rnia il | GIFT OF SEELEY W. MUDD and GEORGE I. COCHRAN MEYER ELSASSER DR. JOHN R. HAYNES WILLIAM L. HONNOLD JAMES R. MARTIN MRS. JOSEPH F. SARTORI to the UNIVERSITY OF CALIFORNIA SOUTHERN BRANCH THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. Philosophical Inquiry is essentially the chief intellectual study of our age. It is proposed to produce, under the title of " The English and Foreign Philosophical Library," a series of works of the highest class connected with that study. The English contributions to the series consist of original works, and of occasional new editions of such productions as have already attained a permanent rank among the philosophical writings of the day. Beyond the productions of English writers, there are many recent publications in German and French which are not readily accessible to English readers, unless they are competent German and French scholars. Of these foreign writings, the translations have been entrusted to gentlemen whose names will be a guaran- tee for their critical fidelity. " The English and Foreign Philosophical Library" claims to be free from all bias, and thus fairly to represent all develop- ments of Philosophy, from Spinoza to Hartmann, from Leibnitz to Lotze. Each original work is produced under the inspection of its author, from his manuscript, without intermediate sugges- tions or alterations. As corollaries, Avorks showing the results of Positive Science, occasionally, though seldom, find a place in the series. The series is elegantly printed in octavo, and the price regu- lated by the extent of each volume. The volumes will follow in succession, at no fixed periods, but as early as is consistent with the necessary care in their production. THE FOLLOWING HAVE ALREADY APPEARED:— In Three Volumes, post 8vo, pp. 350, 406, and 384, with Index, cloth, £1, us. 6d. A HISTORY OF MATERIALISM. By Professor F. A. LANGE. Authorised Translation from the German by Ernkst C. Thomas. "This is a work which has long and impatiently been expected by a large circle of readers. It has been well praised by two eminent scientists, and their words have created for it, as regards its appearance in our English tongue, a sort of ante-natal • reputation. The reputation is in many respects well deserved. The book is marked hroughout by singular ability, abounds in striking and suggestive reflections, subtle nd profound discussions, felicitous and graphic descriptions of mental and social movc- ients, both in themselves and in their mutual relations." — Scotsman. THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. Post 8vo, pp. xii. — 362, cloth, 10s. 6d. NATURAL LAW: An Essay in Ethics. By EDITH SIMCOX. Second Edition. "Miss Simcox deserves cordial recognition for the excellent work she has done in vindication of naturalism, and especially for the high nobility of her ethical purpose." — Athencewm. In Two Volumes, post 8vo, pp. 268 and 288, cloth, 15s. THE CREED OF CHRISTENDOM: ITS FOUNDATIONS CONTRASTED WITH ITS SUPERSTRUCTURE. By W. R. GREG. Eighth Edition, with a New Introduction. "No candid reader of the 'Creed of Christendom ' can close the book without the secret acknowledgment that it is a model of honest investigation and clear exposition, conceived in the true spirit of serious and faithful research." — Westminster Review. Third Edition. Post 8vo, pp. xix. — 249, cloth, 7s. 6d. OUTLINES OF THE HISTORY OF RELIGION TO THE SPREAD OF THE UNIVERSAL RELIGIONS. By C. P. TIELE, Dr. Theol., Professor of the History of Religions in the University of Leiden. Translated from the Dutch by J. Estlin Carpenter, M.A. " Few books of its size contain the result of so much wide flunking, able and laborious study, or enable the reader to gain a better bird's-eye view of the latest results of inves- tigations into the religious history of nations. . . . These pages, full of information, these sentences, cut and perhaps also dry, short and clear, condense the fruits of long and thorough research." — Scotsman. Third Edition. Post 8vo, pp. 276, cloth, 7s. 6d. RELIGION IN CHINA: Containing a Brief Account of the Three Religions of the Chinese, with Observations on the Prospects of Christian Conversion amongst that People. By JOSEPH EDKINS, D.D., Peking. " We confidently recommend a careful perusal of the present work to all interested in this great subject." — London and China Express. " Dr. Edkins has been most careful in noting the varied and often complex phases of opinion, so as to give an account of considerable value of the subject." — Scotsman. Post 8vo, pp. xviii. — 198, cloth, 7s. 6d. A CANDID EXAMINATION OF THEISM. By PHYSICUS. " It is impossible to go through this work without forming a very high opinion of hi: speculative and argumentative power, and a sincere respect for his temperance of state inent and his diligent endeavour to make out the best case lie can for the views he rejects." — Acadi mil. THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. Post 8vo, pp. xii. — 282, cloth, 10s. 6d. THE COLOUR SENSE : Its Origin and Development. AN ESSAY IN COMPARATIVE PSYCHOLOGY. By GRANT ALLEN, B.A., Author of "Physiological ..Esthetics." " The book is attractive throughout, for its object is pursued with an earnestness and singleness of purpose which never fail to maintain the interest of the reader." — Saturday Jieciew. Post 8vo, pp. xx. — 316, cloth, 7s. 6d. THE PHILOSOPHY OF MUSIC. BEING THE SUBSTANCE OF A COURSE OF LECTURES Delivered at the Royal Institution of Great Bkitain, in February and March 1877. By WILLIAM POLE, Mus. Doc. Oxon. Fellow of the Royal Societies of London and Edinburgh ; one of the Examiners in Music to the University of London. " We may recommend it as an extremely useful compendium of modern research into the scientific basis of music. There is no want of completeness." — Pall Mall Gazette Post 8vo, pp. 168, cloth, 6s. CONTRIBUTIONS to THE HISTORY of the DEVELOPMENT OF THE HUMAN RACE. LECTURES AND DISSERTATIONS By LAZARUS GEIGER, Author of "Origin and Evolution of Human Speech and Reason." Translated from the Second German Edition by David Asher, Ph.D., Corresponding Member of the Berlin Society for the Study of Modern Languages and Literature. " The papers translated in this volume deal with various aspects of a very fascinating study. Herr Geiger had secured a place in the foremost ranks of German philologefs, but he seems to have valued his philological researches chiefly as a means of throwing light on the early condition of mankind. He prosecuted his inquiries in a thoroughly philosophical spirit, and he never offered a theory, however paradoxical it might seem at first sight, for which he did not advance solid arguments. Unlike the majority of German scholars, he took pleasure in working out his doctrines in a manner that was likely to make them interesting to the general public; and his capacity for clear and attractive exposition was hardly inferior to that of Mr. Max Muller himself." — St. James's Gazette, Post Svo, pp. 350, with a Portrait, cloth, ios. 6d. DR. APPLETON : His Life and Literary Relics. By JOHN H. APPLETON, M.A., Late Vicar of St. Mark's, Staplefield, Sussex ; AND A. H. SAYCE, M.A., Fellow of Queen's College, and Deputy Professor of Comparative Philology, Oxford. " Although the life of Dr. Appleton was uneventful, it is valuable as iHus1 rating the manner in which the speculative and the practical can be combined. His biographers talk of his geniality, his tolerance, his kindliness ; and these characteristics, combined with his fine intellectual gifts, his searching analysis, his independence, his ceaseless energy and ardour, render his life specially interesting." — Nonconformist. THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. Post 8vo, pp. XXV1.-370, with Portrait, Illustrations, and an Autograph Letter, cloth, 1 2s. 6d. EDGAR QUINET : HIS EARLY LIFE AND WRITINGS. By RICHARD HEATH. " Without attaching the immense value to Edgar Quinet's writings which Mr. Heath considers their due, we are quite ready to own that they possess solid merits which, perhaps, have not attracted sufficient attention in this country. To a truly reverent spirit, Edgar Quinet joined the deepest love for humanity in general. Mr. Heath . deserves credit for the completeness and finish of the portraiture to which he set his hand. It has evidently been a labour of love, for the text is marked throughout by infinite painstaking, both in style and matter." — Globe. Second Edition, post 8vo, cloth, 7s. 6d. THE ESSENCE OF CHRISTIANITY. By LUDWIG FEUERBACH. Translated from the Second German Edition by Marian Evans, Translator of Strauss's " Life of Jesus." " I confess that to Feuerbach I owe a debt of inestimable gratitude. Feel- ing about in uncertainty for the ground, and finding everywhere shifting sands, Feuerbach cast a sudden blaze into the darkness, and disclosed to me the way. : ' — From S. Baring -Gould's "The Origin and Development of Religious Belief" Part II. , Preface, page xii. Third Edition, revised, post 8vo, pp. 200, cloth, 3s. 6d. AUGUSTE COMTE AND POSITIVISM. By the late JOHN STUART MILL, M.P. Post 8vo, pp. xliv. — 216, cloth, 7s. 6d. ESSAYS AND DIALOGUES OF GIACOMO LEOPARD! Translated from the Italian, with Biographical Sketch, by Charles Edwardes. "This is a good piece of work to have done, and Mr. Edwardes deserves praise both for intention and execution." — Athenaeum. " Gratitude is due to Mr. Edwardes for an able portraiture of one of the saddest fignres in literary history, aud an able translation of his less inviting and less known works." — Academy. Post 8vo, pp. xii. — 178, cloth, 6s. RELIGION AND PHILOSOPHY IN GERMANY: A Fragment. By HEINRICH HEINE. Translated by John Snodgrass, Translator of " Wit, Wisdom, and Pathos from the Prose of Heinrich Heine " " Nowhere is the singular charm of this writer more marked than in the vivid pages of this work. . . . Irrespective of subject, there is a charm about whatever Heine wrote that captivates the reader and wins his sympathies before criticism steps in. But there can be none who would fail to admit the power as well as the beauty of the wide-ranging pictures of the intellectual development of the country of deep thinkers. Beneath his grace the writer holds a mighty grip of fact, stripped of all disguise and made patent over all confusing surroundings." — Bookseller. THE EXGLISH AND FOREIGN PHILOSOPHICAL LIBRAE V. Post 8vo, pp. xviii. — 310, with Portrait, cloth, 10s. 6d. EMERSON AT HOME AND ABROAD. By MONCURB D. CONWAY. Author of "The Sacred Anthology," "The Wandering Jew," "Thomas Carlyle," &c. This book reviews the personal and general history of the so-called "Trans- cendental " movement in America ; and it contains various letters by Emerson not before published, as well as personal recollections of his lectures and con- versations. "Mr. Conway has not confined himself to personal reminiscences ; he brings together all the important facts of Emerson's life, and presents a full account of his governing ideas — indicating their mutual relations, and tracing the processes by which Emerson gradually arrived at them in their mature form." — St. James's Gazette. Seventeenth Edition. Post 8vo, pp. xx. — 314, cloth, 10s. 6d. ENIGMAS OF LIFE. By "W. R. GREG. " What is to be the future of the human race? What are the great obstacles in the way of progress? What are the best means of surmounting these obstacles? Such, in rough statement, are some of the problems which are more or less present to Mr. Greg's mind ; and although he does not pretend to discuss them fully, he makes a great many observations about them, always expressed in a graceful style, frequently eloquent, and occasionally putting old subjects in a new light, and recording a large amount of read- ing and study." — Saturday Review. Post 8vo, pp. 328, cloth, 10s. 6d. ETHIC DEMONSTRATED IN GEOMETRICAL ORDER AND DIVIDED INTO FIVE PARTS, WHICH TREAT I. Of God. II. Of the Nature and Origin of the Mind. III. Of the Origin and Nature of the Affects. IV. Of Human Bondage, or of the Strength of the Affects. V. Of the Power of the Intellect, or of Human Liberty. By BENEDICT DE SPINOZA. Translated from the Latin by William Hale White. " Mr. White only lays claim to accuracy, the Euclidian form of the work giving but small .