burying her bird, she left it and returned to the maneaba. Then all the people of the land were happy, and gathered together to play and dance; every day, indeed, they made merry.
  • 5. And when it was again evening on a certain day, they were dancing together, and behold! there appeared a red glow within the maneaba, on its eastern side. The people of the maneaba looked toward the place where the red glow oppeared, and behold! they saw a shining red-skinned man as big as a giant. But when the dancing chant was nearly ended, that man ran away.
  • 6. And when they were next dancing, that man once more appeared: and when the chant ended, he was chased; and behold! he was found in the top of the pandanus-tree upon which the tropic-bird had formerly settled. A great company also dwelt with him in the top of the pandanus-tree. And Nei Tituaabine asked him, “Whence grew ye?” And that man, who was the eldest of them, answered, “I grew out of the pandanus-tree, where the tropic-bird settled upon it.” And thus said Nei Tituaabine, “Thy name is Koura.”34 Then she pointed at the rest of his company in turn saying, “Thou art Iti-ni-koura; thou art Rube-ni-koura; thou art Koura-toa; thou art Koura-iti;
- 75
  • thou art Koura-ma-to-taake; thou art Koura-n-tamoa; and thou Koura-n-Tarawa.”
  • 7. The whole of this company was red-skinned. And after a little while, they were all taken to the maneaba, and later still they were made uea (high chiefs) therein.
  • 8. After a time, they ordered that a canoe should be built for them. When it was ready, it was called “Tebuki-ni-benebene” (The-tip-of-a-coconut-leaf); and then the time came for them to travel, so that they might see the neighbouring islands. And Nei Tituaabine told them, saying, “You shall go first to look at the coconut-tree which I planted over the tropic-bird.” They went to look at it, and there were people in its top. These people were led back to Nei Tituaabine, and when thy came to her she said to the eldest of them, “Thy name is Nei Riki; and thou art Nei Temareve; and thou art Nei Tebaarae; and thou, Nei Tarabainang; and thou, Nei Newi.” And then she gave her divination-set of pebbles to Nei Newi, with a mat of invisibility as its covering; and then (again) she tore off the top of the coconut-tree from which they had all grown, and gave the leaves to Nei Tarabainang as a divination-set.

After this point, the narrative describes the voyages of the tropic-bird people, under the leadership of Koura, down the Gilbert Group, and their colonization of the four islands of Butaritari, Abaiang, Tarawa, and Beru. This carries the tradition beyond the scope of the present subject, but it is worth while to point out in passing the evident criss-cross of immigrant currents that was set up in the Gilbert Islands by the return of the Samoan branch of the race to Micronesia. While the Karongoa clans of the two texts first examined are seen to have entered the group at Tabiteuea in the south, and to have proceeded thereafter up to the extreme northerly islands of Little Makin and Butaritari, the tropic-bird groups of the tradition now presented took the diametrically opposite course of invading the group at Little Makin, and working their way thence down to Beru, an island as far to the southward as - 76 Tabiteuea. This single example must suffice at present to illustrate the restless and complex swirl of clan-movements that vexed the group during the period immediately succeeding the incursion from Samoa.

Regarding now the technique of the Keaki tradition, we have in this narrative a good example of the method common to many clan-histories in the Gilbert Islands. The tale is fundamentally a record of facts, the central event being the immigration of a certain man-eating group from Samoa into Little Makin; but, instead of naming the actual ancestors who took part in the invasion, the historian uses the clan-deity and totem-creature—Nei Tituaabine with her tropic-bird—as social indices, and attributes to them the historic acts of the whole Keaki group of immigrants for which they stand.35 He overlays the whole with myth-material relating to the totem and the deity, which obviously dates from an era much earlier than that of the return from Samoa.

Setting aside the myth-fabric, and rationalizing the account of facts, the tradition may be read as follows: When the ancestors of the Keaki clan were obliged to leave Samoa, they fled northward until they came to Little Makin. There they landed, having secured their first foothold in the neighbourhood of the bathing-pool called Tengare-n-nao. From that centre, they proceeded to attack the local population, their victims being killed and eaten. The practice of cannibalism, however, ceased for a very definite reason, which is made apparent in the prayer to Nei Tituaabine put into the mouths of the victimized people (paragraph 3 of text): “Canst thou, then, not save us from that bird of thine, for thou art indeed our mother ALSO?” This is an excellent example of the characteristically oblique manner in which the Gilbertese historian imparts - 77 his most vital information. The passage means that Nei Tituaabine was the ancestral deity not only of the invaders, but of the invaded: in other words, the immigrants from Samoa were of the same ancestral stock as the people whom they found established upon Little Makin. This is, of course, valuable support to the conclusion dictated by the Tabiteuea text (Appendix 2), that the immigration from Samoa into the Gilbert Group was nothing more that the return of a race—or part of a race—along an ancient migration-route to one of its earlier homes. The intent of the historian, in the passage quoted above, is to explain that the incoming tropic-bird folk, though first obliged to fight their own ancestral kin for a foothold upon Little Makin, nevertheless ceased to practice cannibalism upon them because they shared with them the cult of a single clan-deity, Nei Tituaabine.

