Volume 64 1955 > Volume 64, No. 3 > Language in Rapa, by John F. G. Stokes, p 315-340
LANGUAGE IN RAPA
(Slightly edited and arranged by Donald Stanley Marshall, Research Anthropologist for Polynesia, Peabody Museum of Salem.)
When examining the papers of Mr. J. Frank Stimson of Tahiti, Major Donald S. Marshall found 1 a brief account of the language as spoken in Rapa in 1921, and (2) twenty lists of words as spoken in the same island.
The material was mine and was intended for my report to the Bishop Museum on the ethnography of Rapa after examination and criticism by Mr. Stimson who is a keen student of Polynesian dialects, and who had agreed to collaborate with me. He had made some progress with the aid of five natives from Rapa, resident in Tahiti, when pressure of his own business compelled him to set aside the studies temporarily.
Of the lists, the most important was comprised of 350 general terms with Polynesian comparatives. Other lists included kindred terms, numeration, classes of foods, war, fishing, tools, house-building, games, clan names, winds and plant names. The last, about 150, were collected by my wife, since deceased, who volunteered to gather botanical specimens.
Mr. Marshall has edited the account, and most industriously has collated the many lists. The only other change is the identification of the botanical specimens from the collector's numbers as noted in the published works of the Bishop Museum botanists, Doctors Forest and Elizabeth Brown.
I wish to record my appreciation of Mr. Marshall's attention.- 316
Those who came in contact with the Rapa people in early post-European times give differing reports of the dialect spoken in the island. Ellis (III, 364), no doubt expressing the views of Davies, remarked that although in physical appearance the Rapa people resembled more the Tahitians than the New Zealanders, the dialect bore greater affinity to that of the latter. Moerenhaut (I, 138) found that the dialect differed but little from that of Tahiti. However, this was eight years after the Tahitian missionaries became resident. Hale (p. 141) pronounced it to be “with a few verbal exceptions, pure Rarotongan, and this in its minute peculiarities,”—an interesting statement considering Hale's opportunity for making observations. This is confirmed in general by the Rapa natives themselves.
The Tahitian dialect was introduced into Rapa by native missionaries who, in 1826, brought the Bible and some other books of instruction in their own dialect. Works of this nature comprise the only literature available to the Rapa natives.
At present there are two dialects to be heard in Rapa—the Neo-Tahitian and a Rapa-Tahitian hybrid. The official dialect of the island, as well as the religious, is Tahitian. This is used by men, and the hybrid by women and children. Both are understood by each sex. In talking with strangers all natives use Tahitian, the most frequent lapses being noticed among children.
The inclination of the men towards Tahitian is readily understood. Many of them have been educated in Tahiti, and this island has been visited by most men at one time or other. Some stay there for many years before returning to Rapa. Furthermore, the frequent religious discussions, conducted in Tahitian, are largely monopolized by the men.
On the other hand, only a few of the women have travelled. Most of them never leave Rapa. Furthermore, they have attended school less than the men. Other factors may have contributed to the retention of the Rapa dialect by the women. The women are the field labourers as well as the household drudges. Their voices are very powerful, and quite in keeping with their physiques. The Rapa dialect, masculine in note, seems more adapted to them than the softer Tahitian. It is possible also that the local telephonic system of relaying calls across the island may have had an - 317 influence in retaining the palatal sounds dropped by the Tahitians. The calling was observed to be common among all but the adult men.
The Rapa alphabet adopted here contains 13 letters: a, e, ng, i, k, m, n, o, p, r (=l), t, u, v. They are pronounced as is most general in Polynesian with the following exceptions:—
r—the sound from this consonant gradates between a clear l as in English and a soft r, the preponderance being on the l side. It was found both initially and medially as l. There have probably been modern influences pulling both ways. The Tahitian of the native missionaries and schools inclined towards the r. This is emphasized today by the schoolmaster who still instructs in Tahitian and attempts to teach the children to roll the r. In the other direction may be mentioned the possible influence of some natives from islands such as Niuafou and Tokelau where the l is used. Several men and a woman, kidnapped from these islands, were landed in Rapa about 1863. They intermarried with the Rapa natives, and from these unions nearly half the present population has descended. However, some of those of the pure Rapa stock use the l very strongly, while many descendants of the mixtures incline towards the r. As the Rapa dialect seems to have been maintained through the women, it is very questionable if the Niuafou-Tokelau influence was real. Pickering records “Lapa,” as the name of the island, received from a Rapa native before 1848. It may be reasonably assured that Rapa was among the l speak-dialects since the Tahitian influence, under the most favourable conditions, has been unable to uproot the sound in nearly a century.
