Arts and Crafts of the Cook Islands
A study of the cultural processes which have functioned in Polynesia leads me to agree with Piddington that too much importance has been attributed to diffusion from without and too little attached to local development within Polynesia. In anthropology, as in other sciences, there is a strong tendency to accept without question the opinion of writers who were regarded as authorities at the period when they promulgated their views. Rivers' theory of two distinct cultures being responsible for the variations in Polynesian culture was accepted as a fundamental fact; and subsequent writers, not deeming it necessary to check the accuracy of the theory, have devoted their efforts to arranging the data in their particular lines of study to conform with the two strata theory. The cultural process employed to support the theory was of necessity diffusion from outside of Polynesia. The importance attributed to diffusion reached its peak in the theory formulated by Elliot-Smith thatall early civilizations emanated from Egypt and that the heliolithic culture that developed in that country spread across the Pacific to America, where it provided the foundation of the Maya and other allied civilizations that developed on the American continent. The cause for the spread of the heliolithic culture has been page 502given as the search for gold and pearls to form health-giving amulets, but no explanation has been given as to what human agency carried the culture across the Pacific. The acceptance of this theory involves the belief that an ancient civilization was implanted on the islands along the sea path to America before the Polynesians entered the Pacific. However, the theory that heliolithic culture was the basis of the Mayan, Aztec, and Inca civilizations has been refuted by American anthropologists, who adhere to the belief that these civilizations were developed in America without outside influence.
The supposition of an extinct civilization in Polynesia has formed an intriguing theme for literary writers with little or no scientific inhibitions. The stone images of Easter Island and the relics of previous human occupation on deserted atolls such as Malden Island have been attributed to previous large populations which have been drowned when the larger land areas conveniently provided for them were as conveniently submerged beneath the Pacific. I have already maintained that the manufacture and transport of the Easter Island images from the factory to their present sites were well within the powers of the ancestors of the present Polynesian inhabitants (15, pp. 232, 233). The Easter Islanders found a suitable material in the easily worked volcanic tuff in the extinct crater of Rano-raraku and developed their own technique and form of image, just as their Polynesian cousins in the Marquesas and the Austral Islands developed local forms of large stone images; all three areas producing a conventional human form with different anatomical details. I have also criticized MacMillan Brown's assumption that Malden Island was a sacred island to which a large population came to worship from a nearby fertile archipelago now sunk below the sea with its entire population (15, pp. 143-144). MacMillan Brown's opinions were based on a description of the archaeological remains which were seen by Bloxam in 1825. The description included an illustration of a temple drawn by Dampier to resemble a truncated pyramid. MacMillan Brown saw in the drawing an affinity With the pyramids of the sun and moon on the coasts of Peru, the teocallis of Mexico, and the Metalanim structures of Ponape. Such an affinity, of course, indicated the trail of an ancient civilization across the Pacific to America. He held that an army of men was required to build the temple on Maiden and that "paved roads" which were present were used by the people who came from the nearby archipelago to worship at the temple. An accurate survey of the remains, made by K. P. Emory in 1925, showed that the temple was not a truncated pyramid and that it was no different in type or in the size of the limestone slabs than similar structures in Tongareva which were built by the ancestors of the present inhabitants. The "paved roads" dwindled down to narrow tracks formed of single flat slabs of coral limestone spaced a little apart to form pathways to the sea. Similar tracks are still used by fishermen on the not far distant atoll of Tongareva to avoid the sharp coral points strewn about on the ground. Thus an affinity with Ponape and America never existed and the need page 503for an army of builders with the conjuring up of their home on a submerged archipelago is dispelled by more accurate field observation. The relics ascribed to an extinct ancient civilization resume their proper place as the signs of past temporary occupation by Polynesians with an atoll culture which did not differ materially from the Tongarevan culture which was functioning on first contact with Europeans in 1853.
Another example of wishful misinterpretation is the statement from another source that the slabs erected on some of the open temples were made of concrete and that, as the Polynesians did not know how to make concrete, the slabs must have been made by some previous more advanced civilization. However, the alleged concrete slabs were formed of coral limestone quarried from the natural coral limestone strata that exists in a natural state on the seaward and lagoon sides of all coral islets and on the seaward side of volcanic islands which are bounded by coral reefs. An example of trying to create a mystery out of what should have been obvious is provided by Lamont's description of the small islet of Te Kasi in the Tongareva atoll. The islet is circular in shape, and from the central hollow a number of stone paths radiate to the shore. Lamont concluded that the islet must have been used for some peculiar ceremonies of an unknown nature. Had he known the native language, he would have learned, as I learned from a native informant while sitting in the selfsame hollow of Te Kasi, that the hollow formed a sheltered resting place for fishermen and that the radiating paths led to whatever side of the islet the wind or tide rendered favorable for fishing (15, pp. 132, 133).