-cope for literary finish. We have carefully examined a number of passages with the original, and have in every case found the sense correctly given in fairly readable English. For the purposes of study it may in most cases replace the original ; more Mr. White could not claim or desire." — Athenamm. In Three Volumes. Pest 8vo, Vol. I., pp. xxxii. — 532, cloth, 18s. ; Vols. II. and HI., pp. viii. — 496 ; and pp. viii. — 510, cloth, 32s! THE WORLD AS WILL AND IDEA. By ARTHUR SCHOPENHAUER. Translated from the German by R. II. Haldane, M.A., ami John Kemp, M.A. " The translators have dune their part very wed, for, as they say, their work lias been one of difficulty, especially as the style of the original is occasionally ' involved and loose.' At the same time there is a force, a vivacity, a directness, in the phrases and sentences of Schopenhauer which are very different from the manner of ordinary German philosophical treatises. He knew English and English literature thoroughly ; he ad- mired the clearness of their manner, and the popular strain even in their philosophy, and these qualities he tried to introduce into his own works and discourse." — Scotsman. THE EXGLISH AXD FOREICX PHI1 !CAL LIBRARY. In Three Volumes, post 8v0, pp. xxxii. — 372 ; vi. — ;eS ; and viii.— 360, cloth, £i, 1 is. oil. THE PHILOSOPHY OF THE UNCONSCIOUS. By EDWARD VON HARTMANN. [Speculative Results, according to the Inductive Method of Physical Soience.] Authorised Translation, hy William C. Coupland, M.a. r«n Editions of the German original have been eold siua • - ippearance in 1S6S. "Mr. Coupland has been remarkably successful In dealing with the difficulties of 11 irtm.mn. . . . [t must be owned that the book merited the honour of translation. Its collection of facts alone would he sufficient to deserve this, and the appendix in the third volume, giving a re of Wurdt's payoho-phyaios, Is a valuable addition to English psychology." — Athtiiatim. Three Vols., post 8vo, pp. viii.— 368 : is. — 225 ; and xxvii. — 327, cloth, £i, 1 is. 6d. THE GUIDE OF THE PERPLEXED OF MAIMONIDES. Translated from the Original Text, and Annotated by If. Fkiedlandi k, Ph.D. Vol. 1. has already been published under the auspices of the Hebrew Litera- ture Society ; but it has now been determined that the complete work, in three volumes, shall be issued in the English ami Foreign Philosophical Library. ••It is with sincere satisfaction that we welcome an English trans! itlon of the well- known tractate of Maimonides, Morek Nebhukhim, or, 'Guide of the Perplexed.' . . . Dr. Friedlander has performed his work in a manner to secure the hearty aoknov ment of students, ' — Saturdo;/ R< view, •' From every point ofvlew a SUCOessful production."— Aciuhmii. • Dr. Friedlander has conferred a distinct boon on the .lews of England and America." — Jewish Chronicle. Tost, Svo, pp, xii. and 395, cloth, with Portrait, 14s. LIFE OF GIORDANO BRUNO, THE NOLAN. By I. FRITH. Revised by Professor Moriz Carriers. " Tho interest of the hook lies in the conception of Bruno's character and in the elucidation of his philosophy. . . . His 'writings-dropped from him wherever he went, and were published in many places. Their number is very lari e, and the bibliographical appendix is not the least valuable part of this volume. . . . We are tempted to multiply quotations from the pages before us, for Bruno's utterances have a rare charm through their directness, their vividness, their poetic force. Bruno stands in relation to later philosophy, to Kant or Hegel, as Giotto stands to Raphael. We feel the merit of the more complete and perfect work; but wear.' moved and attraoted by the greater indi- viduality which accompanies t he Struggle alter expression in an earlier and simpler age, Students of philosophy will know at onoe how much labour has been bestowed upon this modest attempt to set forth Bruno's significance as a philosopher. We have oontentod ourselves with showing how much the [enei J reader may gain from a study of its pages, which are never overburdened by technicalities and are never dull." — Athenaeum, THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. Post 8vo, pp. sxvi. and 414, cloth, 14a. MORAL ORDER AND PROGRESS: A\ ANALYSIS OF ETHICAL CONCEPTIONS. I3y 8. AliKXANDEB, Fallow of Lincoln College, Oxford. This work is an account of the Faotors involved in the two oeni ral phenomena of Order or Equilibrium, and Progress, which are shown to be essential to morality. Its method is to group ethical faots under the main working ooncep tions of morality, [t treats Ethics independently of Biology, but the result is to con lirm the 1 beory of Evolution by showing that the characteristic differences of moral action are such as should be expeoted if that theory were true. In particular, Book III. aims at proving that moral ideals follow, in their origin aud development, the same law as natural Bpecies. Post 8vOj pp. xx. and 314, cloth, ids. 6d. THE SCIENCE OF KNOWLEDGE. By J. G. FICHTB. Translated from tie- (ierman by A. K. KtlOKGEB. With a New Introduction by Professor VV, T. Harris. Post 8vo, pp. x. .and 504, cloth, 12s. 6d. THE SCIENCE OF RIGHTS. By J. G. FICHTB. Translated from the German by A. E. Kroeqer. With a New Introduction by Professor W. T. HARRIS. Fichte belongs to those great men whose lives are an everlasting possession to mankind, and whose words the world does not willingly let die. 1 1 is character stands written in his life. a. massive Imt severely si in 1 ile whole. It has no parts, the depth and earnestness on which it- rests speak forth alike in his thoughts, words and actions. No man of his time few, perhaps, of any time exercised a more powerful, spirit -stirring influence over the minds of his fellow-count ry men. The impulse which he communicated to the national thought extended far beyond the sphere of his personal intluences ; it has awakened, it will still awaken, hi^h emotion and manly resolution in thousands who never heard his voice. The ceaseless effort of his life was to rouse men to a sense of the divinity of their own nature, to fix their thoughts Upon a spiritual life as the only true and real life; to teach them to look upon all else as mere show and unreality; and thus to lead them to constant, effort after the highest ideal of purity, virtue, independence and Belf-denial. In Two Volumes, post. 8vo, pp. iv. — 47.S and x. — 5^1 cloth, 21s. JOHANN GOTTLIEB FICHTE'S POPULAR WORKS. THE NATURE OF THE SCHOLAR ; Till'. VOCATION OF THE SCHOLAR. ; Tin: VOCATION OF max; Tilt': DOCTRINE OF RELIGION; CHARACTERISTICS <>!'' THE PRESENT AGE; <>l TLINE8 OF 'MM I"" J 1 : 1 ■ r. OF KNOWLEDGE. With a Memoir by William S.mitii, LL.D. THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. EXTRA SERIES. Two Volumes, post 8vo, pp. xxii. — 328 and xvi. — 358, with Portrait, cloth, 2 is. LESSING : His Life and Writings. By JAMES SIME, M.A. Second Edition. " It is to Lessing that an Englishman would turn with readiest affection. We cannot but wonder that more of this man is not known amongst us." — Thomas Oarlyle. " But to Mr. James Sime has been reserved the honour of presenting to the English public a full-length portrait of Lessing, in which no portion of the canvas is uncovered, and in which there is hardly a touch but tells. We can say that a clearer or more compact piece of biographic criticism has not been produced in England for many a day." — Westminster Review. " An account of Lessing's life and work on the scale which he deserves is now for the first time offered to English readers. Mr. Sime has performed his task witli industry, knowledge, and sympathy ; qualities which must concur to make a successful biogra- pher." — Pall Mall Gazette. " This is an admirable book. It lacks no quality that a biography ought to have. Its method is excellent, its theme is profoundly interesting : its tone is the happiest mixture of sympathy and discrimination : its style is clear, masculine, free from effort or affecta- tion, yet eloquent by its very sincerity." — Standard. ''He has given a life of Lessing clear, interesting, and full, while he has given a study of his writings which bears distinct marks of an intimate acquaintance with his subject, and of a solid and appreciative judgment." — Scotsman. In Three Volumes, post 8vo. Vol. I. pp. xvi. — 248, cloth, 7s. 6d. ; Vol. II. pp. viii. — 400, cloth, 10s. 6d. ; Vol. III. pp. xii. — 292, cloth, 9s. AN ACCOUNT OF THE POLYNESIAN RACE : ITS ORIGIN AND MIGRATIONS, AND THE ANCIENT HISTORY OF THE HAWAIIAN PEOPLE TO THE TIMES OF KAMEHAMEHA I. By ABRAHAM FORNANDER, Circuit Judge of the Island of Maui, H.I. " Mr. Fornander has evidently enjoyed excellent opportunities for promoting the study which has produced this work. Unlike most foreign residents in Polynesia, he has acquired a good knowledge of the language spoken by the people among whom he dwelt. This has enabled him, during his thirty-four years' residence in the Hawaiian Islands, to collect material which could be obtained only by a person possessing such an advantage. It is so seldom that a private settler in the Polynesian Islands takes an intelligent interest in local ethnology and archaeology, and makes use of the advantage he possesses, that we feel especially thankful to Mr. Fornander for his labours in this comparatively little- known field of research." — Academy. " Offers almost portentous evidence of the acquaintance of the author with the Polynesian customs and languages, and of his industry and erudite cave in the analysis and comparison of the tongues spoken in the Pacific Archipelagoes." — Scotsman. In Two Volumes, post Svo, pp. viii. — 408 ; viii. — 402, cloth, 21s. ORIENTAL RELIGIONS, AND THEIR RELATION TO UNIVERSAL RELIGION. By SAMUEL JOHNSON. I.— INDIA. LONDON : TEUBNER & CO., LUDGATE HILL. PRINTED BY BAI.LANTYNE, HANSON AND CO. EDINBURGH AND LONDON. 503 — 21/1/go — G. THE ENGLISH AND FOREIGN PHILOSOPHICAL LIBRARY. EXTRA SERIES. VOLUME III. •■iCI-i AN ACCOUNT OP THE POLYNESIAN RACE ITS ORIGIN AND MIGRATIONS AND THE ANCIENT HISTORY OF THE HAWAIIAN PEOPLE TO THE TIMES OF KAMEHAMEHA I. BY ABRAHAM FOBNANDEE, CIKCUIT JUDGE OF THE ISLAND OF MAUI, H.I. VOL. L Scconto (Edition, LONDON: KEGAN PAUL, TRENCH, TRUBNER, & CO., Lt? 1890. [All rights reserved.] 8Gbib BALLANTYNE, HANSON AND CO. EDINBURGH AND LONDON ! 70 TO MY DAUGHTER CATHERINE KAONOHIULAOKALANI FORNANDER, ©jjte SHorfc 75 AFFECTIONATELY DEDICATED, AS A REMINDER OF HER MOTHER'S ANCESTORS AND AS A TOKEN OF HER FATHERS LOVE. ABR. FORNANDER. "As made up of legendary accounts of places and personages, it (mytho- logy) is history ; as relating to the genesis of the gods, the nature and adventures of divinities, it is religion." — Native Races of the Pacific States, H. H. Bancroft, vol. iii. p. 14. "It is now a recognised principle of philosophy, that no religious belief, however crude, nor any historical tradition, however absurd, can be held by the majority of a people for any considerable time as true, without having in the beginning some foundation in fact." ..." "We may be sure that there never was a myth without a meaning; that mythology is not a bundle of ridiculous fancies invented for vulgar amusement ; that there is not one of those stories, no matter how silly or absurd, which was not founded in fact, which did not once hold a significance." — Ibid., vol. iii. pp. 16, 17. "The fact of an immigration, and the quarter from which it came, are handed down from father to son, and can scarcely be corrupted or forgotten, unless in the case where the people sink into absolute barbarism. " — Essay ii. book vii. of Herodotus, edited by G. Rawlinson. CONTENTS. tt VOL. I. Preface N Resume" of conclusions arrived at Names of places indicating descent of immigrants Names of cardinal points leading to same conclusion Legendary and mythological reminiscences Physical resemblance of Polynesian and pre-Malay tribes in Malaysia ......... Polynesian language, one of the oldest .... Probate route of Polynesians into the Pacific and contact with Papuan tribes . Final halt on the Fiji group Reminiscences of contact with Papuans .... The Cushite people and civilisation as connected with Poly nesian legends, myths, and cult Traces of solar and serpent worship Polynesian god Oro Maui, Shea, Lingam symbolism, sacrificial stones The Hindu Yama and Polynesian Tawjaroa compared The Hawaiian goddess Pele, a personified reminiscence of an ancient religious schism ; also considered etymologically Polynesian national appellations : Take, Menehune . Further recollections of a "Western origin .... Polynesian accounts of creation ; comparison with Hebrew and Chaldean versions ; Hawaiian chants .... The Hawaiian paradise, Kalana-i- Uauola, Pali-tili, the fountain of life, Wai-ola-loa-a-Kane ...... The tree of life, Ulu-iapu-a-Kane ; the forbidden fruit and fall of man, with chants ....... Creation of angels, spirits, and their rebellion ; the proper place of Kanaloa, Tanyaroa ...... ^Remarks on the genealogies from the first man, Kiunuhonua, to the Flood Legends of the Flood The descendants of Nuu, and apparent resemblance of Poly nesian and Hebrew legends ..... Two hypotheses of such resemblance reviewed . Circumcision Manner of burial PAGES i-xvi 1-2 2-i5 15-18 18-26 26-^0 34-36 37-42 42-45 45 45-51 5i 51-53 53-57 57-59 59-77 77 -79 79 -83 «3 -85 85 88 -87 -96 96- IOI IOI- 103 104- 106- 105 108 CONTENTS. PAGES Of castes ........... . 