Nevertheless, though cannibalism ceased, it is clear that the immigrants established themselves as conquerors of the land, for the evidence of paragraphs 6 and 7 of the text is that the Koura people, who “grew out of the pandanus-tree” of the tropic-bird from Samoa, were made high chiefs in the maneaba of Little Makin. This naturally raises the question why the people of the Keaki clan are not, to this day, high chiefs of the island. The answer is implied in the final paragraph of the Little Makin text (Appendix 1) which we have examined. While the tropic-bird group was invading the extreme northerly Gilbert Islands, that branch of the Karongoa group which was led by Rairaueana the Man-of-Matang, child of Batuku the skull, was immigrating into Tabiteuea. The genealogical details supplied in the closing sentence of the text show that, in a later generation, the Karongoa clan moved northward, in the person of Rairaueana's descendant Teietoa, to Butaritari. By that time, the tropic-bird folk must have been well established as overlords of Little Makin, and probably also of Butaritari, but such was the sacred prestige of Karongoa among the Samoan immigrants that it is very doubtful whether the tropic-bird folk withstood—or even desired for a moment to withstand—the prerogative of Rairaueana's group to supersede them. - 78 However this may have been, it is certain from the records of the high chiefly dynasty now established upon Butaritari and Little Makin that, but a few generations after Rairaueana the Man-of-Matang first invaded Tabiteuea, his three famous descendants, Rairaueana the warrior, Na Atanga, and Mangkia—whose names appear in the final paragraph of the Little Makin text—represented the only ruling caste upon the two northern islands.36 The reign of the tropic-bird immigrants was thus of short duration, but a great many of their descendants still form part of the local population.37

A somatological point of great interest stressed by the Keaki historian in paragraph 7 of the text is that “the whole of the company of Koura”—which is to say, all the tropic-bird invaders from Samoa—were red-skinned. This is evidently the Keaki rendering of that Karongoa tradition already examined, concerning “the breed of Samoa, the breed of red men, who were called the people of Matang.” That the tropic-bird folk were indeed also “people of Matang” is clear from the fact that their ancestral deity was Nei Tituaabine, this goddess being numbered, as the sister-paramour of our tree-god Auriaria, among those fair-skinned beings believed to feast upon the red food called te renga in Matang-of-the-West. But perhaps the clearest evidence of the racial identity of the tropic-bird clans with the breed of Matang38 is contained in the intimate relationship of their ancestral beings, Koura and Nei Tituaabine, with that very index of the Karongoa-Matang culture, the ancestral pandanus of Auriaria called Kai-n-tikuaaba, the tree of the sacred mountain. Out of - 79 that tree, according to the Little Makin text in Appendix 1 (paragraph 4), sprang not only the Karongoa god Tabuariki, but also every other great clan-deity of the head-hunting people, including Nei Tituaabine, who grew from one of the branches, and Koura, child of the first bloom. The inference is that all the social groups who believed themselves descended (through their gods) from the tree were, equally with Karongoa, of the breed of Matang. The following very explicit passage from the Beru text already quoted confirms the conclusion: “And in the men Kanii and Batuku appeared the breed of Samoa, the breed of red men, who were called the people of Matang, the people of the tree Kai-n-tikuaaba: Tabu-ariki, Taburimai, Nei Tituaabine, Koura, Riiki, Nei Tevenei.39

The tree of Auriaria may thus be regarded henceforth as the index, not only of Karongoa, but of a whole congeries of red (i.e., tawny)-skinned clans who practised head-hunting and cannibalism, and brought with them into the Pacific memories of betel-chewing out of a western land called Matang.