An instance of cumulative error arose from Lamont's description of an open temple in Tongareva (74, pp. 180, 181) which states, "It was encircled by tall, flat stones, some six feet in height… a sort of 'Stonehenge' in a small way." Westropp, from Lamont's description, described it as "a stone circle in one of the Penrhyn Islands like the sepulchral circle of Stonehenge and the stone circles of Khassia." Perry, from Westropp's reference, described it as "a megalithic stone circle" that indicated that sunworship was practiced on Tongareva by an archaic civilization. Thus Lamont's description of an open space "encircled" by six-foot stones resembling Stonehenge "in a small way" became a stone circle like Stonehenge and finally a megalithic stone circle used for sunworship by an archaic civilization. I made a field survey of the Tongarevan temples in 1929 and found that all the temples, including the one described by Lamont, were rectangular in ground plan. They all had vertical slabs of coral limestone erected at intervals along the four sides of the temple courts and they conformed to an atoll type of Polynesian temple. As late as 1853, Lamont saw a religious ceremony being conducted on one of the temples in which offerings of coconut husk were made to the local gods among whom the sun was not included. Thus, another piece of evidence in support of an archaic civilization was based on the ambiguous and erroneous use of the word "encircled."page 504
As shown by the examples cited, the theory of a pre-Polynesian civilization has been based upon unverified statements, misinterpretations, and exaggerations to make certain things appear as if they were beyond the powers of the Polynesians to accomplish. The most astounding flight of imagination is that wherein, without geological confirmation, large areas of inhabited land have been submerged to bolster up a theory by destroying the evidence. A certain amount of emergence and submergence has perhaps occurred and may account for certain problems in the distribution of plants, insects, and landshells; but, if so, they occurred in the dim past before the coming of man. The volcanic islands of the western Pacific were peopled at an early period and the Melanesians spread as far as Fiji, but there is no scientific evidence that the islands of the central and eastern Pacific were inhabited before the arrival of the Polynesians.
Having cleared the way, the two strata theory may now be considered. If, as the names imply, the so called archaic, pre-Tangaloan, Indo-Tangaloa, or palae-Polynesian culture formed the earlier stratum, it may be identified with the original culture brought in by the first Polynesian settlers. The date when the Polynesians entered Polynesia is, at best, uncertain. Attempts have been made to fix the date by genealogies. However, genealogies transmitted by memory alone are subject to the vicissitudes of their human carriers and to the influences that effected change in other elements of human culture. They provide a sequence of historical events associated with some of the ancestors on the particular lineage, but beyond a period of five or six centuries, they are unreliable as a means of fixing dates. In spite of their unreliability, it has been passively accepted from the writings of S. Percy Smith and others that the Polynesians began to move out into the Pacific at about the beginning of the Christian era and had reached Samoa by the end of the fifth century. Samoa is specifically mentioned because Percy Smith and others held that the Polynesians entered Polynesia by way of a southern route through Melanesia. However, I believe that they entered Polynesia by the northern route through Micronesia, from the eastern end of which they fanned out southward to Samoa and Fiji and eastward to the Society Islands. As a working hypothesis, Percy Smith's date of the fifth century may be accepted tentatively. It may have been some centuries earlier, but is not likely to have been later.
The former existence of a second distinct culture in Polynesia is based on the assumption that some of the elements in Polynesian culture are so highly developed or so different that they must have been introduced by another group of people. If the assumption is correct, when did they enter Polynesia and from where did they come?
If a second culture did enter Polynesia, it must have been after the fifth century and before the spread of the later Polynesian culture (p. 475). This later spread commenced approximately in the twelfth or thirteenth century, to page 505 Hawaii and the Cook Islands; and it was as late as the fourteenth century to New Zealand. Thus there occurred a period of about 700 years in which a later intrusion could have taken place, but such a period also allowed for considerable local development. Genealogies, legends, and historical traditions supply information concerning movements within Polynesia, but they are strangely vague about such an important event as a second influx from without.
Information as to where a second culture could have come from is equally vague. It would have had to come fully developed from Indonesia or to have been developed on the way to Polynesia. In either case, it would have had to enter Polynesia by way of Melanesia or Micronesia. The objections to the Melanesian route apply equally well to the second culture as to the first. If it came by Melanesia, one would expect some Melanesian elements in the later culture that spread from central Polynesia. That the Polynesians were capable of accepting some Melanesian elements is indicated by the culture of Samoa and Tonga. In religion, Samoa and Tonga accepted the Fijian concept of Bulotu (Pulotu) as the haven of departed souls. In social organization, they included brother-and-sister avoidance and the power of the sister's son over the property of his maternal uncle and his family. In material culture, they adopted the Fijian type of double canoe, some types of weapons, neck rests, and rectangular plaiting. None of these Melanesian elements are present in the culture that spread from the Society Islands. It is evident from their distribution that they diffused from Fiji to Samoa and Tonga because of shorter interisland distances, frequent communication, and intermarriage, but they spread no farther east. This diffusion in the west was probably comparatively late, after communication between Samoa and the Society Islands had ceased.
There are two elements in the local culture of the Society Islands which are regarded by some writers as due to diffusion from Melanesia. These are the 'Arioi society and the mourner's mask. Rivers held that not only was there affinity between the 'Arioi society and the Melanesian secret societies, but that the secret societies were created in Melanesia by the Polynesians on their way through and that the 'Arioi society was a survival in Polynesia. The fundamental principles of the Melanesian secret societies are that membership was confined to males, that special lodges were absolutely restricted to its male members, and that a secret ritual was conducted at the lodge. In the 'Arioi society, membership was open to both sexes, the buildings erected for the society, in which to give performances in plays, singing, and dancing, were open to the general public of both sexes, and there was no secret ritual. The 'Arioi society was not a secret society but a highly organized union of strolling players, and the Melanesian secret societies have a greater affinity in fundamental principles with the secret societies of civilized Europe and America than with the 'Arioi society of the Society Islands. Furthermore, in the Marquesas, the adolescents who sang and danced at public entertainments were page 506termed ka'ioi, and the karioi houses of Mangareva and the kariei houses of Rarotonga and Aitutaki were built for entertainment in singing and dancing by adolescents of both sexes. Thus not only does diffusion lack support, but the various stages which indicate how the 'Arioi society reached its peak by the process of internal development are found in Polynesia.