109 -112 The tabu . 113-114 Tattooing 115 Holy waters 1 1 5-1 17 Cities of refuge 118 Division of the year . 11S-125 Names of months and days 125-127 Superstition . 127-129 Human sacrifices 129-130 Cannibalism 130-132 > Review of legend of Hawaii-loa 132- 1 "7 The mixed condition of the Polynesian family .... 137-1^9 The Polynesian language no kin to the Malay .... 139-141 Rev. S. J. Whitmee's opinion reviewed 141-144 The numeral system 144-158 Resume of foregoing data 159 Time of Polynesian arrivals in the Pacific — The Wakea period . 160-168 Approximate division of periods 168 Hawaiian history : Wakea probably a chief in Gilolo . . 169-172 Marquesan migrations ........ 173-178 The name Samoa considered 179 Samoans and Hawaiians came to the north of New Guinea, and Tongans and Marquesans to the south of it . . . 180 - Hawaiian Genealogies — Kumuhonua 181-184 Kumu-uli .......... 184-185 Opukahonua ......... 186-187 Welaahilani, &c 187 Nana-Ulu — Genealogy 188-189 Ulu-Hema , 190-192 Hana-laa-iki ,,........ 193 Ulu-Puna „ 194-196 >. Remarks on the genealogies, their discrepancies, and their inter- polations during the migratory period .... 197-204 Legend of Wakea and Papa . 205 From Kii to Maweke and his contemporaries .... 206-209 Appendix — 1. Legend of the world -egg ...... 211 Other legends, Brahminical and Javanese . . . 212-213 2. Te Vanana na Tanaoa (chant of), Marquesan chant of creation 214-210, 3. Tahitian chant of Creation ...... 220-224 4. Te Tai Toko (the Deluge), Marquesan chant . . . 225-235', 5. Baptismal chant, New Zealand ..... 236-237 6. Hawaiian signs and omens 238-239 7. Sundry Hawaiian customs compared .... 240-243 8. J. Crawford's view of the introduction of Malay words in the Polynesian language reviewed . . . 244-247 9. Comparative genealogical tables of Ulu and Nana-Ulu . 248 PREFACE. When a gentieman, whose genius and talents have secured for himself one of the curule chairs in the republic of letters, introduces a blushing aspirant, his name becomes a voucher for the respectability of the latter, and his "favete Unguis" ensures an attentive hearing until the close of the performance. But we are not all born with a silver spoon, and many an author, like myself, has had to bear the double burden of introducing himself as well as his subject. But when a writer presents himself with new discoveries, and new ideas based upon them, the reader has a right to inquire who the writer is, and if his discoveries are genuine, before he exercises his judgment upon the ideas submitted for his acceptance. It is meet and proper, therefore, on entering upon ground so little travelled as that of Polynesian Archaeology, on presenting myths and legends to the inspection of the literary world some of which have never darkened a sheet of paper before, that I should state my right to present them, how I came by them, and also the lights which guided and the aids which assisted me on the journey. x PREFACE. Thirty-four years' residence in the Hawaiian group ; nineteen years' position in various offices under the Government; a thorough local and personal knowledge of every section of the group, acquired during numerous journeys ; my knowledge of the language, and the fact — though with all due modesty I state it — that I am well known, personally or by reputation, to every man within the group, from the King on the throne to the poorest fisherman in the remotest hamlet ; — all these considerations give me a right to speak on behalf of the Polynesian people, to unveil the past of their national life, to unravel the snarled threads of their existence, and to pick up the missing links that bind them to the foremost races of the world, — the Arian and the Cushite. Thus much, though reluctantly, I have felt bound to say in vindication of my right to be the spokesman of a people whom no one knew till a hundred years ago, and whom no one even now recognises as a chip of the same block from which the Hindu, the Iranian, and the Indo-Euro- pean families were fashioned. When first I entertained the idea of preparing myself for a work on Polynesian Archaeology, I employed two, sometimes three, intelligent and educated Hawaiians to travel over the entire group and collect and transcribe, from the lips of the old natives, all the legends, chants, prayers, &c, bearing upon the ancient history, culte, and customs of the people, that they possibly could get hold of. This continued for nearly three years. Sometimes PREFACE. xi their journeys were fortunate, sometimes rather barren of results ; for the old natives who knew these things were becoming fewer and fewer every year, and even they — as is well known to every one that has had any experience in the matter — maintain the greatest reserve on such sub- jects, even to their own countrymen ; and to a foreigner, unless most intimately and favourably known, any such revelation is almost impossible. The labours of my em- ployees, however, were crowned with results exceeding my expectations, and I am now in possession of probably the greatest collection of Hawaiian lore in or out of the Pacific. It took me a long time, during leisure moments from offi- cial duties, to peruse, collate, and arrange these materials, and, though they are filled with much that was worthless for my purpose, yet I found very many pearls of invalu- able price to the antiquarian and historian. To this expost of my own pursuits, I would only add that, during my many journeys from one end of the group to the other, I never omitted an opportunity in my inter- course with the old and intelligent natives to remove a doubt or verify a fact bearing upon the work I had in hand. Among Hawaiian authors and antiquarian literati, to whom I gratefully acknowledge my obligations, are, in the first place, his Majesty King Kalakaua, to whose personal courtesy and extensive erudition in Hawaiian antiquities I am indebted for much valuable information; the late Hon. Lorrin Andrews ; and the late David Malo, whose arii PREFACE. manuscript collections were kindly placed at my disposal by the Honourable Board of Education ; the late Dr. John Eae of Hana, Maui, who, in a series of articles published in the " Polynesian " (Honolulu, 1 862), first called atten- tion to the extreme antiquity of the Polynesian language ; the late Hon. S. M. Kamaeau, with whom I have conferred both often and lengthily ; the late Kev. Mr. Dibble, whose "History of the Sandwich Islands" (1843) contains many gems of antiquarian value ; the late Hon. Naihe of Kohala, Hawaii, and the late S. N. Hakuole. Mr. J. Kepelino has furnished some valuable chants, and the groundwork of the " Kumuhonua " legends, most of winch was con- firmed by the late Mr. Kamakau above referred to. The current communications, from time to time, in the Hawaiian journals on antiquarian subjects, by different authors, have been carefully culled, and are thankfully remembered. Mr. Jules Eemy is personally and kindly remembered since his sejour on the Hawaiian islands, and his Introduction to and edition of the " Moolelo Hawaii " (Paris and Leipzig, 1862), as well as his "Eecits d'un vieux Sauvage, pour servir a l'Histoire ancienne de Hawaii" (1859), have been carefully considered and found of great value. From the Marquesas group, the author is under obliga- tion to Professor W. D. Alexander for access to a collec- tion of ancient legends and chants as told to and written down by the late Mr. T. C. Lawson, for many years a resident of Hivaoa (St. Dominica). From the Society group, and several others of the South- PREFACE. xiii Pacific Islands, Eev. Mr. Ellis's " Polynesian Researches " is replete with much and valuable legendary lore. Mr. Moerenh^ut's " Voyage aux Isles du Grand Ocean " has been carefully referred to ; and in Lieutenant De Bovis's " Etat de la Societe Taitienne a l'arrivee des Europeens," was found a cautious, critical, and reliable author, though on some points we must necessarily differ. From the Tonga group, "Mariner's Voyage" has fur- nished the greatest amount of information. From New Zealand, Dieffenbach's " Travels," and Sir \J George Grey's " Polynesian Mythology " and " Proverbial and Popular Sayings of the Ancestors of the New Zealand Race," not only bring up the common property of the Polynesian race in its legendary lore, but throw an unex- pected light on some very ancient passages of Hawaiian history. From the Samoan (Navigators') group, I regret to say that I have but scant information, collected piecemeal from various sources. What I have, however, coincides strongly with the leading features of the legendary lore of the other groups. From the Fiji group, the " Fiji and the Fijians," by Thomas Williams and James Calvert, has been found to J be good and reliable authority. Various other utterances from Polynesian folklore have been collected and utilised from the best accounts obtainable of voyages undertaken at public expense or prompted by private enterprise ; and among the former, I xiv PREFACE. consider the highest praise is due to the Ethnological and Philological section of the United States' Exploring Expedition under Commodore Wilkes, prepared by Mr. Horatio Hale ; and among the latter, I have found Mr. M. G. L. Domeny de PtiENZi's work " Oceanie" (Paris, 1836), which is a resumi of his own and other voyages in Malaysia and Polynesia, still stand unrivalled for fulness and accuracy. Touching the philological questions arising from a con- sideration of the Polynesian language and its relation to others, I have consulted the great work of William v. Humboldt, "Tiber die Kawi Sprache;" that of Francis Bopp, " Uber die Verwandtschaft der Malayisch-Poly- nesischen Sprachen mit den Indo-Europaeischen;" J. Crawfurd's " Grammar and Dictionary of the Malay Language ; " Adolph Pictet's " Origines Indo-Euro- peennes ;" Professor Max Muller's " Lectures on the Science of Language," and his " Chips from a German Workshop," and such dictionaries as I could procure. Mr. George Smith's " Assyrian Discoveries," and his " Chaldean Account of Genesis ;" Colonel Henry Yule's edition of, and notes to, "The Travels of Marco Polo;" Mr. G. PcAWLiNSON's edition of "Herodotus," and his " Five Great Monarchies ; " and Sir Stamford Eaffles's various essays and writings, have furnished me many valuable points of contact and much light, where other- wise I must have groped my way in darkness. But, while such are my right to speak, and the lights PREFACE. xv which aided me in compiling this work, yet the work itself might possibly never have been published, had not the Hon. H. A. Widemaxn, an acquaintance and friend of thirty years' residence in the Hawaiian group, kindly exerted himself in my behalf to procure the means to defray the cost of publication. And to him and to those who so promptly came forward to aid the enterprise my grateful acknowledgments are herewith tendered. Painfully conscious that my long seclusion from literary labours has cramped my hand, even though the spirit be unflagging as ever, yet with the treasures of legendary lore around me, with my affection for the people with whom I have associated my lot in life for so many years, and with the certainty that each year is fearfully dimi- nishing the chances of ever again procuring an equal col- lection