But the full lore of the tree—its myth, its history, its head-hunting rituals—was peculiar to Karongoa. Only such fragments of that lore as Karongoa passed for circulation were permitted to subsist in the traditions of other social groups, and these were so cryptically presented that, in the course of time, their meaning was lost to the uninitiate clans who purveyed them. Examples of such morsels of occulted truth are to be seen in the Keaki text. In paragraph 1, the man-eating tropic-bird of Nei Tituaabine is pictured as having settled at Little Makin upon the branch of a pandanus-tree, whereof one of the three recorded names is Ara-maunga-tabu (Pandanus-of-the-sacred-mountain). In paragraph 6, the ancestral being Koura is shown to have grown from this same tree.40 These allusions carry a merely local significance for the Keaki - 80 historian, but it is clear to any one acquainted with the Karongoa tradition in Appendix 1 that, far from being local, they are disjunct fragments of the authentic tree-myth. The pandanus of the tropic-bird on Little Makin is but a projected shadow of the ancestral pandanus of Auriaria, the parent of the Keaki deities, the mother-tree of a cannibal race, where of Keaki was a member, and whose kings once received sacrifice of human heads on the slopes of a sacred mountain.

(e) The Matang tradition of Karumaetoa.

A tradition of the social group called Karumaetoa41 concerning an ancestor named Tewatu or Towatu links itself usefully with the Keaki text, and emphasizes the association of cannibalism with the land of Matang. A free translation of the Karumaetoa narrative appears in Appendix 4, and will now be very shortly analyzed.

The first five paragraphs of this story form an effective digest of the Keaki tradition concerning the invasion of Little Makin by the tropic-bird people, and confirm the account of their man-eating habits (duly inhibited by Nei Tituaabine) which we have already examined. The text indeed adds somewhat to our knowledge of the invaders' movements before their onset upon the Northern Gilberts, by describing in figurative language (paragraph 1) what seems to have been their unsuccessful attempt to establish themselves first upon the island of Beru. But as far as the mythical content of the narrative is concerned, it is interesting to note the Karumaetoa historian's complete silence as to the association of Koura and Nei Tituaabine with a pandanus tree. The reason is, that the origin of the Keaki ancestral deities is no concern of Karumaetoa's: this clan is therefore left in ignorance of even such garbled - 81 fragments of the pandanus-myth as serve, in the Keaki text which we have seen, to connect the tropic-bird people with the tree of the sacred mountain.

From paragraph 6 onward, the narrative is purely Karumaetoa history. The tropic-bird invasion, leading to the establishment of a Keaki dynasty upon Little Makin, caused the flight of one of the vanquished autochthones, named Tewatu, to Tabiteuea. But, according to paragraph 7, this personage did not dwell long on Tabiteuea, “for there was war in that land”—which probably means that he was driven out. He with all his people then fled overseas until “after a long time, they came to the land of Matang.” Matang is described as a great land in the west, where dwelt Tewatu's anti, or ancestral spirit, Tabuariki. In other words, Tewatu fled to Matang because he was descended from ancestors to whom that land had once been home, which is to say, he was of the breed of Matang.

The same conclusion as to this personage's race may be reached from another angle, for we recognize his deity Tabu-ariki as one of the principal clan-gods sprung from the ancestral tree of the Samoan invaders who, as we have seen, called themselves the breed of Matang. There can thus remain little doubt as to the validity of our previous finding upon the Keaki text (paragraph 3) that the immigrants from Samoa were of the same ancestral stock as the people—including, as we now know, Tewatu—whom they found established upon Little Makin; and this again lends support to the evidence of the Tabiteuea text, first, that the Gilbert Group was colonized by the tree-folk en route from a western fatherland into southern Polynesia, and second, that the invasion from Samoa was nothing but the reflux along its old migration-track of that part of the race which had passed into the South Pacific.

It was Tewatu's descendant and namesake, Tewatu-of-Matang, who (paragraphs 8 and 9 of text) came back from Matang to Beru two generations after the flight from Little Makin. We cannot take the text too literally as to the number of generations passed in Matang, but the return to the Gilbert Group is dated for us with some accuracy by the reference made in paragraph 11 to - 82 Taane-n-toa the second, a famous high chief of Beru, with whom Tewatu had dealings. According to the best Beru pedigrees in male lines, Taane-n-toa flourished 19 generation ago, so that probably not less than 400 and not more than 500 years have elapsed since the immigration described in the Karumaetoa history.