The mourner's headdress of the Society Islands has large pearl shells fixed to conceal the face, and it has been regarded as a mask of Melanesian origin. The only other masks known in Polynesia are the tapa masks of Mangaia (pl. 2, B) and the curious gourd masks drawn by Webber during Cook's third voyage off one of the Hawaiian islands. I regard the Mangaian masks as of recent local origin, perhaps even post-European. The Hawaiian masks were evidently a late sporadic invention, for they are not known outside of Webber's drawing. In the Tahitian mourner's headdress there is no attempt to delineate the human face, as there is in the Melanesian masks. Moreover the mourner's mask was part of an elaborate mourner's dress which had a wooden shoulder piece ornamented with pearl shells, an apron of narrow pieces of perforated pearl shell strung together with fine cords, a poncho jacket of tapa, a skirt of yellow tapa ornamented with circular discs of coconut shell, and a shoulder cloak of feathers. None of these individual parts of the dress have any affinity in form or technique with anything in Melanesia. I feel sure that the mourner's costume, including the mask, was developed in the Society Islands long after the later Polynesian culture spread to the other parts of Polynesia. The only support in favor of diffusion from Melanesia appears to be the fact that masks were common in Melanesia and rare in Polynesia.
The objection to the Micronesian route is the difficulty of a highly organized culture maintaining its high standard after infiltrating through the chain of atolls at the eastern end of Micronesia. The loss of basaltic stone, fibrous plants, paper mulberry, cultivable food plants, and domestic animals would have formed as serious an obstacle to a second, later culture as it did to the first. The possibility that the coiled coir cap and the Y-shaped canoe stanchion diffused from the Gilbert Islands has been mentioned, but they were elements of an atoll culture that reached the Cook Islands after the spread of the later Polynesian culture. Though social and religious complexes may pass through an atoll chain, the highly developed techniques that characterize the variations in Polynesian arts and crafts could not have been introduced by a later diffusion through Micronesia.
Having dealt with the possibility of diffusion of a second culture by the two available routes, I find no tangible evidence in favor of it. It must be remembered that the second culture is purely a hypothesis created to account for the cultural variations of the different Polynesian groups of islands. The more developed cultural complexes that were assigned to the later intrusive culture were found in the various islands by the first European visitors, at the page 507end of the eighteenth and beginning of the nineteenth centuries. Thus, a period of at least 600 years had passed since the later Polynesian pattern was carried from the Society Islands to the various Polynesian groups. The later Polynesian pattern was also the last that could be strictly termed Polynesian, for, though it formed the basic foundation of the culture in other parts, each island group, including the Society Islands, developed and elaborated along directions whither their local environment and particular genius guided them. On the advent of Europeans, there was no standard Polynesian culture which applied to all branches of the Polynesian people any more than there is a standard English culture which applies to all English-speaking people whose ancestors came from England.
Internal development with local differences has been described for the individual Cook Islands, and what applies to the islands of a group applies with still greater force to groups of islands more widely separated. The part played by diffusion within Polynesia has been shown to be small and usually limited to groups which are fairly close; and the most marked variations have been due to local internal development. These developments apply to all branches of culture.
In social organization we see such marked variations as the customs surrounding the sacred person of the Tui-tonga of Tonga, the carpenters' trade unions of Samoa, the ritual and human sacrifice attending the installation of a king in the Society Islands, the bird cult of Easter Island, brother and sister marriage in Hawaii, and the development of military genius in New Zealand. Those peaks of development were not due to diffusion from abroad but may be traced locally to developments and efflorescences having their roots in some earlier elements of Polynesian culture.
Polynesian religion is based on the deification of ancestors with the development of family gods. The family gods rose in power with the increased status of their followers from families to tribes and from tribes to tribal confederations. Offerings of food and material objects also rose with the economic condition of their worshipers from a piece of coconut husk, a pebble, a fish, or a piece of turtle on an atoll to large quantities of food, whole pigs, and even human sacrifice on the richer volcanic islands. The simpler' less developed form of religion was retained in Samoa and Tonga; but a more complex system grouping certain gods into a divine family exercising influence over the various activities of human endeavor was developed by the priests of Opoa in Raiatea. This complex, forming part of the latter Polynesian culture, was carried with it to the north, east, and south by the emigrants who settled in those regions. The major gods, Tane, Tu, Rongo, and Tangaroa varied in the seniority given them in different island groups, and both the Tahitian and New Zealand priests have been credited with creating supreme gods which were surrounded with an aura of esoteric lore. Some authors have regarded the page 508concept of a supreme creator as a survival from an ancient monotheistic religion, but I agree with Piddington (82, p. 263) who "suggests that the esoteric knowledge of Tahitian and Maori priests was a specialized elaboration, systematization and collation of a body of mythology which was originally less full, less organized, and less unified."