It is possible that the l had almost the same stability in Rapa as in Samoa (Newell, Pratt's grammar, 4th ed., p. 2) and in Hawaii in both of which it occasionally carried the sound of r or d. However, the d sound was not noticed in Rapa, nor did the rule quoted by Newell for Samoa fully apply in the former island.
As written today, the character is r, which will be retained for convenience of comparison with the dialects spoken in the neighbouring islands. In some of the genealogical books written one to two generations ago the sound was represented by the character l.- 318
Rapa is not alone in the region in possessing the l, which was found to have been also present in Raivavae, the nearest inhabited island. The writer could form no judgement as to how strong the l had been in Raivavae, as it has been completely submerged by the Neo-Tahitian now spoken there.
v—This letter has a sound in accord with the Tahitian, which today approximates the sound in English. In this respect the Rapa sound differs from that of Rarotonga, noted by Trotter as almost equivalent to w.
Occasionally the v is interposed between u and a vowel following. Kainuvi is found in place of Kainui and Kauva'i in place of Kaua'i.
ng—This letter represents the velar nasal. It is written variously as g, ng, and k. However, no interchange with the palatal mute was detected in speech.
PHONIC MUTATIONS AND ELISIONS.
An important feature of the Rapa dialect for comparison with those in the neighbourhood is the elision of the lingual and labial aspirates, respectively written elsewhere as h, and hw, wh or f. In Rapa they have been almost completely elided, but occasionally the vowel following seems to be slightly breathed upon. One might imagine the substitution or addition of another and very faint vowel as was later found in the Tuamotus. Here, while conversing with men from Reao and Hikueru, it was found that the sound they had written down as h frequently included, either preceding or following, what seemed to be faint vowel, such as e or i. For instance koha as written was pronounced kō iha. Hao, the island, was variously spoken of as He ao and Heao. Though written as the lingual aspirate, the character should have represented the labial aspirate. In Rapa, however, the sounds were still fainter, if they existed at all.
The r is occasionally elided. The elision was noted in two instances, kio'e, which existed side by side with the complete form kiore, and Maunga'oa. The latter is a place name, and was said to stand for Maungaroa.
A further elision in the causative 'aka occurs as ka, generally in nouns and adjectives derived from causative verbs.
While the velar nasal was pronounced strongly, it - 319 occasionally changed places with n when the two lay in proximity. For instance: No'ongorupe would sometimes be heard as Ngo'onorupe. This is an example of interchange which has been frequently noted.
There is a certain degree of indifference to the vowel sounds e and i. While first attributing apparent faults in spelling to lack of hearing, I later ascertained that the interchange was real. The place name Anakiri had been noted carefully from one good authority, and later was heard as Anakere from another man equally well informed. The latter, on being questioned, stated that both were correct, and repeated each pronunciation distinctly.
An instability of vowels within the Polynesian dialects has been noted by others and Rapa apparently is no exception. In running through the names on the genealogical lists one notes occasionally the e or the o expressing a in the same person's name. Va'ine, present in compounds is frequently Ve'ine, and once Vi'ine. This recalls Hale's note in reference to the Rarotongan analogy in Rapa of the verbal directive mei in place of the general Polynesian mai. I did not detect it in speech, as the Tahitian dialect now predominates and I was not at the time aware of Hale's comment.
For the purpose of comparison, a brief list of about 500 Rapa words was obtained, principally as the result of specifically questioning the older people and culling from ancient sayings written down by them. Throughout it was attempted to exclude all words not assuredly of Rapa origin, and on this account no doubt very many true Rapa words, identical in form with the Tahitian, have been omitted.
It will be useful to add here some notes that indicate a custom of word changing in Rapa, and conclusions suggested by its presence and the phonic mutations. Many of the words or terms used in ordinary life in Rapa were found, in form or application, to be without affiliation in Polynesian vocabularies. A few others are of very limited affiliation.
Tentatively noted as such are nineteen indicating action, and mental and physical processes; six body parts; nine relating to foods; four to relationship; four of common qualities; seven of plants; four meteorological; ten geophysical; nine miscellaneous.