of the Polynesian folklore, I submit this work without hesitation to the favourable regard of the Hawai- ians and the Polynesians, whose past I have endeavoured to rescue from the isolation and oblivion which were fast closing over it, and whose echoes were growing fainter and fainter in the busy hum of a new era and a new civilisation, derided by some, disputed by others, un- heeded by all. To the literati of foreign lands I address myself with that respectful diffidence and cautious reserve which be- come a pioneer in an almost untrodden field. With the data before me, drawn from Polynesian sources, my con- clusions could not well be other than what they are. If xvi PREFACE. at times I have erred in comparative philology, mythology, or history, it will be kindly borne in mind that over forty years of an adventurous and busy life have crept between me and the Alma Mater on the Fyris, where the classics flourished, and where Geyek taught history ; that my own library is very small ; and that there is no public institu- tion worthy of the name within two thousand miles of the Hawaiian group. . In attempting to solve the ethnic riddle of the Polynesian race, I may have stumbled in the path ; but that path alone, I feel convinced, can lead to a solution. ABR. FORNANDER. Lahaina, Hawaiian Islands, March 30, 1877. ORIGIN AND MIGRATIONS OF THE POLYNESIAN RACE. Before I offer ray contribution to Hawaiian history proper, I think it justice to the reader and to the cause of truth to state my view of the Origin and Migrations of the Poly- nesian Family, of which the Hawaiian is only one, though at present the foremost and best known branch. The singular spectacle of a people so widely scattered, yet so homogeneous in its physical characteristics, in its lansuase and customs, has not failed to exercise the minds of many learned and worthy men, both of past and present time, who have written much and differed widely about the origin of the Polynesian family. North and South Americans, Malays, Papuans, Chinese, and Japanese, and even the lost tribes of Israel, have all, at different times, and by different writers, been charged with the paternity of this family, and made responsible for its origin and appearance in the Pacific Ocean. , These writers formed their opinions, undoubtedly, according to the data that were before them ; but those data were too few, often too incorrect and too unconnected as a whole, to warrant the conclusions at which they arrived. A more intimate acquaintance with the Polynesian family itself, with its copious folk-lore, and its reminiscences of the past still vol. I. A 2 THE POLYNESIAN RACE. floating about with dimmer or brighter outlines through its songs and sagas ; a better insight and a truer apprecia- tion of the affinities of its language ; and, lastly, a small amount of renunciation of national vanity on the part of those different writers, might have removed many of the errors and misconceptions in regard to this interesting family of mankind. It would be presumption in me to pretend that I have fully solved so great a problem as the origin and descent of the Polynesian family. Yet I trust that the sequel will show that my conclusions are not only plausible, but ex- tremely probable, and that, only by following the guide which the data now offered afford, can we account in a satisfactory manner for the ethnic, linguistic, and social phenomena connected with that family, for their appear- ance in the Pacific and their distribution within it — from New Zealand to Hawaii, from Easter Island to Eotuma. - That the reader may know at a glance the result to which my investigations in the Polynesian folk-lore, as well as its comparison with that of other peoples, have led me, it may be proper here at the outset to say that I believe that I can show that the Polynesian family can be traced directly as having occupied the Asiatic Archipelago, from Sumatra to Timor, Gilolo, and the Philippines, pre- vious to the occupation of that archipel by the present Malay family ; that traces, though faint and few, lead up through Deccan to the north-west part of India and the shores of the Persian Gulf; that, when other traces here fail, yet the language points farther north, to the Aryan stock in its earlier days, long before the Vedic irruption in India ; and that for long ages the Polynesian family was the recipient of a Cushite civilisation, and to such an extent as almost entirely to obscure its own consciousness of parentage and kindred to the Aryan stock. THE POLYNESIAN RACE. 3 ' "Were every other trace of a people's descent obliterated by time, by neglect, by absorption in some other tribe, race, or tongue, the identity of the nomenclature of its places of abode with that of some other people would still remain an cb priori evidence of the former habitats of the absorbed or forgotten people. • Were every other record and tradi- tion of the descent of the present ruling races in America, North and South, obliterated, the names which they have given to the headlands, rivers, cities, villages, and divisions of land in the country they inhabit, would primarily, and almost always infallibly, indicate their European descent — English, Spanish, Portuguese, Erench, &c, &c. The practice of naming new abodes in memory of old homes is a deep-rooted trait of human nature, and displays itself alike in the barbarous as in the civilised condition of a people. We find it in the wake of all great migrations, from the mosc ancient to the most recent. History is full of illustrations to this effect, to prove the presence of the mother race, through its migrations, in foreign lands where every other vestige, except this one, has been trodden out by time or by succeeding migrations of other peoples and races. • Following the clue which this evidence affords, I hope to be able to show that the Polynesian family formerly occupied, as their places of residence, the Asiatic Archi- pelago, and were at one time in the world's history closely connected by kindred, commerce, or by conquest with lands beyond, in Hindustan, the shores of the Persian Gulf, and even in Southern Arabia. ' From what I have been able to glean of the old Javanese annals, and of their ancient language, the Kawi, I am led to believe that of the two words, which in the present Malay tongue signify an island — "Nusa" and "Pulo" — the former is by far the older, and obtained exclusively before the latter was introduced by the comparatively modern Malays. In those old annals may be found such names for Jawa, or different portions of it, as "Nusa-Kin- 4 THE POLYNESIAN RACE. dang," " Nusa-Hara-Hara," ! and " Nusa- Jawa ; " "Nusa- Kautchana" for Borneo; " Nusa-Antara " for Madura; " Nusa-Kambargan " for Bali, &c, and in several of the eastern parts of the archipelago, such as Ceram, Bulu, Arnboyna, the ancient word " Nusa " still prevails over the modern " Pulo." This word " Nusa," the old ante- Malay designation of an island, reappears under a Polynesian form in various quarters of the Pacific. "We have " Nuka-tea," one of Wallis' group ; also " Nuka-tapu," " Nuka-lofa," the prin- cipal town on Tonga-tabu ; " Nuka-Hiwa " (in some dia- lects contracted to " Nuuhiwa "), one of the Marquesas group ; " Nuku-nono " 2 of the Union group ; " Nuku- fetau " of the De-Peyster's group ; " Nuku-ta-wake " and "Nuku-te-pipi" of the Paumotu Archipel; and some others in the Eastern portion of the Viti group, which lias received so large a portion of its vocables from Polyne- sian sources. But in none of the Polynesian dialects does the Malay word for island, " Pulo," obtain, nor has it left any marks of ever having been adopted. In regard to this word "Nusa," as signifying an island, among the old ante-Malay inhabitants of the Indian Archipelago, and having been brought by them into various parts of the Pacific, it maybe interesting to remark that we meet with the same word, signifying the same thin"-, in the Mediterranean, at a time anterior to the Hel- lenic predominancy, as far back as the Phoenician supre- macy over that sea, and probably older. We thus find that " Ich-nusa " was one of the oldest names of Sardinia ; " Oe-nusce," some islands in the iEgean Sea, off Messene ; " Sire-nusre," islands off Cape Surrentum, Campania, Italy; " Argi-nusa," below Lesbos, off the JEolian coast, and others. Of this word I have found no etymon in the Greek lan- 1 "Hara," or " Hara-Hara," was present case it may be as much a cor- one of the many names of Siwa. ruption of " Nuka-nono " as " Nuu- 2 This may derive from Nuku or hiwa" is a corruption of " Nuka- Nuu, elevated, raised ; but in the hiwa." THE POLYNESIAN RACE. 5 guage, and it is no kin to Nasos or Naesos, the Doric and Ionian names for island. It is justifiable, therefore, to trace it back to the Cushite Arabs, who traded, colonised, and conquered up to and beyond the pillars of Hercules in the West, as well as to the confines of the Pacific in the East. 1 I will now give the names of a number of places within the Polynesian area, which I think may be identified With others situated in the Indian Archipelago and beyond. Were my acquaintance with the older pre-Malay names of the latter greater than it is, I have no doubt the num- ber could have been greatly increased. 2 1. The first island whose name I will thus trace back will be the island of Hawaii, the principal one of the Hawaiian or Sandwich Islands. That name in the principal Polynesian dialects is thus pronounced : — 1 In The Five Great Monarchies of the Ancient Eastern World, by G. Kawlinson, vol. i. p. 112, the author says : " We can scarcely doubt but that, in some way or other, there was a communication of beliefs, a passage in very early times, from the shores of the Persian Gulf to the lands washed by the Mediterranean, of mythological notions and ideas." If so, why not of names of places, capes, islands, &c, also? In Col. Yule's edition of "Marco Pclo," London, 1875, vol. ii. p. 406, it is said that the people of St. Mary's Isle, off the east coast of Madagas- car, in lat. 17 , as a sign of their Arab descent, "call themselves the children of Ibrahim, and the island JSfusi Ibrahim." 2 Mr. Crawfurd's Grammar and Dictionary of the Malay Language, vol. i. p. 282, says : " AVith the ex- ception of a few places in the Philip- pines and Madagascar, no Malay or Javanese names of places are to be found beyond the limits of the Archi- pelago. We look for them in vain in the islands of the Pacific." As Mr. Crawfurd properly distinguishes the Malay and Javanese languages from the pre-Malay and pre-Javanese lan- guages of the Archipelago, he is pro- bably correct ; but the names which I am going to refer to, came without doubt with the earliest Polynesian settlers from the Indian Archipel, and their not being Malay or Java- nese is another proof that the Poly- nesians had departed from the Archi- pel before the Malays and Java- nese had been so long domiciled there as to introduce their own nomenclature of islands and places. The number of old names of places retained and adopted by these in- vaders must have been very great, though perhaps now impossible to define. b 6 THE POLYNESIAN RACE. In Hawaiian, Hawa-ii. „ Society group, .... Ditto. 