It is sometimes dangerous to use social and religious complexes for comparative purposes, because some of the elements which compose them may have been added by native informants after the culture underwent change in post-European times. This is particularly true of religious complexes. Again, students of another race may attach undue importance to the elements they select for their particular project. Too much depends upon how a subject has been recorded and upon individual enthusiasms and idiosyncracies in interpreting behavior and emotion. In material culture, however, the artifacts themselves speak even though the nimble fingers that made them have long been still. But the artifacts must be gathered together from all parts of the world before they can tell the consecutive tale of their origin and development. Let us consider the story of the beautiful feather cloaks of Hawaii which reached a peak of perfection in Polynesia if not in the whole Pacific area. A rough rain cape in the Bishop Museum at Honolulu tells how a rectangular piece of fish net had split ti leaves tied to it by their stalks in overlapping rows to provide a portable thatch as protection against the rain. A rectangular cape in the Vienna Museum tells how the coarse feathers of the domestic fowl and seabirds were substituted for ti leaves to form a decorative garment but lacking the long stalks of the ti leaves they were tied on with thread by overhand knots, a separate thread to each feather. Another rectangular cape in Florence shows a transition from a coarse net to a fine net, from coarse feathers to fine red feathers, and from single knottings to a continuous thread. Other capes in many museums show the progress in the change of shape from rectangular to circular by cutting, shaping, and joining pieces of netting with a mesh so fine as almost to resemble woven cloth, and with red, yellow, black, and green feathers arranged in various motifs to form chiefly regalia. Finally, the cloaks which reach from neck to heel complete the tale of evolution in which a technique that commenced with a fisherman's rain cape reached its culmination in a royal robe for a king. This record of local development has been preserved by the garments themselves, for the simpler forms retained their use. Though their manufacture decreased, they were not entirely abandoned, because the higher forms were beyond the means of the lesser chiefs. All the factors for internal development were present in Hawaii and there is no evidence of diffusion from elsewhere.
The flaxen garments from New Zealand (69) tell a similar tale of local development from rough rain capes to dress cloaks of great variety. The capes and cloaks were a substitution for tapa cloth because, though the paper mul-page 509berry had been carried to New Zealand, the tapa cloth which had proved suitable in tropic climes was inadequate for the cold and wet of the new home. The search for a substitute raw material resulted in the selection of the fiber of the local flax and the technique adopted to form a textile was the single-pair twine, a technique so widely used in making fish traps that it must have formed part of the later Polynesian culture which came with the settlers from the Society Islands. The inclusion of rough flaxen tags on the flax fiber warps by means of the single-pair twined weft provided both textile and thatch for the necessary rain cape. Development took place in the lengthening of the rain cape into a rain cloak. The attachment of colored tags and feathers instead of rain tags resulted in decorative cloaks. The single-pair twine was improved upon by the use of a two-pair interlocking weft, and invention proceeded a step farther in decoration by the creation of colored borders worked in geometrical patterns with a wrapped-twine technique. The variety of garments retained specific uses, and thus the simpler techniques remained in use. As in Hawaii, the various types of capes and cloaks indicate clearly through their technical details the sequence of local development.
Each island group developed special skill in certain techniques and the attainment of high standards resulted in the many variations in the arts and crafts of the Polynesian area. The Samoans specialized in the plaiting of fine mats in check with a lower border trimming of red feathers which were worn as skirts to mark social distinction. The Tongans were expert in making baskets in four different techniques: plaiting, coiling, netting, and twining. The twined baskets were unique in the use of geometrical patterns with coir fiber in the natural color and dyed black, and they were also ornamented with white and black beads of native manufacture. The Cook Islanders excelled in plaiting sleeping mats with colored borders worked in geometrical patterns by a twilled technique. The Society Islanders created a tall cylindrical headdress, closed at the top, by using vines with a single-pair twine and attaching a large curved frontispiece covered with tapa to form a foundation for feather decoration. The Marquesans excelled in making coir head bands ornamented with pearl shell, turtle shell, feathers, and tufts of hair from old men's beards. The Easter Islanders were skillful in making circular feather headdresses in which long feathers were attached to foundation coils in such a manner that they projected horizontally instead of vertically.
The objects enumerated illustrate the diversity in specialization that occurred in the Polynesian area and even in those of a similar class, there is no affinity in the techniques selected to produce somewhat similar effects. There is no affinity in the techniques adopted in Hawaii and New Zealand to produce feather cloaks, and the feather cloak that formed part of the Tahitian mourner's costume is totally different in technique from the other two. Even in the plaited mats of Samoa and the Cook Islands, there is no affinity in the tech-page 510niques used to produce the decorative results. Diversity in form and technique is even more marked in the headdresses. If in addition to the headdresses of the Society Islands, the Marquesas, and Easter Island, we consider also the feather helmets of Hawaii, the decorated coiled caps of the Cook Islands, and the bleached hair headdresses of Samoa, we find that in six of the most important Polynesian groups there is nothing in common in the techniques used to produce a headdress. There is a slight affinity in the feather fixation of the Marquesas and Easter Island, but, with this minor exception, each island group differs entirely from each of the other five groups. Practically the only thing in common among the six groups is the idea of producing a form of head decoration to distinguish social rank or military eminence; and this idea is so universal that it does not need diffusion to account for it in Polynesia.
Diversity is also illustrated by the weapons of Polynesia. Though the spears and slings show little variation from some early pattern, the clubs present a problem. Some of the Tongan and Samoan clubs are similar, and the Society and Austral Islands have a practically identical lozenge-bladed club with a shoulder ornament. Such similarities are undoubtedly due to late diffusion between those neighboring pairs of islands. If we group these two pairs, we may say that the clubs of Samoa-Tonga, Society-Austral Islands, Hawaii, the Marquesas, Easter Island, Cook Islands, and New Zealand are entirely different, each group from all the others. If a second, more developed, culture entered Polynesia, it is difficult to understand why it did not contribute some established type of club shared by at least some of the groups affected by the composite culture which was supposed to have resulted from the later diffusion. The great diversity and variation in the arts and crafts, instead of being explained by a later diffusion from without, appear to form strong evidence against it.