After searching unsuccessfully through one Malay, one - 320 Micronesian and two Melanesian dictionaries (Malay, Ponape, New Hebrides and Solomon Islands) for affiliations with the Rapa term for “parent,” “women,” “taro,” “banana,” and “sugar-cane,” also strange to Polynesian vocabularies, it appears that these and most of the other unaffiliated terms were locally coined. In a few instances it is clear that the Polynesian term was known in ancient days:—
Vai, a nancient Polynesian name for “water” occurred in place names in Rapa, and in uvai, an aqueduct. The present term is kota'e, evidently derived from Polynesian tahe to flow. The local account is that kota'e replaced kare (kare=wave), which had previously supplanted koringiringi (ringi=to pour) the oldest, as claimed. Pea is locally regarded as the ancient term for female or wife, but the Polynesian vahine, ve'ine, and so forth, occurs in early personal names and as the root of tua‘ine, sister. Umu, Polynesian for the ordinary ground oven has been replaced by kauatu, yet it was known, since umukopuke, the oven for human flesh is a term still extant. Some other examples are given in Table I, page 329.
The terms for “water” mentioned above illustrate one method of coining the new words. Takaviri, generic name for eels, is another illustration: taka “to roll,” viri “to twist”; tekerima, perhaps tekirima “arm suspend” is the term for shoulder.
The terms of limited affiliation as well as they can be identified, are rare: kami'a, is the present term for a canoe, and is apparently only found in addition in the Mangarevan kamiha, a clear case of borrowing by Rapa, as it uses the Polynesian vaka in its legends. Ko'a, “taro food” may be from the Reao (Tuamotu) koha, food in the Tridacna shell; raupaka, “taro leaf” has its affiliation in the Tuamotuan raupaka “a leaf.” Naku, given as naku atu “go,” nakumai “come” and pronounced variously as naku and nako is said by the Rapa natives to be terms also used in Rarotonga. The Rurutu natives made the same statement, but I cannot find the term in Trotter's Rarotongan glossary. It may be a slangish warping of nako “that place yonder,” but suggests the Fijian lako, “to proceed.”
A few terms suggest a survival of very early Polynesian words. Kakona “sweet in taste” is found in Easter Island - 321 hakakonakona and old Rarotongan and Tongarevan kona. In general Polynesian kona is “sweet smelling” in some dialects, “bitter” in others. Raro, “firewood,” “a dry branch” in general Polynesian indicates a branch, green or dry, but the Rarotongan term rara meaning “to desiccate” conforms to the Rapa term. The presence of rara in Rapa is an indication that pake “the sun” has replaced the Polynesian ra.… Ngara'u, “fire,” “live coals” in Rapa has the same meaning in the Tuamotus, although the general Polynesian term indicates “charcoal.” The list is not large.
The majority of the terms referred to are unaffiliated, however, and while their derivation may be ascertained in a few other instances than those given, most appear as strange words. They are also, as stated, terms of ordinary life.
It has been shown by the Murrays Encyclop. Britt., 11th Ed., that in a mixed language like English, the bulk of the early English “words of ordinary life” have persisted in spite of the many foreign introductions (590c, 597b) which now comprise the bulk (3/4 to 4/5) of the language (596d). With such an example it would seem possible to regard the Rapa common words, unaffiliated with the General Polynesian, as relics of some other dialect now all but forgotten.
Tregear (xxii) came to a similar conclusion with respect to the Tuamotuan dialect which he regarded as Polynesian in the bulk, and “crossed” with some foreign tongue: “The numerals and many of the vital words are utterly strange to the Maori linguist; but on the other hand, the Polynesian words have been preserved with great purity of sound and accuracy of meaning.” The paragraph quoted applies almost equally well to Rapa, except that the Polynesian numerals have remained. However, the presence of Polynesian words preserved with purity of sound and accuracy of meaning side by side with many words of daily life of foreign origin, creates a difficulty not explained in the development of the other mixed language referred to.