1 „ Somoan (Navigator's), . . . Sawa-ii. „ South Marquesan and New Zealand, Hawa-iki. „ Earotonga, Awa-iki. „ Tonga (Friendly Island), ( . . Habai. This word is manifestly a compound word : Hawa and ii or iki. Whether the ii or iki is accepted as meaning " small, little," the apparent sense of the New Zealand, Earotongan, and South Marquesan form of the word, or " racjin^, furious with heat," the sense of the word in the North Marquesan, and which has its analogy in the Tahi- tian and Hawaiian, it is evidently an epithet, a distin- guishing mark of that particular " Hawa " from any other. I am led to prefer the North Marquesan sense of the word, in as much as in a chant of that people, referring to the wanderings of their forefathers, and giving a descrip- tion of that special Hawaii on which they once dwelt, it is mentioned as : Tai mamao, uta oa tu U Ii ; " a distant sea (or far off region), away inland stands the volcano " (the furious, the raging)?) This" Hawa," referred to by the Polynesians of all the principal groups as an ancient place of residence, corre- sponds to Jawa, the second of the Sunda islands, which name, however, seems to have been applied principally to the eastern part of that island, the western portion being known from ancient times as " Sonda." In the second century a.d., Ptolomy called the Sunda Isles by the general name of Jaba-dios insulce, or Jaba-din. In the ninth century a.d., two Muslim travellers, re- ported by Eenandot, spoke of the island and its grandeur as the empire of Zaba-ya or Za/pa-ge, evidently an Arabic pronunciation of Jaba or Jawa. 1 An ancient name of the sacred place "Opoa," in the island of Raia- tea, Society group, was " Hawa-ii." THE POLYNESIAN RACE. 7 In the fourteenth century a.d., Marco Polo mentions the island under the name of Claiva, and refers to both Sumatra and Java under that name. 1 Javanese historians indicate that the name of "Java" was given to the island by emigrants from Kling, Kalinga, or Telinga, on the north-east coast of Deccan, who in the first century a.d. invaded and settled on the island, under one Aji Saka, or Tritestra; but it is understood that Java, 1 On this subject, Colonel Yule in his edition of " Marco Polo," 1875, vol. ii. p. 266, remarks : "Polo by no means stands alone in giving the name of Java to the island now called Sumatra. The terms Jaioa, Jaui, were applied by the Arabs to the islands and productions of the Archi- pelago generally, but also specifically to Sumatra. Tims Sumatra is the Jawah both of Abulfeda and of Ibu- Baluta, the latter of whom spent some time on the island. Javaku again is the name applied in the Singhalese chronicles to the Malays in general. Jau and Dawa are the names still applied by the Battaks and the people of Nias respectively to the Malays, showing probably that these were looked on as Javanese by those tribes who did not partake of the civilisation diffused from Java. De Barras says that all the people of Sumatra called themselves by the common name of Jauijs. There is some reason to believe that the appli- cation of the name Java to Sumatra is of very old date. It is by no means impossible that the Jabadin or Yavadvipa of 1'tolomy may be Suma- tra rather than Java." In a note to page 359, same work, Colonel Yule says, " Sonagar or Jonagar is a Tamil corruption of Yavanar, the Yivauas, the name by which the Arabs were known, and is the name most com- monly used in the Tamil country to designate the mixed race descended from Arab colonists." As tames of places and peoples are older than the chronicles which re- cord them, it is well to bear in mind that the Singhalese and the Tamil speaking peoples of Southern India recognised a Jawa to the east of them, the laud of the " Javaku, ''and a Jawa to the west of them, the land of the Yavanar or Jonagar. But the Singhalese chronicles were written after the Malays had occupied the Sunda Isles, became the leading peo- ple there, and appropriated the name of the country to themselves ; while the Tamil appellation of the Arabs must have been infinitely older than the commercial revival during the early Mohammedan times, seeing that Arab intercourse with India was fre- quent and continuous as far back as the times when the Cushite race ruled supreme in Arabia, and their Zaba was yet an emporium of commerce and a cradle of colonisation. The inference, therefore, seems to me almost irresistible that the people, known to the Tamils as "Yavanar," extended their operations to the Sunda Isles, and called that country after their own home, a name which in after ages was borne back to Cey- lon by Malay cruisers and invaders. The Tamil expression "Java-ku," is thoroughly Polynesian. In Hawaiian legends (for the words are obsolete in modern parlance), the suffixes ku and moe to names of places indicate east and west. Thus Kahiki-ku and Ka- hiki-moe, Holani-ku, Holani-moe, sig- nifying " Eastern Kahiki, Western Kahiki," &c, &c. * THE POLYNESIAN RACE. which in Sanskrit means larky, does not grow o'jp. the island. Evidently those emigrants found the ltiame already existing, and with national vanity found a mean- ing for it in their own language, and in process of jtime believed the fiction. The name occurs, however, in cither parts of the Archipelago, as Djaiva, a river on the east coast of Borneo, near \ oti', as m Sawa-it, a place in south-west Borneo ; as Saioa-i, a place on the north coast of Ceram ; and as Awaiya, a village on the south coast of Ceram. For the origin of the name, and its expansion in the Asiatic Archipelago, and thence into Polynesia, we must look beyond the Kalinga invasion, beyond India, to that nation and race whose colonies and commerce pervaded the ancient world in pre-historic times — the Cushite Arabians ; and among them we find as a proto-nom the celebrated Saba, or Zaba, in Southern Arabia, a seat of Cushite empire and commercial emporium " from the earliest times," according to Diodorus Siculus and Aga- tharcides. "We shall see in the sequel how Polynesian legends confirm the opinion of an early intercourse be- tween the Polynesians and the Cushites, and the close adoption by the former of the culture, and many of the beliefs and legends, of the latter. That the influence of this Cushite " Saba," as a name-giver, extended to the nations of the West as well as in the East, may be, inferred from the epithet of Dionysius " Sabazius," and probably also from the names of the town " Saba-te," in Etruria, and of the " Sabini," one of the most ancient indigenous peoples of Italy, and of their god " Sabus," from whom Cato derived their name. 2. The next case of identity will be found in the name of the island of Oahu, one of the Hawaiian group, which evidently refers to Ouahou, a tract of country in Central and South-east Borneo, occupied by Dyak tribes ; and to THE POL YNESIA N RA CE. 9 Ouadju, 1 a State or territory in Central Celebes, occu- pied by a Buguis population. We shall see further on that both the Dyaks and the Buguis, as well as other tribes in that Archipelago, are pre-Malay inhabitants, and kindred to the Pacific Polynesians. 3. We now come to Molokai, another island of the Hawaiian group ; in the ancient songs and sagas called Molokai-a-Hina. This island finds a striking confirma- tion of the derivation of both the names in Morotay, Moroty, Morty (according to different ortho- graphies), one of the Moluccas, north-east of Gilolo. The Moluccas were called by ancient geographers the Sindas, 2 and this name is referred to by Spanish navigators in the sixteenth century, as having obtained before the islands were called collectively the Moluccas. The " Molokai-a- Hina," therefore, or " Molokai-a-Sina," as it would be called in the Samoan dialect of the Polynesian, points with remarkable directness to the derivation of the name, and to the people who named it ; and, allowing for pho- netic variation, we find the same name in Borotai, a place or village among the Sadong hill-Dyaks inland from Sarawak, Borneo. 4. Lchua, Lcfuka, and Levuka, of the Hawaiian, Tonga, and Fiji groups respectively, and Lefu, one of the Loyalty Islands, refer themselves to Labouk, a province on the north side of Borneo. 5. Niihau, one of the Hawaiian group, corresponds to Lifao, a place on the island of Timor. 1 By other writers called and pally on the island of Oahu, may re- written Wajo. See Asiatic Journal, fer to some half-forgotten remem- August 1825. Mr. Crawfurd, in his brance of a former national appella- Grammar and Dictionary of the tion of Wugi. Malay Language, vol. i. p. 85, says : 2 j n fa e Histoire de la Conquete that the Bugis call themselves Wu- des Isles Moluques, par d'Argensola, gist. As we shall observe more than Amsterdam, 1706, vol. iii., it is said one reference to Celebes in this work, that the Moluccas were formerly it is not improbable that the Hawa- called "Sindas" by Ptolomy, espe- iian appellation of Wold, as a title cially Amboyna, Celebes, and Gilolo. for a certain rank of chiefs, priuci- I o THE POL YNESIA N RA CE. 6. Morca, or Eimeo, one of the Society group, west of Tahiti, corre spends to Morca, name of a mountain range in the east of Jawa, and east of Mount Ardjouna. 7. Bora-Bora, one of the Society group, and Pola-pola, name of lands in Ewa, Oahu, in Koolau, Molokai, and in Lahaina, Maui, of the Hawaiian group, refer to Bulo-Bora, an island off the coast of Menangkabau, in Sumatra. 8. Uuahine, one of the Society group, refers to Onjein or Ujein, a town in Malva, on the Nerbudda river, India; in Sanskrit called Ujjayini, also called "Vi- sala." Oujein was also called " Avanti." This town boasted of a most remote antiquity. It is mentioned in the Pu- ranas, also in the Periplus of the Erythrean Sea, and Ptolomy mentions it under the name of " Ozene." ! 9. Vavao or Wawao, one of the Habai group in the Friendly Islands, in other dialects pronounced " Wewau " or " Vevau ; " and Mature- Wawao, or Acteon Island of the Paumotu group, correspond to Babao, an ancient name of the Bay of Coupang, Isle of Timor ; also of a village and district there, and probably the name of the whole island before the Malays conquered it, settled, and named it Timor. 10. Namuha, one of the Tonga islands (Friendly Island), also one of the Fiji group, refers to Namusa, one of the Menguis group in the Moluccas. 1 1 . Kauai, one of the Hawaiian group, refers to 1 In the Asiatic Journal, Febru- Oujein ' means in part nothing more ary 1821, p. 118, B. Tytler, speak- than sovereign of some undefined ing of Vikramaditya, says : " Al- country situated in the west, or to though he is called king of Oujein, the westward of India." Its other this does not by any means prove him name "Avanti,'''' being an equivalent to have been monarch of any portion of Oujein as meaning west, confirms of Hindostan ; because Oujein is uni- the above reasoning. "Avanti,'" from formly made use of by the natives of Sanskrit Ava, away, off, down ; Ava- the upper provinces in the sense of nati, setting of the sun. 4 the west. ' Consequently 'King of THE POL YNESTA N RA CE. i r Tawai, one of the Batchiau islands, west of Gilolo, in the Moluccas ; Kaivai, in south-west of Sumatra. 1 2. Pangai-motu, one of the Tonga islands ; Pango- pango, or Pago-pago, harbour and village on the island of Tutuila, Samoan group, and Pao-pao, a land in Kohala, Hawaii, Hawaiian group. Paopao, or Cook's Harbour, on the island of Eimeo, Society Islands, correspond to Pampanga or Papango, a district in Luzon, Philippine Islands, and to Pagai or Poggi, two islands off the west coast of Sumatra. 13. Puna, name of districts on the islands of Hawaii and Kauai, Hawaiian group ; and Puna-auia, a district in Tahiti, Society group, and Puna-he, district on Hiwaoa, Marquesas group, refer themselves to Puna, the name of a mountain tribe in the interior of Borneo, and to Puna, a district in Deccan, India, south of Bombay, as well as to a river of that name in North ern India, sup- posed by Eemusat to be the Jamuna or Jumna. It recalls, moreover, the old Egyptian name of Pun for Yemen, in South Arabia ; a name older than the twelfth dynasty. 14. Kohala, a district on the island of Hawaii, Hawaiian group; also name of a land in Kumuele, Molokai, Hawaiian group ; also Ta-hara, the south-west point of Matawai Bay in Tahiti ; also Haraikc, one of the Paumotu Islands, cor- respond to Koshala or Kosala, the ancient name of the kingdom of Oude, in India. I mentioned above that Ujjayini or Oujein, in Malva, India, was also formerly called " Vi-sala;" and Arrian mentions an island off the coast of Mekran, the present Beluchistan, called " No-sala," whence the Ichthyo- phagi were said to derive Sohdr, or Soer as Marco Polo calls it, the former capital of Oman, Arabia, shows another singular family likeness to the constituent part of the above names, Hara, Hala, with different prefixes. 1 5. Ka-papala, name of a land in Kau, Hawaii, Hawaiian 1 2 THE POL YNESIAN RA CE. group; also a district called. Papara in Tahiti, Society group, find their reference to Papal, name of a district in Borneo, inhabited by Dyaks. 1 6. Analwla, a land in the district of Koolau, Kauai, Hawaiian group, refers to Ankola, one of the six districts of the Batta country Sumatra. 17. Laie, a land in Koolauloa, Oahu, Hawaiian group, and a land in Kula, Maui, Hawaiian group, recalls Laye, a place in the country of the Keyangs, in Sumatra. 