A reason advanced for the presence of certain highly developed complexes in some islands and their absence in others is that they are survivals of a higher culture which has undergone degeneration. In religion, the sporadic occurrence of a supreme creator has been regarded as evidence that a monotheistic religion formed the high level of a theology once held by the Polynesians. In social matters, reference has been made to Rivers' theory that the 'Arioi society was a survival of secret societies created by the Polynesians. The assumption of such survivals implies that peaks of development indicate the high-water mark once attained and that the cultural tide has receded to a lower level. However, there is no evidence of any such recession, and the so-called survivals may be regarded rightly as special developments that have risen independently above the general culture level.
Handy, who had the advantage of making a field study of the history and culture of the Society Islands (40), evidently accepts the two strata theory for he divides Society Islands culture into two strata which he terms Old page 511Tahitian and Arii. The Old Tahitian stratum he associates with the first inhabitants of the island of Tahiti, the Manahune, who were regarded as commoners or plebeians. The Arii stratum he associates with the Hui Arii or chiefly families who claimed descent from Taaroa and who developed their polity and religion in the district of Opoa on the island of Raiatea. It is at once evident that this division may indicate the social cleavage between the cultures of the commoners and the aristocracy.
According to two independent sources (40, p. 6), chiefly courts (marae arii) which derived their origin from Opoa, were established in Tahiti and Maupiti 62 generations ago. Handy, on a basis of 20 years to a generation, gives the courts an approximate antiquity of 1,240 years which indicates the seventh century as the era of the spread of the dynasty from Opoa. Allowing not more than 100 years (5 generations) for the colony at Opoa to become firmly established, Handy supposes that the Hui Arii came over the horizon and descended upon Raiatea first in the late sixth or early seventh century A.D. Such a date is as uncertain as the date of the fifth century attributed to the general arrival of the Polynesians on the western fringe of Polynesia. With such uncertainty, the difference of 100 years does not carry much weight in constituting a gap of any significance between the two events. It is quite possible, therefore, that the ancestors of the Hui Arii took part in the general entry into Polynesia and did not come as a distinct movement at a later date.
A difficulty in the use of genealogies for dating historical events, apart from their accuracy in the first place, is the uncertainty as to the number of years to be given to a generation. Percy Smith and others have allowed 25 years in New Zealand genealogies; Handy has set 20 years for Tahitian genealogies; but one of his chief informants, the late Queen Marau Taaroa of Tahiti, held that 15 years was more likely to be accurate owing to the early marriages in chiefly families. By Marau's measure, the 62 generations quoted by Handy would diminish to 930 years and thus advance the era of the spread of the Opoa dynasty to the tenth century instead of the seventh. I believe that the later date would coincide more nearly with the historical events that led to the conquest of Tahiti by the Hui Arii. Thus Marau's measure suits one end of a chain of historical events and Handy's measure suits the other end. As they cannot both be right, the doubt created serves to illustrate the uncertainty attending the use of genealogies.
Handy (40, p. 8) describes the method adopted by him in distinguishing between the two cultures as follows:
Through analysis and comparison, it is possible to arrive at an approximate definition of the culture of these early Tahitians. Certain traits characteristic of the mass of the people are like those typical of the Maoris in New Zealand. Since the Maoris are unquestionably "old Polynesians" I infer that these traits are "Old Tahitian," and attribute them to the ancestors of the manahune. On the other hand, other elements of culture, which contrast with the above, are clearly attributable to descendants of Taaroa whom traditions reveal as invaders.
The statement that the Maoris are unquestionably "old Polynesians" with a culture similar to that of the ancestors of the manahune is open to question. It would have been applicable to the tangata whenua first settlers of New Zealand but it must be remembered that the later Polynesian culture developed in the Society Islands was carried to New Zealand in the fourteenth century. The Maoris, therefore, cannot be regarded as "old Polynesians" any more than the Cook Islanders and the Hawaiians who left the Society Islands at an earlier date. It is true, however, that elements belonging to the early stage of Polynesian culture have been retained in New Zealand, but so they have in all parts of Polynesia including the Society Islands.
The other elements of culture which contrast with the simpler elements regarded as old Tahitian are attributed by Handy to the Hui Arii or descendants of Taaroa "whom traditions reveal as invaders." In view of the second stratum in the two stratum theory being regarded as an intrusive culture from outside of Polynesia, Handy's use of the term invaders might be regarded as supporting evidence by believers in the theory. However, it must be clearly understood that the traditions reveal not an invasion from without Polynesia but the invasion of Tahiti from Raiatea, an interisland invasion within the same group. Such invasions from within have occurred in the history of practically all the main island groups of Polynesia. In further distinguishing the culture elements attributable to the Hui Arii, Handy (40, p. 19) states that he has been helped by a comparison with Samoan civilization which he regards as having preserved the Tangaloa (Taaroa) culture in its greatest purity. My own opinion is that the cultures of Samoa and Tahiti, by whatever names they may be designated, have less in common than the culture of Tahiti with the cultures of the other main groups of Polynesia. Upon analysis, the affinities between Samoa and Tahiti are more theoretical than real. Handy (40, p. 67) lists in two columns the elements attributed to old Tahitian and Arii. Some of the material elements will now be discussed.