In place of regarding these unaffiliated words as relics, or survivals (which possibly some are), the writer now believes that most of them will prove to be new terms, coined, metaphorically derived from other terms, or borrowed from other dialects, but in the majority of cases originating in the island itself. Illustrations have already been given of the replacement of general Polynesian terms, known formerly in Rapa, by derived terms.- 322
Such local word-coining has been common among the Polynesians, and yet there is a general lack of information as to the extent to which it has been carried out, and the motives underlying it. The most detail has been given by Newell, in regard to the “chief's language” of Samoa, which, as he explains (p. 85) was originally used in the address of and reference to chiefs and gods. It has become to a large extent “just a language of politeness,” “used in public discourse” and “applied in courtesy to every stranger.” Accompanying this are the complimentary titles for the address of chiefs (Newell, p. 89). These are present in this language, words apparently coined, the use of which is limited to particular classes of chiefs. The language, however, is in general formed by the metaphoric use of common words, or by their replacement with “figurative or picturesque expressions.” Newell (p. 89) quotes Christian's opinion that most of the words employed are of extra-Polynesian origin, and probably many are. A second phase of this language is the prohibition and change in the chief's presence of a common term when present in his name. Similarly, common words present in the names of gods were forbidden to their worshippers.
The common Samoan vocabulary was no doubt affected in time, but the effect could only have been gradual.
Among Newell's examples, a chief is not “sick” (although perhaps seriously ill) but is “indisposed,” “weary,” “turned aside,” “wrapped in a covering” and so forth. He is not “buried,” but “housed in leaves of sandalwood.” His “belly” is his “front.” An “Adze” in his presence becomes a “strong thing” or “turned away,” while a “pig' is a “live thing” or a “four-legged thing.” A “dog” becomes “led by a string.” The chief does not nofonofo, “sit” but does alala, in common parlance, “sits awake at night,” a derivation from ala, “to wake.” On the other hand, he does not “wake” (ala) but does maleifua, perhaps “emits a cough.” The principle seems to be analagous to the indirect address of the Oriental, of European aristocracy and of official life generally, as well as the indirect reference of expression to be found in what is termed “polite society.”
Of the two phases described, Hale (p. 287) noted the first as present in Tonga, and in still less degree in Tahiti. - 323 It occurred in the Hawaiian native speech, and perhaps among the Polynesians generally, although not developed to the same extent as in Samoa.
Williams and Tregear (p. 164) noted in New Zealand the presence of the second phase, namely the change of common terms in the presence of a chief (Maori Dictionary, Auckland, 1892, XI, XII). It also occurred in Hawaii. It was in Tahiti, however, where it was most developed, and to an extent beyond the similar phase in Samoa. It has had the effect of permanently modifying the Tahitian vocabulary.
Under the Tahitian system, termed pi, a word or a syllable present in the name of a god or a chief was banned from common speech, and another substituted. “Every inhabitant is under the necessity of adopting” the change (according to Vancouver in 1791, I, 135) “as any negligence or contempt of it is punished with the greatest severity.” Pomare I even reprimanded Vancouver for using the discarded words. The occasion for the particular change was the accession of Tu, Pomare's son, shortly before Vancouver's visit, when forty or fifty of the “most common words which occur in conversation” were dropped and replaced by forms unaffiliated with those lost. In addition, all the chiefs changed their names. Of the words which now may be traced, some were borrowed from other dialects, some derived as in the Samoan, while there are others which seem to have been newly created sounds, as their affiliations cannot be traced.
Vancouver regarded the new words as permanent and thought the custom was a recent innovation. Hale (p. 288) states that the change was only temporary, and that the original terms were restored after the death of the high individual responsible for the innovations. The permanency of many of these words which has actually resulted is attributed by Hale to their use in the missionaries' writings. However, the most recent example given by Hale is 'amu “eat” substituted for 'ai, on account of Queen Ai-mata. This lady was born about 1810 (13 years after the missionaries' arrival) and succeeded to the throne in 1827. (Ellis, III, 263.) The Scriptural translations were mostly completed before the second date, yet 'amu is the current expression in the Tahitian dialect today.
What the original custom in Tahiti was, probably can - 324 only be surmised today. The effect would be permanent, as the words would be in the vocabulary of the youth and if the king survived the span of an ordinary reign, would thus be perpetrated unless affected by a subsequent change.
There is a possibility that the Tahitian pi as last practiced was derived from the Tuamotus, and one cannot but be impressed with Vancouver's view that it was a recent introduction. Vancouver saw it in operation through Pomare I, who was even so insistent on its observation that he chided his friend Vancouver for the unconscious errors of the latter. From the Memoirs of Ariitaimai (Paris, 1901) we learn that Pomare I (originally Tu) was the grandson of a Tuamotuan who became associated with the chief of the district of Pare-Arue in Tahiti and married into a high family. Pomare inherited the chiefery of Pare-Arue and through intrigue became the predominating chief of Tahiti. In the “Memoirs” he is described as cruel, crafty and ambitious, characteristics which the contemporary missionary accounts confirm. He was hated and feared by the members of Tahitian chiefdom in general, and had to depend upon “savages from the Paumotus” and outsiders for aid in satisfying his political ambitions. Once when a combination of the other Tahitian chiefs threatened to be too strong for him, he took refuge in the Tuamotus.