Paragraph 9 makes a particular point of the transport by Tewatu of his parents' skulls from Matang to Beru. In another version of this story which I possess, it is declared that the hero used the two skulls as his drinking vessels, but paragraph 13 of the text exhibited indicates that, on his arrival at Beru, he buried them near the boua42 (stone pillar) at which he practised the cult of his ancestral deity. As we have seen, the cult of an ancestral deity together with an ancestral skull was the basis upon which rested the head-hunting and cannibal rituals of the tree-folk in Samoa, who called themselves the breed of Matang. The texts which have made this fact apparent, however, belong to the secret lore of Karongoa, and are confined to a narrow circle; futhermore, though they give rise to the plain inference, they fall short of furnishing the direct statement, that the cult of god and skull emanated from Matang or, indeed, from any land in the west; thus, the concrete evidence of the Karumaetoa text that Tewatu came direct from Matang with his cult of one of the best-known tree-gods linked with that of his parents' skulls, plus the habit of cannibalism, is of high value, inasmuch as it is of a domestic rather than a sacerdotal kind, and concatenates in a single stroke all the main features of the Matang-religion with which this section is chiefly concerned.

Paragraph 10 gives a terse but unequivocal description of Tewatu's cannibal activities on Beru, and paragraph 11 relates how, “after many men were eaten at Teteirio.” he was called to parley with the high chief, Taane-n-toa. The honour with which he was treated by the chief—including the ascription to him of a sitting-room in the maneaba at Tabiang—plainly indicates that he was a person to be placated rather than opposed: that is to say, the - 83 force of the immigrant host from Matang was a strong one. What, then, prevented it from continuing to pursue its cannibal habits on the island?

It was seen in paragraph 3 of the Keaki text that the tropic-bird invaders of Little Makin ceased to prey upon the local population because both immigrants and autochthones practised the cult of the same ancestral deity, Nei Tituaabine. Paragraph 13 of the Karumaetoa story now provides us with an exactly parallel situation, the historian using the same oblique or allusive method of conveying his facts as that observed in the former instance. Tewatu-of-Matang is pictured to us as taking the sitting-room in the maneaba just accorded to him, when the high chief asks, “Who is the ancestral god of the stone where those two skulls of thine are buried?” In other words, “What is thy ancestry?”—the first of all questions asked by any Gilbertese native to this day of the stranger who assumes a sitting-place in his maneaba. Tewatu answers, “Tabuariki is my ancestral god.” “Ours also!” replies Taane-n-toa, 43 and, this point having been made clear, the conclusion follows naturally, “It is enough! Thou shalt not after this eat the people of Beru.” The breed of Matang, composed of social groups all claiming descent from the ancestral tree of Auriaria, did not prey upon itself. This accords perfectly with the evidence of the Little Makin text (Appendix 1), which shows the children of Batuku-the-skull organizing their head-hunting raids, not against the people of the tree in Samoa, but against the inhabitants of Nikumaroro (Nieue), Futuna, and Tonga, about 250 miles overseas. The skulls of kinsmen evidently played an important part in the rituals of the head-hunters, but it was the heads of strangers that formed the sacrifice, and the flesh of aliens that made the food of the tree-folk in the cannibal meal that followed.

- 84

Such, presumably, were the reasons which led to the extinction of head-hunting and cannibalism on a national scale in the Gilbert Group. The raiding of island by island was inhibited, because all units—including Baanaba—were populated by tree-folk, and the gods of the tree demanded only alien heads in sacrifice. Isolated in mid-Pacific, far from any foreign population, the race was starved of material for its rituals, and since the eating of human flesh was but the by-product of those rituals, that too fell into abeyance. The biting in two of enucleated eyeballs in time of war, described in the opening part of this section, may perhaps be taken as a vestige of the ancient headhunting practice, while the sporadic acts of cannibalism recorded might be regarded as rare individual reversions to a habit engendered by the old tree-god cult. That the eating of human flesh was regarded as anti-social within a few generations of the immigration from Samoa is clear from the local history of that personage named Mangkia, who is named in the final paragraph of the Little Makin text (Appendix 1). Mangkia was a member of the high chiefly group on Butaritari, directly descended from Rairaueana, “the child of Batuku-the-skull, ” who invaded Tabiteuea from Samoa. According to a well-authenticated tradition of his island, Mangkia was expelled from Butaritari because he developed, and too often satisfied, a taste for human flesh, at which his brothers were ashamed and his fellow-islanders indignant. He was nevertheless followed into exile by a good many retainers, which enabled him first to win a foothold upon Abemama, in the Central Gilberts, and finally to establish there a dynasty of high chiefs, of whom the infamous Tern Binoka (all too romantically depicted by R. L. Stevenson) was a descendant.