The houses of the old Tahitians are given as rectangular and those of the Arii as having rounded or apsidal ends appended to a rectangular frame. These apsidal houses are peculiar to the Society Islands, Samoa, and Tonga and form what Handy considers an affinity between the descendants of Tangaloa in Samoa and the descendants of Taaroa in Tahiti. However, an analysis of the detailed structure of the apsidal houses in the two areas shows clearly that they are constructed by means of totally different forms of architecture. The Samoan apse (73, pp. 19-36) is curved horizontally and vertically by means of a series of graduated arches passing obliquely from side to side, one set with its ends resting on the curb plate and another, upper, set on the end rafters of the middle rectangular frame. The Tahitian apse (41, pp. 12-16) is formed by a series of straight rafters that radiate from the ridgepole of the rectangular frame to the curb plate and is thus curved only in the horizontal page 513direction. The highly developed locking joints of the short curved pieces that form the Samoan arches is unknown to the Arii culture. Furthermore, the roof of the middle rectangular section in the Samoan house is curved convexly outward from the ridgepole to the wall plate by using a series of graduated collar beams to strut opposite main purlins to the required distances apart, and the flexible rafters conform to the curve so provided. The middle section of the Arii house has thick straight rafters and the straight roof differs in no respect from the roof of rectangular houses. Thus, not only is there a difference of structure in the apsidal ends, but there is also a difference in the rectangular frame to which the apses are attached. If technique is any criterion, there is absolutely no affinity in structure between the architecture of the two areas. The only thing in common is a general idea of a rounded end and if the idea of roundness was diffused from Samoa, only one half of a two dimensional idea reached the Society Islands, where it was produced with such a different technique as to constitute a new invention. On the evidence, the two types of apsidal houses may be regarded as distinct local creations which developed independently of each other. As the Arii type of apsidal house is confined to the Society Islands, it may be regarded as a late development which occurred after the later Polynesian culture had spread from that area. It must also be remembered that the apsidal houses were made as residences for chiefs and for use as houses of assembly and entertainment, whereas the commoners ruled over by the same chiefs dwelt in rectangular houses. The apsidal houses and the rectangular houses may thus denote class distinction rather than distinct cultures, much as differences occur in the same civilized culture between the mansions of the rich and the humble dwellings of the poor.
House furniture such as stools, headrests, and bowls are attributed by Handy to the Arii culture, and he draws particular attention to their characteristic feature, that of being supported on legs. The four-legged seats were made in one piece (their distribution has already been described, p. 421). Attention has been drawn to the fact that they do not occur in Samoa and Tonga. The distribution in the Society, Cook, Austral, and western Tuamotu indicates that the seats were probably developed in the Society Islands and diffused to the neighboring islands after the spread of the later Polynesian culture.
Wooden headrests occur also in Samoa and Tonga. The Samoan headrest (73, pp. 76-78) is a length of bamboo to which two pairs of legs formed of forked branches of light wood are lashed. The Tongan headrests consist of a variety of forms, but, though a few are made of one piece of wood, most of them are formed of a round bar or a narrow, flat piece of wood to which paired legs are attached at each end. A great variety of headrests are found in Fiji where the narrow bar was specifically designed to support the neck and so prevent the elaborate Fijian hairdress from being disarranged by contact with page 514the floor while resting prone. The Tongan forms resemble the Fijian types so closely that diffusion must have taken place between the two groups and to a lesser extent between Samoa and Tonga. In all three groups, the articles were used as neck rests, and, as neither the Tongans nor the Samoans had an elaborate hair dress to protect, the credit of origin must be given to Fiji. Furthermore, the Fijian name of kali for the headrest was used in both Tonga (kali) and Samoa ('ali). The Society Islands headrest was made in one piece with a wider horizontal part for use as an actual pillow to support the head, beneath which a piece of folded tapa was often placed for greater comfort. The western headrests made with separate legs and used as neck rests were derived by diffusion from Fiji, whereas the Society Islands headrest made in one piece and used as a pillow for the head was derived locally, in all probability from the wooden seat which is absent in Samoa and Tonga. Though wooden seats were used sometimes in the Cook Islands as pillows, the smaller, specialized headrest was confined to the Society Islands. It was thus developed after the wooden seats had diffused to the nearer islands, which confirms the theory of its development from the wooden seats.
Bowls with legs are characteristic of the Society and Cook Islands, whereas the food bowls of Samoa and Tonga are without legs. The Samoan food bowls are characterized by an elliptical rim opening with horizontal projections at each end which are used as handles. The Society and Cook Islands bowls are ovoid or beaker shaped with an upward projecting lug at the broad end of the rim. Though the kava bowls of Samoa and Tonga have legs, they were probably a late development to distinguish them from food bowls when ceremonial kava drinking increased in social importance. There is also a strong possibility that both the round shape and the legs were derived by diffusion from the Fijian kava bowls. The western type of kava bowl is not present in the Society Islands. The differences in detail of the food bowls of the Society Islands and Samoa are sufficient to show that they are due to different cultural developments.
In articles of dress, a cultural distinction is made by Handy between the maro loin cloth that passed between the legs and the pareu skirt. It must be remembered, however, that women always wore the short skirt and never the loin cloth. The cultural distinction made is therefore in the use of the two forms of garment by men. The loin cloth is undoubtedly old, but it formed an element in the later Polynesian culture. In Hawaii, it continued in use as the male garb of chief and commoner alike. In New Zealand, influenced perhaps by the abandonment of bark cloth with its resultant change in clothing technique, the usual raiment for the lower part of the body was a kilt or skirt. In the Cook Islands, the loin cloth was worn by fishermen and agriculturists at work but in the evening after bathing, the pareu was worn. The loin cloth was thus a working garb and the skirt, a garment of relaxation. It is natural that the chiefs, who did no manual labor, should have worn the skirt as their page 515usual dress. In Samoa, on the other hand, the malo is referred to as a narrow girdle worn during war, but the common article of dress worn by chief and commoners alike during the day was a kilt (titi) made of ti leaves. In the evening after bathing, the ti leaf kilt was laid aside for a tapa skirt. Plaited kilts and skirts were also made for wear on festive occasions. In the Society Islands, the use of the loin cloth and the skirt was much as it was in the Cook Islands, but the secondary chiefs (ra'atira) wore plaited skirts (vane) made of hibiscus bast. Handy (40, p. 21) considers that the presence of the skirt form of dress in Samoa, Tonga, and the Society Islands may be safely attributed to the Taaroan invaders. However, the poncho (tiputa) type of garment, the presence of which in the Society Islands is also attributed by Handy to the Taaroa people, is absent in Samoa and Tonga. The fact that the poncho was developed after the spread of the later Polynesian culture is evidenced by its limited diffusion to the neighboring Cook and Austral Islands, hence it cannot be regarded as an element of an old Taaroa culture shared by the Samoans. If the plaited skirt was associated with the plaited poncho, it must also be regarded as a development in the Society Islands which was independent of a similar development in Samoa.