Another fact of interest is the presence in the neo-Tahitian vocabulary of many words which affiliate only with those in the Tuamotus. I draw attention to Nos. 18, 19, 21, 22, 4 and 1 in the accompanying table. Some others noted from Hale and Tregear are 'e'a “road” (Tuamotuan keka), fene “six” (Tuamotuan hene), mori “oil,” utari “to follow.” Hale (p. 289) also noticed the neo-Tahitian and Tuamotuan affinities. Two of these are ru'i, “night” and hota, “cough,” which replaced po and mare, when Tu assumed the name of Pomare.
Vancouver did not list the forty or fifty words he referred to, but tu was evidently one of them as the change he noticed took place on the accession of Tu, the son of Pomare. The word tu was formerly present in the Tuamotus as is shown by its survival as a root; (tuhou, turotu, and so forth), and analogous Polynesian term tika was substituted, as was done in Tahiti. This presupposes a system of pi in the Tuamotus, and that Tu being the name of a god or of - 325 some early predominant chief was responsible for the adoption of tika for the meanings of tu.
It can only be assumed, with the information at hand, that the second phase of the Samoan “chiefs' language” was the prototype of an ancient Tahitian system. The Samoan system only modified the words in address to the chief, while the Tahitian pi, as known, forced the words on the common tongue.
It is not improbable, therefore, considering the coincidences mentioned above, that there was a system of pi in the Tuamotus, more drastic than that of Samoa, and that through Pomare's Tuamotuan affiliations it modified the existing Tahitian system, and from the combination perhaps, produced one even more drastic than either of the originals.
While we have not the evidence of systems of pi in all the Polynesian islands, the fact that they have been identified in four Polynesian groups, Samoa, Tahiti, New Zealand, and Hawaii and that forms of a ceremonial language existed in Samoa, Hawaii and Tahiti, is sufficient to indicate the presence of analogous systems in Polynesia generally.
The presence then of apparent extra-Polynesian words used in ordinary life in a vocabulary largely composed of Polynesian words of great purity and in a dialect with a grammar and phonology differing inappreciably, if at all, from that of any other dialect in the language, is thus explained. Tregear himself (p. 201) suggests that the second phase of Polynesian neology “accounts for much of the difference in dialects found between certain tribes.”
Such an explanation is not a denial that foreign words may have crept into the Polynesian dialects, but it demonstrates that if they did, they were adopted progressively and did not materially affect the purity of the tongue. The presumption is that most of the new terms arose locally, and that their affiliations should not be sought outside Polynesia until it has been ascertained that they were not local creations. It must be remembered that while it was the words of ordinary life which were principally affected by these changes, this result was natural under the pi system, since such words composed the personal names of the Polynesians, whether commoners or chiefs.
That some form of the system existed in Rapa there can be no doubt. It is demonstrated by the examples given, but - 326 I do not know of its details. The descendants of Rapa's royalty, resident in Tahiti state, in reply to enquiries through Mr. Stimson, that the pi of Tahiti was not present in their native island. It is of interest to note that of the unusual word forms with affiliations which could be recognized, more were traced to the Tuamotus than to the other groups.
There are differences in application to be noted between the Tahitian and Rapa systems. In Tahiti, ancient place names suffered, as in Vaiete, Vaiari, Mehetu, and so forth, which became Papeete, Papeari, Mehetia, and so forth. In Rapa we still find Vai'ou, Vaitokerau, and many others. In Tahiti also the syllables were attacked: as in anapape from anavai, and Mehetia above. Anavai and uvai remained unchanged in Rapa.
The basis for the described forms of Polynesian neology, as may be observed, is a respect willingly accorded to the chiefs or demanded by them. It does not suggest the survival of a foreign language used by chiefs, since in order that the system may be put into operation the words must be known to the commoners.
The impulses underlying the systems were no doubt various, and from the facts already at hand, I am inclined to believe that the more important were affectation and the desire for novelty which one cannot fail to understand when the people and the limitations of their environment are considered. Their readiness, almost eagerness, for change is shown by their desire for new names which are attached not only to persons but to lands and canoes. The system is ancient, and can be traced back in Polynesian traditions for very many generations. It may be regarded as a custom.