1   Bird-headed Woman: Nei Karamakuna, who pecks out the ghost's human eyes and gives him spirit's eyes in return, provided that he can give her the only food she desires, namely, the tattoo-marks on his arms or body. If the ghost lack these he must pass blind into the Land of Shades.
2   Old Man of the Cats-cradle: Noubwebwe, who displays before the ghost the series of string-figures collectively called by his name. The ghost is bound to name correctly that figure of the series, which is also called Noubwebwe, whenever it appears. If he fail, he will be either strangled in the string or else impaled by the old man's staff, and die forever. The being Noubwebwe is represented as a stunted, black, curly-haired person; he appears in a servile capacity in some versions of the Creation Myth, and is believed by some to have invented the string-figure and displayed it for the first time while heaven was being separated from the earth
3   Western home of the dead … Bouru: For a general description of the path of a Gilbertese ghost to the Land of Shades, see “From Birth to Death in the Gilbert Islands,” J.R.A.I., January-June, 1921. Much additional information concerning the rituals of the dead, their burial, and despatch to Paradise, has been collected since 1921, and will be published in due course. In connection with footnote 45 above, it may be stated that certain string-figures used to be made over the dead during the course of the ritual for “straightening the path of the ghost” to its western home. The religious significance of the cats-cradle is thus very definite in the Gilbert Islands. Vide also the ritual connected with the sun-figure, “Gilbertese Astronomy and Astronomical Observances,” J.P.S., December, 1931, pp. 213-215.
4   E.g.: te ara-bouru (the pandanus of Bouru), te ara-matang (the pandaus of Matang), te ara-maunga-tabu (the pandanus of the sacred mountain), are some of the hundred and seventy-odd pandanus names used by the Gilbertese. Ara changes euphonically to an- and ani- before n and k respectively—e.g., to an-nabanaba (the pandanus of Nabanaba), and te ani-koura (the pandanus of Koura).
5   The three social groups are named Karongoa, Ababou, and Maerua. See 2 and 9 in the table of totems exhibited in section 5, Part 1, p. 20.
6   See footnote 43, section 10, Part 1, p. 49, for the meaning of kamaraia.
7   Bu-kiroro, Bongiroro: These names have already appeared in connection with a certain form of cooking-oven. See antepenultimate paragraph of section 2 (a), Part 1, p. 10.
8   The eel-totem group of Nukumauea.
9   The word has, for obvious reasons, acquired a second meaning, i.e., red dye. The verb rengana signifies to mix.
10   As to the distribution of betel in Oceania, see Rivers, History of Melanesian Society, vol. 2, pp. 249-251 and passim.
11   The soil of Ocean Island, consisting mostly of phosphate of lime in an insoluble form, supports only the wild almond in addition to the pandanus and coconut.
12   Haddon, “The Outriggers of Indonesian Canoes, ” J.R.A.I., January-June, 1920, p. 71.
13   See preceding section.
14   See for preference Admiralty Chart No. 942a, Eastern Archipelago, eastern portion.
15   The high chiefly group of Little Makin is the local equivalent of the Karongoa sibs on other islands.
16   “Canoe Crests of the Gilbert Islands,” Man, June, 1921.
17   I was present when a Karongoa elder, to whom the real facts were perfectly well known, purveyed the cryptic form of the story to a large audience with great gravity and conviction. When I talked with him a few hours afterward, he explained, “These things are shameful, and they are also kamaraia for both reasons, they may not be squandered (bakataeaki) to the mass of the people.”
18   Rotuma—Romilly, Letters from the Western Pacific and Mashonaland, London, 1891; Mangaia—Gill, Myths and Songs from the South Pacific; Admiralty Is.—Meier, “Mythen and Sagen der Admiralitätsinsulaner,” Anthropos, 2, 646-67, 933-42; New Zealand—White, Ancient History of the Maori, 1; Paumotu Is.—Leverd, “The Paumotuan Version of Tafa'i,” J.P.S., 1911, pp. 172-78.
19   Dixon, Oceanic Mythology, p. 59.
20   The name of the creator, Na Areau, means spider. For variant versions of the Gilbertese Creation Myth, see Folklore, 1922, pp. 91-112. As a rule, the creator does not appear as the trickster, this role being filled by his son Na Areau the younger, who is portrayed sometimes as a spider, but more often as a little malicious black man, with close curly hair and a flat nose. This personage is often called in the Northern Gilberts Na Areau-Tekikitea, and in the south Na-Areau-Tekitekite or Tekikinto. He is always found in conflict with the large-bodied, tawny-skinned ancestral deities Tabuariki, Auriaria, and others, from whose righteous anger his cleverness enables him invariably to escape. The trickster tales whereof he is the hero seem to represent an intermediate stage between the purely animal trickster tales of Indonesia and the stories of impishness grouped around the name of Polynesian Maui-Tikitiki. Observe how closely Na Areau's second name, Tekitekite, assimilates to Maui's second name, Tikitiki.