1 8. Mana, a district of Kauai, Hawaiian group, points to Mana, a district near Bencoolen, Sumatra.* 19. Ninole, name of lands in Kau and in Hilo, Hawaii, also on Molokai, Hawaiian group, refers to Ninore, a place in the Bajpootaua, north-west India, in the Bheel country. 20. Kipu, name of lands on Molokai and at Keei, Kona, Hawaii, Hawaiian group, correspond to Tibuu, the south-west point of the Island of Buru, in the Moluccas. 21. Plana; name of numerous districts and lands in the Hawaiian, Marquesas, and Tahitian groups, either singly or in compounds, as " Hana," " Hana-vi," " Hana-manu," " Hana-pepe," " Mala-e-ka Hana," " Olo-hana," and others, refer themselves ultimately, doubtless, to Sana, one of the ancient Cushite emporiums in Southern Arabia. I am not aware of any place in the intermediate Indian Archipelago that has preserved the single form of this name, but there are several places with the compound name, such as Eata-han on the north-east prong of Celebes, Asa-han in the north of Sumatra. 22. Taioa ; name of place and bay in Nukahiwa, Mar- quesas group; and Kaioa, a land in Koolau, Oahu, Hawaiian group ; refer plainly to Kaioa, one of the Molucca Islands, west of Gilolo. 1 Also to Mana-toa, a place in northern part of Timor. THE POL YNESIAN RA CE. 1 3 23. Lawai, a land on Kauai, Hawaiian group, corre- sponds to Laivai, a river in Borneo, province of Succadow, near the centre, inhabited by Dyaks. Besides these references — and their number could be greatly increased were my means of knowing names of localities in the Asiatic Archipel, their present and more ancient names, greater — there are numerous places on all the principal Polynesian groups which preserve the name of water, wai, under various combinations, in their names, such as " Wai-kapu," " Wai-luku," " Wai-aka," " Wai-ehu," "Wai-pa," " Wai-pipiha," " Wai-tiarea," Wai-rao," &c, while the same combination still obtains in many of the islands of the Asiatic Archipelago, as "Wai-gama" in Mysol, "Wai-puti" and "Wai-apo" in Bum, " Wai-kiu" in Timor, &c. The formation of names of places in Poly- nesia with the final compound of hai, as " Ka-wai-hai " on Hawaii, " Tai-o-hai " on Nukahiwa, &c, has its counterpart in such names as " Wa-hai " in Ceram, and " Ama-hai " in Celebes, all showing the ethnic current of the people who named them. I now have to refer to some names which occur several times in the ancient Hawaiian chants, legends, and prayers, as the names of places or islands inhabited and visited by the remote ancestors of those who composed said chants or legends, but whose location is very vaguely defined, and of which, with one exception, I have not been able to find a name-sake among the present Polynesians, though, pos- sibly, many such existed in former times. I note, first — 24. O-lolo-i-mehani, in some legends said to have been the residence of TVakea. The word is composed of the vocative 0, the name proper Lolo, and the epithet Meliani. The latter, so far as I know, has no kindred now existing in the Polynesian dialects ; but it remains entire in the Amblaw dialect, where " Mehani " means " Red ; " and in the Ceram, Ahtiago dialect, "La-hanin" means also " Red." This name refers itself to the island of 1 4 THE POL YNESIAN RA CE. Gi-lolo or Ji-lolo, the principal of the Moluccas. In other Hawaiian legends (that of " Keanini "), refer- ence is made to a group or country called the O-pae-Lolo, literally, " the Lolo group," thus indicating that Lolo is the proper, original name, and " Mehani " but a subsequent and descriptive appellation. This reference of " O-lolo-i- mehani " to " Ji-lolo," receives a farther confirmation from another legendary name of Wakea's residence, which in other legends is said to have been in 25. O-lalo-waia, and which finds its counterpart and original in Lalo-da, a village or district on that same island of " Ir-lolo," on the west coast, opposite Galela. I know not the meaning of the Jilolo suffix-da; the Hawaiian waia means " strong smelling, filthy, dirty." 26. Fatu-hiwa, one of the Marquesas group, refers itself to Baton, a place on the south side of Timor; and to Batou-bhara in the north of Sumatra. 27. Halawa; name of several lands in the Hawaiian group, refers to Salaway, the north-east cape of the island of Jilolo, of the Moluccas ; also to two districts in Beluchistan, men- tioned in Lieutenant Pottinger's travels under the names of Jhalawan and Saraivan. One of the ancient names of the island of Sawaii, Samoan group, was Salafa-ii, which, as well as the Hawaiian " Halawa," indicates its connec- tion with the Jilolo nomenclature of places. 28. In the western part of the Fiji group occur the names of such places as Oto-wawa and Ka-wawa, and in Hawaiian legends reference is made to places named Wiwa and Wawa. The comparison " Wawa " probably refers to Baba, an island south of the Banda group, Indian Archipelago. 29. Kepa, a land on Kauai, Hawaiian group, refers itself to THE P OL YNESIA N RACE. 15 Tcpa, a village on the above mentioned island of Baba. 30. Manoa, a valley on Oahu, Hawaiian group, points to Manoa, islands off the south-east prong of Celebes. 31. Holani-ku and Holani-moe, corresponding to East ami West Holani, also occurring in the chants as Helani ; refers to the island of Ceram, which, according to Craw- furd and others, was formerly called Strang, and is still so called by the natives. Another ancient Indonesian name might also stand sponsor to the Polynesian appellation, viz., Siren-dwip, the ancient name for Ceylon. For the etymology of Ceylon, see " Travels of Marco Polo," edited by Colonel Henry Yule, C.B., vol. ii. p. 296. Under this head of evidence — properly, perhaps, called the geographical evidence — of the previous residence of the Polynesian family in the Asiatic Archipelago, and, probably, lands beyond, the following observations may be worthy of consideration. While the present names for the north and south points of the compass may, or may not, have been adopted by the Polynesians since their irruption and dispersion in the Pacific ; yet^injthe ancient Hawaiian chants and legends, with which 1 am best acquainted] there occur names for north and south which indicate a residence on islands or lands whose configuration and physical surroundings were different from those which they now inhabit. Thus for the north we find such names as Ulu-nui, Uli-uli, Haka- lauai, Mele-mele ; but these are known and handed down by tradition as having been names of lands as well, situated to the north of some former habitat of the people, of which all knowledge and remembrance were lost, save this that they were so situated to the north of them, and were visited at one time by that famous voy- ager, "Kaulu-a-Kalana," whose exploits survive in song and saga. Of Ulu-nui and Mele-mele, I shall have occasion to speak hereafter. Of HaJcalauai I find no further mention in Polynesian legends and manuscripts ; but the Dyaks in 1 6 THE POL YNESIA N RA CE. the province of Succadow, Borneo, on the river Lawa% have a tradition that, having been driven out by war from a country which they called Lawai, they embarked in canoes or praus, and arriving at Borneo, settled there and named the river after their former home. 1 The geogra- phical relation of Borneo to Jawa, or the Sunda Isles generally, gives point and application to the Hawaiian legends, that Lauai or some particular portion of it, dis- tinguished as Hakalauai, was situated to the north of some of their former habitats, and in course of time became synonymous with north. . The other Hawaiian legendary name for the north, just referred to, was Uli-uli. This word throughout the Poly- nesian dialects means " a dark colour, black, blue, dark- green, dusky, sombre," and its application to the northern point of the heavens recalls the observation of M. de Kienzi, when, speaking of the most ancient Javanese division of time into weeks of five days, he says : 2 " Les denominations de noir et de Nord demontrent d'une maniere incontestable que cette subdivision a pris nais- sance dans l'Hindustan, ou le soleil n'est jamais boreal comme a Java et dans les contrees equinoxiales." 3 May not the Hawaiian denomination of the north suggest the same inference ? 1 Asiatic Journal, August 1821, dus invaded and obtained supremacy p. 118. in the island; and the etymology of 2 Oceanic, vol. i. p. 167. those names plainly indicate their 3 The five days of the Ancient Java- Polynesian affinity. Thus the Laggi, nese week were called respectively— blue, is the Polynesian Langi, Lani (1.) Laggi, (2.) Pahing, (3.) Pov, (4.) (.Fiji), Lagi, the sky, the blue ex- Wagi, and (5.) Kliwon ; and repre- panse ; Pahing, red, doubtless refers senting, 1st, the blue and the east ; to the same root as the Polynesian 2d, the red and the south ; 3d, the Hina or Sina, white, bright, in Fiji, yellow and the west ; 4th, the black Siga, the sun, Siga-Sigau, white, and the north ; and 5th, a mixed while in Ceram (Wahai), Mo-sina is colour and the hearth or centre, red ; Port, yellow, finds its relation in This division of the week is said to the Polynesian ; Hawaiian, Poni, a have obtained before the Hindus in- mixture of colours, purple, the early troduced the Brahminieal week of dawn of the morning ; Waggi, black, seven days. This division and those refers to the Polynesian Wake and names, then, belong to the people Wake-wake, the black liquid of tho that inhabited Jawa before the Hin- squid. L THE POL YNESIA N RA CE. 1 7 Liiiong the Hawaiian names for the south occur those ancient ones of Zisso, 1 and Lepo. The former signifies " blue, black, or dark," and hence " the deep water in the ocean ; the latter is synonymous with ' Moana' the deep, open ocean." But there is no land to the north of the present Hawaiian islands, within reach or ken, that could have suggested those other names as epithets or synonyms for the north only: the " Moana-lipo," the dark fathomless ocean, approaches them, not on the south only, but on every side. Nor were these names acquired or adopted while the Hawaiians yet lived in some of the southern groups of the Pacific, for the situation and surroundings of none of these would justify such designations for either north or south. Those names, therefore, refer to a period when the Polynesians occupied the Asiatic Archipelago, and probably lands further west, with the Indian Ocean as their " Moana-lipo," their " dark, unbounded sea," their southern quarter of the heavens, Kuana-lipo, their south ; and with lands of various names all along their northern horizon. The expressions Tonga, Kona, Toa (Sam., Haw., Tali.), • to indicate the quarter of an island or of the wind, between the south and west, and Tokelau, Toerau, Koolau (Sam., Tab., Haw.), to indicate the opposite directions from north to east — expressions universal throughout Polynesia, and but little modified by subsequent local circumstances — point strongly to a former habitat in lands where the regular monsoons prevailed. Etymologic ally " Tonga," "Kona," contracted from "To-anga"or " Ko-ana," signi- fies " the setting," seil. of the sun. " Toke-lau," of which the other forms are merely dialectical variations, signifies " the cold, chilly sea." Mr. Hale, in the Ethnological por- tion of the United States Exploring Expedition under Commodore Wilkes, considers the application of Tonga to the south-western quarter as subsequent to the dispersion 1 One of the Greek names for the south-west wind was Lips, g. Liboa TOT,. T. I* 1 8 THE POL YNESIAN RA CE. of the Polynesians in the Pacific {viol, p. 180). But Mr. Hale, in the very same article, has very lucidly shown that " Tonga " was a term applied to the very first settlement of the Polynesians in the Pacific, on Viti-lewu, signifying " the Western," seil. people, in contradistinction from the Yiti proper, or " Eastern " people. Hence it is reasonable to infer that the Polynesians brought the term with them as an already existing appellation of the western quarter, as much so as they did the other term of " Toke-lau," to designate the eastern quarter. In the Tonga Islands, Hahagi means the northern and eastern side of an island, and Hihifo means the southern and western side. The first is derived from the preposi- tion Hagi, " up, upward ; " the latter from the preposition Hifo, " down, downward." In many of the other Polyne- sian groups the expressions " up " and " down " (Haw., iluna or manae and Halo) are used with reference to the prevailing trade- winds. One is said to " go up " when tra- velling against the wind, and to " go down " when sailing before it. But the relative situation of the Tonga Islands reverses this order of things, and thus precludes the idea of its application being original there. These terms, then, are older than the residence of the Polynesians in the Tonga Islands, and indicate, in connection with other considera- tions, that they originated in some continental abode, or on the southern side of one of the larger Asiatic islands, where, from time immemorial and from constant use, they had hardened into synonyms with the cardinal points, and where to go to the northward was equivalent to " going up," to ascending an upland or mountain, and where the south became identified with the idea of descent, of " going down," towards the sea, from the interior highlands. 