The Polynesians, as they have been described, were an intellectual people with great mental vivacity. They were conservative in the matter of religion which gripped them so completely, and their material development was limited by their natural resources. With such limitations one would expect that their intellectural development would be hampered, but we find among them historians, poets, orators, and actors. Punning contests and mimicry were also sources of entertainment. Proverbs formed part of their speech. They were sociable and hospitable by nature, always asking and receiving or spreading the news. The - 327 language, where it did not touch the ritual, was “free,” and it will thus be understood why the vocabulary could readily become subject to modification, as well as the rapidity with which a word, once launched, would become current. Combined with the desire for novelty is the affectation attributed to many neologists of other places.
It is possible that the Polynesian consonantal mutations and elisions were due to similar causes. It will be remembered that Polynesian is a “vowel language,” and that the so-called Polynesian dialects are groupings principally based on the varying consonants present.
We have an interesting illustration of the rapid change from t to ts in Niue. Pratt (Newell, XII) noted in 1876 that fifteen years previously only a portion of the younger people changed the t to ts before i, and that since then the practice had spread generally, some of the people using ts before any vowel. It is also of interest to note that Dr. Loeb, when he returned from Niue, used a sound for t which was almost a clear s, when speaking the dialect. It is evident then that a change from t toward s is possible in less than three generations in a cultivated Polynesian language. Possibly the ts was a borrowing from Tonga.
Pratt also remarked that the Samoans, who had previously elided the k, were then in process of substituting that sound for t, similarly to a previous practice among the Hawaiians. Mr. Stimson has informed me of an occasional use of k for t by Tahitians who had been to the Tuamotus.
The rapid progress of the kappation of t and other dialectical changes in Samoa may be judged from the protest in the governor's official address in 1911 to the local Teachers' Institute and the appeal for the preservation of the language in its purity. (Quoted by Churchill, Rapanui, p. 25). In the report it is stated that “the majority of the Samoan teachers speak this gibberish all the time.”
A curious reversion of k back to t, occurring in Hawaii, has been drawn to my attention by Mr. Emory of the Bishop Museum. In Honolulu today the younger Hawaiian singers are using t instead of the Hawaiian k, which was formerly marked. The change was first noticed by Mr. Emory in 1921, after an absence of some years. In 1920 organized bands of Samoan entertainers and singers began to visit Honolulu and became very popular with the younger - 328 Hawaiians. These Samoans used the t in singing. The sequence is obvious. The change is due in part to the affectation of the young people, and also, perhaps to a recognition of the musical quality of the t.
An instance of affectation following mimicry was also noticed by Mr. Emory, who found the young people of Lanai using a diphthonic syllable, kio in place of ko, when singing. It was traced to a vocal peculiarity of the teacher—a native.
The instances quoted are of course only modern “straws” but they may well indicate the course followed by the Polynesians of earlier days, since they are supplied by the descendants of these people.
Brown suggests as an explanation of these mutations, a change in climatic environment towards the equator (p. 19) through a relaxation of the tissues, including those of the speech organs (p. 17). This may be so in some cases, but it does not explain why the Tokelau people (being nearer the Equator than Samoa and derived therefrom according to Hale), retained the k while the Samoans dropped it; nor why the Samoans, having dropped the k, have begun to restore it in substitution for their t. It is equally unsatisfactory when applied to the temperate climate of New Zealand, where some tribes substitute n for ng. (Williams, W. L., p. xii).
I believe a more reasonable explanation may be based on the facts presented, which point to an indifference on the part of the young Polynesian to the preservation of the language in its purity, leading to the mutation or elision of consonants as the fad is taken up and becomes general. That it began in ancient days I have no doubt, since the same indifference is shown in the name and word changing.
Applying these explanations to the origin of the Rapa dialect I believe, with Brown, that within the Polynesian area the phonology of the dialect is a more important key to the affinities than the vocabulary. This is well illustrated by the remarkable purity of such words as are recognized as Polynesian among all the dialects.- 329
(Collated by D. S. Marshall from material collected by J. F. E. Stokes, Mrs. Stokes and J. Frank Stimson.)
1 In order to accord with standard practice in written Maori, we have substituted the letters ng for the phonetic symbol used by the author. This does not represent a final decision as to the orthography to be used in future in the Journal. We intend to consider the question further before coming to such a decision.—Editors.