21   Little Makin Text, Appendix 1, Section 2 (6).
22   Section 1 (5).
23   It will be seen later that the divergence of externals is more apparent than real, inasmuch as Taranga's tree, stolen in the underworld by Auriaria, was a pandanus, and, as such, essentially a woman (see also Part 1, p. 16, footnote 20, for evidence of this). The tree of the Little Makin account thus equates perfectly with the wife of the Tabiteuea tale stolen by the god.
24   Ria is used of ships, luminaries, or lands appearing over the skyline. The duplicated form riaria is frequentative in force, denoting habitual or recurrent action.
25   See footnote 92, Appendix 2, for a discussion of the canoe-names in question.
26   Na Areau-Tekikinto was the progeny of Na Areau the elder, the creator. See footnote 63 ante.
27   Matang is now a common place-name up and down the Gilbert Group.
28   History of Melanesian Society, vol. 2, pp. 260-61 and passim.
29   Op. cit., p. 261.
30   For other totems of the Keaki clan, see the table appended to Section 5, Part 1, p. 20.
31   Te Uribaba: c.f. the Little Makin text, Appendix 1, Section 4, where Te Uribaba is named as the traitor who betrayed the people of the tree to the inhabitants of Tonga, Futuna and Nieue. It is probable that this name (which signifies “the tap-root of a tree”) stands rather for a social group of the tree-people than for an individual. The phrase “breaking of the tree” used in the present context obviously refers to the catastrophe which caused the break-up of the race in Samoa, and is a good example of the cryptic idiom originally imposed by the Karongoa clan upon other social groups, such as Keaki.
32   Te Ani-koura—The Pandanus-of-Koura. It will be seen a little later that Koura was an ancestral being of the Keaki clan. Te ara-maunga-tabu—The-pandanus-of-the-sacred-mountain; Tara-kai-mate—The pandanus-tree-of-death. See discussion of the nature of the tree of life in Section 11 (a) ante. The present context seems to put it beyond doubt that the trees of life and death were regarded as pandanus trees.
33   For a short description of the use of te kirikiri (pebbles) for divination, see “Canoes in the Gilbert Islands,” J.R.A.I., Jan.-June, 1924, pp. 102-3.
34   Ura, in modern Gilbertese, signifies red or burning. Cp. Polynesian kura.
35   It is still a common Gilbertese practice to designate a whole group of people by the name of their totem or clan-deity: e.g., E roko Taburimai i aba-ra (Taburimai arrives at our island) means, “Some people of the clan whose deity is Taburimai have arrived.” E noraki te Taake i Tarawa meang (The tropic-bird is seen at north Tarawa) signifies, “There are people of the tropic-bird clan living on north Tarawa.”
36   Rairaueana the warrior migrated to Mille in the Marshall Islands, and there established a chiefly group which (according to a Butaritari claim which I have not verified from the Mille end) is still extant. The third brother, Mangkia, migrated to Abemama, where his descendants are still high chiefs. The second, Na Atanga, stayed at Butaritari and became the ancestor both of the local high chiefly dynasty now in power and of that established upon the island of Abiang, about sixty miles south of Butaritari.
37   See footnote 40, Part 1, p. 48.
38   See Section 11 (b) ante,
39   See page 71 ante.
40   Allied to the tradition of Koura's growth from a pandanus-tree is the widespread Keaki belief that any pandanus-drupe represents the body of this ancestral being, whose name, signifying red or burning, has reference to the orange glow of the ripe fruit. The incident of Koura's appearance as a red glow in the maneaba, related in paragraph 5 of the Keaki text, is clearly related to this belief.
41   The totem of Karumaetoa is the shark, the ancestral deity Tabu-ariki. See table annexed to Section 5, Part 1, item 13. Karongoa also numbers the shark amongst its totems, and the being Tabu-ariki amongst its ancestral deities.
42   A description of one of these stone pillars will appear in Part 3.
43   The high chief Taane-n-toa of Beru was demonstrably of the breed of Matang, being descended from a line of Tarawa high chiefs belonging to the clan of Karongoa. There will be some discussion of the Tarawa line in Part 3.