A cultural distinction is made in boats, a va'a canoe with a dugout hull, bow and stern pieces, gunwales, and outrigger float with two booms being the older Tahitian form and the built up pahi with a keel being attributed to the Arii. Though the difference in construction is marked, it must be taken into consideration that both types of vessels existed in the one culture at the same time for different purposes. The dugout canoe was a fisherman's craft, managed usually by one man and it formed part of the necessary equipment of practically every family which obtained an important part of its subsistence from the sea and lagoon. They could be made by any average man handy with tools; and, because they were of every day practical use to the mass of the people, their manufacture continued from the earliest times with some variations in details. The pahi on the other hand was a sailing vessel used by chiefs and their retinues in traveling by sea between coastal villages, between islands, and on long sea voyages. Because of the size needed for the greater number of crews and passengers, they were built up from a lower middle element or keel. It took time and labor to assemble material, split and shape planks, and fit and lash the parts together. Their construction required the employment of skilled labor. Experts and material could be commanded only by chiefs, who could call upon their people to assemble material, provide food during construction, and goods for the payment of the experts on completion of the work. The circumstances were similar to those surrounding the building of the chief's houses with apsidal ends.
The master builders were usually expert in building both houses and canoes. Thus, though it is true that the Arii were responsible for the building of the superior pahi, it could be attributed to them as a result of their authority and page 516wealth rather than as a result of a different culture. The pahi form of vessel is attributed to Hiro, an early Arii of the leeward islands. It must be admitted that Hiro in his position as a high chief gave orders for the building of a voyaging craft, but the design and construction were carried out by skilled artisans, whose names are mentioned in the story. Skilled experts follow a course of local development conterminous with their crafts. Furthermore, Hiro was an ancestor shared by the people of New Zealand and the Cook Islands, and he lived in the period of the spread of the later Polynesian culture to the Cook Islands. Thus any particular type of ship that was created in Hiro's time must be regarded as a late development and it cannot well be included as an element of a distinct theoretic culture which entered Raiatea some centuries earlier. The earlier Polynesian culture must have made large seagoing vessels in which the original Polynesians entered Polynesia and in which they made successful voyages to New Zealand and Hawaii. The use of planks in ship building is not necessarily an indication of a higher, distinct culture, for the people of the Tuamotu atolls were forced to use planks instead of dugout hulls in the construction of their small fishing canoes in order to make their meager supply of suitable timber go as far as possible.
Another cultural distinction is made by Handy between agriculturists and fishermen, land operations being attributed to the old Tahitians and marine activities to the Arii. It is difficult to reconcile agriculture as an element of the early Tahitian culture with my theory that the earliest Polynesians lost their cultivable food plants during their progress through the atoll end of the Micronesian chain. Neither the Menehune of Hawaii nor the tangata whenua of New Zealand, both of whom belonged to the same culture as the Manahune of Tahiti, had cultivable food plants. Traditions indicate that they were food gatherers, who having nothing to cultivate, explored the endemic resources of their respective groups in the search for edible berries, roots, and other foods that nature could provide. At the same time, they must have devoted considerable attention to fishing. The fishponds of Hawaii are generally credited to the Menehune. If, as I believe, the cultivable food plants came in later from Samoa and probably from Manua to Raiatea, agriculture would perforce be primarily associated with the Arii descendants of Taaroa. The Arii chiefs themselves would not cultivate, for such menial work was beneath them; but the commoners who formed their following must have cultivated the plants which provided the richer food supply that materially assisted in the development of a richer culture in Raiatea. The food plants reached Tahiti later from Raiatea, and it was only then that the Manahune of Tahiti could become agriculturists. Their kindred offshoots in New Zealand and Hawaii had to wait until the spread of the later Polynesian culture brought the food plants to them. Upon the conquest of Tahiti by the Arii invaders from Raiatea, the Manahune were driven back from the coast to the upper valleys and mountains, and became the serfs of the Arii. As serfs and tenants of lands that page 517were divided up among the Arii chiefs, they had to cultivate food for their overlords in order to be allowed to live. They suffered the fate of a conquered people both as regards the land they were allowed to occupy and the work they had to do. Thus the occupation of agriculture was forced upon them primarily by their Arii conquerors.