1 1 In Malay Utara means north, ventionally called Raro, " down," and evidently derived from the ancient the south Buncia, or " up." Leaving and universal Polynesian word Uta, the Samoan islands, when bound to Uka, meaning "inland " from the sea- New Zealand, they were going to the eliore, up the mountain side. south, or " up "against the south-east Iu New Zealand the north was con- trades ; when going back to the north- THE POLYNESIAN RACE. 19 Among the many traditions and legends which obtained currency among the Polynesian groups, it is not always easy to distinguish which are of an earlier and which of a later date. Almost every group of any note — that does not confessedly derive its inhabitants from some other group, like New Zealand, the Hervey Islands, Eotuma, and some others — has more than one tradition upon the crea- tion of the world and of man, &c. Many of these tradi- tions are not exempt from the vanity of sister nations in the Old World, and make the first inhabitants autochthones of their respective groups. But even those who thus begin their national history almost invariably derive their gods and demi-gods from some far off western country. The Tonga islanders say that they were created, or rather de- scended on their group, from their gods, but that the gods themselves dwelt on an island far to the north-west, called Pulutu. This name strikingly points to the island of " Bum," near Ceram, of the Banda group in the Asiatic Archipelago. In the Fiji group, where so much of early Polynesian tradition as well as language was engrafted on the Papuan stock, the abode of departed spirits is called Mbulu, and the Fijian Elysium is called Mbulotu, thus bringing the Tonga tradition back to the time of the Polynesian sejour on the Fiji group. And the tenacity of the tradition, as well as the universality of a western original home of gods or ancestors, is evidenced by the fact that in all the prin- ward, they were going before the wind, "the front," ke alo, of the land, and or "down." the eastern side was called "the According to Dieffcubach, Tra- back," ke kua. The reason of such vels in New Zealand, the natives designations must be sought in the also call the south by the name of fact of the arrival of the inhabitants Tonga. But as none but the Sa- from the west."' Compare with this moans and Tahitians call the south by the practice of the Aru Islanders in that name, it indicates the source the south-east of Malaysia, who also whence the New Zealanders came, call the eastern parts of their group and thus confirms their own tradi- " the back of the islands." See tions. Voyages of the Dutch brig of war (Iu the Hawaiian group, the western "Doiirga,"by D. H. Kolff, jun., p. wrtion or side of an island was called 175. London: 1840. so THE POLYNESIAN RACE. cipal groups, on the western or north-western side, there are certain places, set off from time immemorial, as the points of departure from which the spirits of the dead plunged in the unknown hereafter to join the society of the gods or to be food for them. The Marquesans are the only people who own to a dis- tinctive national name, and retain a tradition of the road they travelled from their original habitat, until they arrived at the Marquesas Islands. They call themselves te Take, " the Take nation." They say that they were created in a country far far to the west, iao-oa, called TaJec-hee-hee ; and of two different traditions reporting the same fact, one mentions thirteen places of stoppage and sejour during their migration eastward, iuna, ere they arrived at the Marquesas, and the other mentions seventeen places. (in one of their legends J or religious chants, that of the creation of the world, te Pena-pcna, by the god Atea, the then known world extended from Yavau to Hawa-ii, " me Vevau i Hawaii" and after the earth was made, or rather, brought to light, the order was given — " Pu te mutant me Vevau A anu te tai o Hawa-ii ; Pu atu te metani me Hawa-ii A anu te ao o Vevau." (Blow winds from Vavao and cool the sea of Hawaii ; blow back winds from Hawaii, and cool the air or region of Vavao.) And the burden of each stanza or act of crea- tion is — " Vevau me Haiva-ii." Again, in the Marquesan chant of the Deluge, Tai- Toko, it is said that after the flood the ribs of the earth and the mountain ridges of "Hawa-ii" and of "Matahou" 2 rose up, and extended far and near over the sea of Hawaii — 1 Collected by T. C. Lawson, Esq. of Punahou College, Oahu, Hawaiian of Hivaoa, Marquesas Islands, in Islands. MS. kindly furnished the author by 2 New Zealand legends and pro- Professor W. D. Alexander, formerly verbs also refer to a country called THE POL YXESIA N RA CE. 2 1 " Una te tai Haica-ii." I know not the age of these chants ; but from the ab- sence of any mention or allusion in them to the present Marquesas Islands, or the " Ao-Maama " as they are called in other chants, specially that of the Migrations, where they are brought in as the closing scene, so to say, of the epic period of this remarkable branch of the Polynesian family — I infer that they were composed in some other habitat, under physical and geographical conditions entirely different from any that the Polynesian groups afford for the solution of the question which were the Vevau and Hawa-ii to which the chants refer. To seek for them in the Vavau of the Tonga group, and the Sawa-ii or Hawa-ii of the Samoan and Hawaiian groups, would be an arbitrary distortion of the obvious sense of the legend, and incorrect as to the relative position of those islands among them- selves and in regard to the prevailing winds. We must therefore seek the scene of these chants beyond the Pacific, where the winds alternate regularly — in fact, where the monsoons blow ; and by so doing we not only conform to the sense of the chants, but we also find that the relative position of these two points w T as east and west, and not north and south. I have already stated that the large bay of Coupang, on the island of Timor, was formerly called Babao. This bay and surrounding country were, at the time of the first European settlements there, an independent kingdom or state, and it is highly probable that in ancient times, before the Malay element preponderated in the Asiatic archipel- ago, it had given its name to the whole island. But Babao is and would be Vavao or Vevau in any of the Polynesian dialects, and as such is preserved in the names of several places within the Polynesian area of the Pacific. It thus becomes intelligible why Yavao or Timor should have been " Mataaho," destroyed by a flood, Ancestors of the New Zealand Race, " Ce hurianga i Mataako." Vide by Sir George Gray, p. 85. 1'roverbs and Popular Sayings of the 22 THE POLYNESIAN RACE. quoted in the chant as the one terminus of the known world to the people then occupying the Asiatic Archipelago from there to Jawa or Sumatra. To those people, at that time, it was the eastermost land then known, and when subsequently the Malays became dominant, they called it Timor or " the East," plainly indicating thereby that it was also by them at that time considered, as the extreme east. They merely translated the old name, or the idea associated with it, into their own language. I have before stated that I consider the Polynesian word Hawa-ii or Saiva-ii as corresponding to Jawa, whether applied to Jawa proper or to Sumatra, or both, and the frequent allusion made in the chant referred to the " sea of Hawa-ii," te tai o Hawa-ii — the Jawa sea points with sufficient accuracy to those islands as the western terminus of the world as known to those who composed that chant, unless future investigations may enable us to extend that boundary so as to include the Arabian Saba. In this way the expression used, regarding the winds, receives a force and application which, under no other con- struction, it could have received ; and it then applies to the regular monsoons which blow over that part of the world : " Blow winds from Vevao (from the east) and cool the sea of Hawa ; blow back winds from Hawa (from the west), and cool the region or air of Vevao." (The Hawaiian traditions which bear upon the origin of the islands and the derivation of the inhabitants are many and diversified, both in substance and colouring. National or dynastic vanity and priestly speculations have appar- ently at different periods re-cast and re-arranged some old primordial tradition, whose features either retreated to the back-ground of by -gone ages, or were overlaid or altered to suit local necessities, or the pressure of newer ideas Enough, however, remains of that old primal tradition, the groundwork of nearly all the others, to show that the earliest reminiscences of the Hawaiian branch of the Poly- nesian family refer to a far western habitat on some very THE POLYNESIAN RACE. 23 large island or islands, or perhaps continent, as the birth- place of their ancestors. This land was known under many names, hut the most frequently occurring is " Kapa- kapa-ua-a-Kane." It is also called " Hawaii-kua-uli-kai- 00 " (Hawaii with the green back, banks or upland, and the dotted sea). It is said to have been situated in Kahiki-ku, or the lar^e continent to the east of Kalana-i- Hau-ola, or the place where the first of mankind were created, while Kahiki-moe was the name of the large land or continent to the west of this same " Kalana-i-Hau-ola." According to the tradition, there lived many generations after the flood (ke kai-a-Kahinalii), on the east coast of a country situated in or belonging to " Kapakapa-ua-a- Kane," and called Ka Aina Kai ITelemelc-a-Kane, 1 "the land or coast of the yellow or handsome sea," a chief of high renown and purest descent called Hawa-ii-loa, or, also, Kc Kowa-i-Haivaii. This chief was a noted fisher- man and great navigator, and on one of his maritime cruises, by sailing in the direction of the star lao (Jupiter when morning star) and of the Pleiades, he discovered land, arrived at the eastermost of these islands which he called after his own name, and the other islands he called after his children. Delighted with the country, he returned to his native land after his wife and family, and having performed the same eastern voyage, in the direction of the 1 "Among the various traditions volution of the elements." History regarding the maimer in which Jawa of Jawa, by Thomas S. Raffles, chap, and the eastern islands were originally x. , vide Asiatic Journal, DecemKr peopled, and the source whence the 1817, p. 586. The mention of the population proceeded, it has been re- " Laut Mira " or Merah by the Ja- lated that the first inhabitants came in vanese tradition as the direction vessels from the Red Sea (Laut Mira), whence their ancestors came, gives a and that in their passage they coasted singular confirmation and importance along the shores of Hindostan, that to the " Kai Melemele" of the Hawa- peninsula then forming an unbroken iian legend ; but whether it should continent with the land in the Indian refer to the Red Sea or theErythrean Archipelago, from which it is now so may be a question which, under the widely separated, and which, accord- bearing of the Hawaiian legend, 1 ing to the tradition, has since been should be inclined to answer in favour divided into so many distinct islands of the latter. by some convulsions of nature or