I cannot accept the different cultural origin, stressed by Handy, between the people who occupy the coast and those who live inland. All Polynesians loved the sea and preferred to live near it. The question of habitat was influenced largely by the varying outcomes of the struggles for supremacy which took place between ambitious chiefs. The defeated shared the fate of the Manahune in being driven inland while the victors occupied the coast. The victors, however, assumed ownership over the fertile valleys, and cultivable lands were leased to their own people. The people, whether chiefs or commoners, obtained their sustenance from both land and sea. The tenants of valley lands, though they might live near their cultivations, had access to the sea and they fished a great deal, for the work of cultivation did not absorb all their time. Even the defeated Manahune were allowed access to the sea, though under restrictions imposed by their Arii masters. In some regions, specialization took place in which full time agriculturists exchanged their products with full time fishermen. This was particularly true in Hawaii, where the ohana family group, occupying a strip of land extending from the sea coast to an inland range, divided themselves into an inland group and a shore group. The inland group cultivated the valleys for taro or the uplands for sweet potato and they supplied the timber and olona fiber needed by their shore cousins for canoes, lines, and nets wherewith to catch fish. The shore group supplied their inland cousins with fish and shellfish in return for taro and sweet potato. Thus in a well-balanced community there was a division in occupation which necessitated a difference in habitat, but it occurred as an economic arrangement between the people of the same ancestry and did not necessarily date back to different cultural origins.
The term ari'i is the Society Islands dialectial form of the widely spread Polynesian term (ariki, ali'i) used to denote a high chief. Thus Handy's use of a class term to denote a culture which underwent development in Raiatea may lead to some elements associated with the social position of chiefs being regarded as elements of the general culture of the people. The statement that the Arii did not cultivate applies to Arii chiefs throughout Polynesia and cannot be accepted as an element of a distinct culture. The Arii chiefs did not cultivate but their people cultivated for them, whether conquered serfs or their own free commoners. This attitude is well illustrated by an incident in which an ariki chief of New Zealand was asked by a judge of the Native Land Court to indicate to the Court where his food cultivations were situated in the block of land under investigation. The ariki drew himself up with dignity, and page 518waving his arm over his people assembled in the Courthouse, he said, "These are my food cultivations."
Space does not permit of a further detailed analysis of the elements of the two cultures postulated by Handy for the Society Islands. I would point out, however, that the material, social, and religious elements attributed to the Arii culture include the changes and developments occurring during the long period of time between the settlement of Raiatea and the advent of Europeans, a period of 13 or 14 centuries, or perhaps longer. After such a lapse of time, it is difficult if not impossible to detach or reconstruct the elements that may have formed part of an original distinctive culture. The elements selected from Samoan culture to show an original affinity with a common Tangaloa or Taaroa culture are also taken from a culture which has undergone change and development for a long period of time. I believe that there was an early contact between the ancestors of the Samoans and the Raiateans when a god named Tangaloa or Ta'aroa was accepted as their principal god by both groups. Intercourse ended early, however, and the Samoan Tangaloa retained an early primitive form whereas the Raiatean Taaroa underwent considerable development. The other Society islands had their own gods. In Tahiti, the principal gods were Tane and Tu. After the conquest of Tahiti, the god Ta'aroa naturally shared the victory of his followers.
After peace was made, however, the Arii chiefs and their priests evidently assumed a diplomatic course with regard to religion. Just as temporal power was shared among the chiefs, so divine power was shared among the major gods of the various islands in the group. Thus Ta'aroa was given authority over the sea and its contents, Tane over the forests and all therein, Tu over war, and Ro'o (Rongo) over peace and agriculture. The major gods thus became departmental gods and they were united in one family under the common parentage of the Sky-father (Atea or Rangi) and the Earth-mother (Papa). It was at this stage of religious reconstruction that the later Polynesian culture spread to the islands to the north, east, and south and carried with it the religious pattern then in operation in the Society Islands. The religious pattern underwent changes in the different islands but the original form of the pantheon seems to have been retained with greater fidelity in New Zealand than elsewhere. In spite of changes, however, the gods Tane, Tu, Rongo, and Tangaroa retained their major rank and the departments previously delegated to them. It is the absence of Tane, Tu, Rongo, Atea and Papa in Samoa that leads me to believe that any contact between Samoa and Raiatea occurred and ended at an early period which preceded the reorganization of the gods in the Society Islands. The influence of the Arii of Raiatea continued to increase and their priests at Opoa created further changes in their mythology and religion. They duplicated their temporal power not only by elevating their god Taaroa to seniority among the gods, but by making him the creator of the other gods. They also gave him a new son named 'Oro who succeeded his father as page 519the principal god of the Society Islands. Ta'aroa seems to have been retired from active service and to have been satisfied with the honor of his new academic position. The fact that this pattern of Ta'aroa and 'Oro was not shared by the other island groups indicates that it was a local development, which took place after the spread of the later Polynesian culture. Though some of the later developments in material culture did diffuse from the Society Islands to the neighboring islands, the failure of a later development in religion to do like-wise indicates that material things may be readily accepted, whereas a change of gods from an established pattern may require enforcement by military conquest.
This digression into religion has been made for the purpose of showing that change and development occurred in religion as well as in material culture. The same processes would be revealed by an analysis of the complexes in social organization. Thus the whole culture underwent change and development during the long period of movement and settlement within Polynesia. Though the dates of changes cannot be accurately determined, some idea of the sequence of their occurrence may be deduced from the traditions of movement and settlement and from the distribution of culture elements in the various island groups. Piddington (82, pp. 201-206) was the first to seriously challenge the two-culture theory by drawing attention to the over emphasis placed on diffusion and the failure to appreciate other cultural processes, such as the factors for change, as a possible cause of the cultural variations which existed. The object of this section on cultural processes has been to give some account of the part that the various processes have played and to determine from this discussion whether the cultural variations in Polynesia were due to diffusion from without or to spontaneous development within one original culture.