re- 24 THE POLYNESIAN RACE. morning star and the Pleiades, crossing the ocean which is called by the diverse names of Kai-liolo-o-ka-ia, " the sea where fish do run," Ka Moana-kai-Maokioki-a-Kanc, " the spotted, many-coloured ocean," and also Moana-kai- Popolo, " the blue or dark-green ocean," — he arrived the second time to the Hawaiian Islands, and he and his family and followers were their first human inhabitants. So runs the legend. That this legend embodies the oldest remembered know- ledge of the Hawaiian people regarding the origin of the world, the creation of mankind, the deluge and some prin- cipal events in the national life of that branch of the human family which we now call " the Polynesian," there will, in my opinion, be little room for doubting. The principal facts, and some of the episodes connected with them, are repeated or alluded to, mutatis mutandis, yet in a recognisable shape — and thus corroborated as an heir- loom of the entire Polynesian family, previous to its dis- persion in the Pacific — in the traditions and legends of the principal groups. The universality of the tradition proves its antiquity. This is not, perhaps, the proper place to critically con- sider the historical merits of the Hawaiian legend of Kumulionua, of which the story of Hawaii-loa is but an episode; but, admitting the antiquity of the legend, it shows that, according to the earliest recollections of the Hawaiian people, as handed down by tradition, they sprang from a country lying far to the westward of their present abode, and that, whether the Hawaii to which the legend refers be the Hawaii of the North Pacific, the Sawaii of the Samoan group, or the Jawa of the Asiatic Archipel, they did not come there from the east, north, or south, but from lands and seas in the far distant west. The Hawaiians considered themselves as emigrants, not as autochthones, of the Hawaii of which the legend speaks. But there are three of the Polynesian groups, the THE POLYNESIAN RACE. 25 Hawaiian, the Samoan, and the Tongan, having each an island whose name, with a slight dialectical difference, is precisely the same — Hawa-ii — and each one claiming for itself the honour of having been the first peopled and first named in the Pacific. Yet all concur, however, in point- ing to the far west as the birthplace of their ancestors, or the abode of their gods. 1 In the far west, therefore, beyond the Pacific, we must look for the original "JEawa" or " Hawa-ii," after which they named their new abodes in the various quarters of the Pacific. And here the legend, to which I have already referred, gives another landmark which, in a peculiar manner, points out the direction in which to look for the special and primary "Hawa" which the Polynesians so fondly remembered. The name of that wandering chief, w r ho is said to have discovered the Hawaiian islands and first settled upon them, is not only Hawaii-loa, " the great burning Hawa," but his name is also repeatedly given as he Kowa Hawa-ii, " the straits of the great burning Hawa." If, as I think, there is sufficient ground for identifying the Poly- nesian Hawa-ii with Jawa in the Asiatic Archipel, then this " Kowa-o-Hawa-ii " can be no other than the Straits of Sunda, or a personified remembrance of them. The Polynesian mind had also another mode of express- ing this vague remembrance of a far off home. In many, if not most of the groups, the Moku-Huna or Aina-LTuna- a-Kane, " the hidden, concealed land of Kane," was as much a reality as the existence of Kane himself. This land of plenty and bliss would occasionally loom up in the far off western horizon to the sight of the gifted and faithful. In the Hawaiian traditions its situation was vaguely indicated to be in a north-westerly direction from the group or the particular island of the beholder, and though firmly believed in, yet the belief seldom stimulated 1 According to M. de Bovig, " Etat tions expressly state that the cradle de la Societe Taitienne k l'arrivee of the human race was where the sun des Europceus," the Tahitian tradi- seta. 26 THE POLYNESIAN RACE. to action. But in the Marquesas group numerous ex- peditions have from time to time, up till quite lately, been started in search of this traditional land of mys- tery and bliss, and their course was invariably to the westward. 1 This looking to the west, this longing after the home of their forefathers, was doubtless brought by the Polyne- sians from their earlier Asiatic homes. From India to the Arran Isles, off the coast of Ireland, the belief in some " happy island of the west " was a conspicuous trait of the Indo-European family ; and whether the Polynesians derived it from Aryan or Cushite sources, or whether it developed itself on each particular group as a common expression of regret and desire after that happier land where they had formerly dwelt, its existence as a fact is none the less pertinsnt to the argument rahminical theogony as a compromise and conciliation with his ancient worshippers ; yet that for long ages afterwards the original repugnance was still so great that no true officiating Brahmin would serve at the altars or before the idols of Siwa. How far any distinct remembrance of the Siwa worship may be traced in Polynesian traditions and customs is not easy to determine precisely. The blood-thirsty wife of of the sun and lived mournfully in formerly lived in a zone where the deep obscurity, and when fruits would inequality of day and night was not ripen, how he stopped the sun and greater than in the tropics? regulated its course, so as to make x Rawlinson's Herodotus, vol. ii. day and night equal." Does not that p. 243. legend indicate that the Polynesians THE POL YNESIA N RA CE. 47 Siwa still survives in name and attributes in the Tongan God of War, " Kali-ai-tu-po!' The name itself of Siwa recurs in the Polynesian word Hlwa, primarily " dark- coloured, black or blue ; " secondarily, " sacred," as a sacri- ficial offering. In different dialects the word occurs as Siwa, Hiica, or Hchm, and is applied as an adjective witli derivative meanings, but in all the idea of sacredness underlies and characterises its application. Thus Nuka- Hiwa, one of the Marquesas, undoubtedly meant origin- ally the " dark or sacred island ; " Fatu^Hiwa or Patu- Iliwa, another of the same group, meant the " sacred rock or stone ; " Hiwaoa, still another of the same group, meant the " very sacred or holy." In Hawaiian Pitaa-Hiica means the " black or sacred hog," offered in sacrifices. Hiwa-hhca was an epithet applied to gods and high chiefs. The name of the Siwaite Lingam has unquestion- ably its root and derivation from the same source as the Tongan Linga, the Hawaiian Lina, occurring in such words as Ta-ringa, " the ear," Papa-Una " the cheek," et al. What the Hawaiians called Pohahu-a-Kane, upright stones of from one to six and eight feet in height, the smaller size portable and the larger fixed in the ground, and which formerly served as altars or places of offering at what may be called family worship, probably referred to the Lingam symbolism of the Siwa culte in India, 1 where similar stone pillars, considered as sacred, still abound. 2 . But Siwa, as before observed, was not a Vedic 1 Dieffenbach, in his Travels in Koonbees, living to the eastward of New Zealand, p. 64, says that Phal- the western Ghauts, worship their lie sculptures are common there on principal gods in the form of " parti- tombs, symbolic of the vis generatrix cular unshapen stones." A black of male and female originals. etone is the emblem of Vishnu; a In the Fiji group, also, rude stones grey one of Siwa or Maha-deo. So, resembling milestones are conse- also, stones are consecrated to or em- crated to this or that god, at which blematical of Mussooba, the god of the natives deposit offerings, and be- revenge ; of Vital, the god of demons; fore which they worship. — Fijiandthe of Bal-Bheirow or Bharoo, the bene- Fijians, by Thomas Williams, p. 173. ficent god ; Khun-dooba, the priuci- - Iuihe " Asiatic Journal," Febru- pal household god of the whole Dec- ary 1828, I find that in Deccan and can, is represented at Jejourg by a iu the collectorship of Punah, the Lingam. / 48 THE POL YNESIAN RA CE. god, and his rites were held in abomination by the earlier Vedic Arians. These stone symbols refer, therefore, to a period of pre-Arian occupation of India, and to the Cushite civilisation or race. In the Hawaiian group these stone pillars were sprinkled with water or anointed with cocoa-nut oil, and the upper part frequently covered with a black native tapa or cloth, the colour of garment which priests wore on special occasions, and which was also the cloth in which the dead were wrapped. Singularly enough, the Greeks called Priapus the " black-cloaked," and the Phallus was covered with a black cloth, signifying the nutritive power of night. The veil of Latona, the mother of Apollo and Diana (Sun and Moon), was black. The Hindus of north-west India still worship " Suria," the sun, under the emblem of a black stone. The colour of the Egyptian bulls Apis and Mnevis was black, and in the hieroglyphic representations of acts of consecration or anointing, the officiating priest is painted black, and the recipient of the ceremony is painted red ; this more especially in upper Egypt. Hence the black colour would seem to indicate superior sacredness. It is possible that from these or similar considerations of supe- riority or sacredness arose the Polynesian proverb (in Hawaiian), " he weo ke kanaka, he pano he alii" red is the common man, dark is the chief. 1 ; 1 In " Polynesian Kesearches" the The connection of the hlack colour Rev. Mr. Ellis explains a similar ex- with Siwa's symbols may be found pression in Tahiti, from the fact that in the Hindu legend, according to a dark and bronzed complexion was which, at the churning of the sea of looked upon, among the chiefs, as a milk for the production of Amutham sign of manliness, hardihood, and ex- (the Ambrosia of immortality) Siwa, posure to fatigue and danger, and a the supreme, was appealed to by the pale complexion was considered a other gods to remove the poison sign of effeminacy. The probable vomited in the Ambrosia by the ser- reason and explanation of the pro- pent Vasuke. He complied with verb may be found in the greater their request by drinking up the amount of tattooing with which the poison, but from that time he was bodies of the chiefs were adorned, known by the name of " the azure- As late as the time of Kamehameha necked one," because the colour of I. of Hawaii, his rival Kahekili, the poison remained on his neck as a King of Maui, had one half of his sign of what he had done. — See Crien- body entirely blackened by tattooing, tal Illustrations, by J. lioberts, p. 6. THE POL YNESIAN RA CE. 49 That these sacrificial stones were closely connected with the Phoenician Bcetylia, dedicated to the same purposes, and indicative of a similarity in creeds and symbols, may be shown from the name itself. " Bcetylia," or Bacrvkia, is evidently a composite word, but may not, as some lexicographers indicate, be of Semitic-Hebrew extraction. The thing and its name must be older than the adoption of a Semitic dialect by the Phoenicians. But, as often happens in transition periods, the term " Bcetylia" may be a compromise between the older Cushite and later Semitic languages spoken by the Phoenicians. I consider, there- fore, the word as composed of Batu and II, Illu or El. The latter term is evidently Semitic, and, through all its dialects, signifies God, the God. The former, however, I take to be a Cushite word. It certainly has no Arian connections. But in nearly all the Polynesian congener? we find this word retaining both its primary and deriva- tive sense, both " Stone " and " God " or Lord. In the ancient Madura dialect " Batu " means a stone. At Pulo Nias " Batu " is used as a name for the deity who has charge of the earth, and is called Batu-da-Danau. The expression Battala, used by the pre-Malay Battas in Sumatra, and Bitara of the Bali Islanders, for their deities, may reasonably be referred to the same origin. In the Polynesian dialects proper, we find Batu and Pata-patu, " stone," in New Zealand ; Fatu in Tahiti and Marquesas signifying " Lord," " Master," also " Stone ; " Haku in the Hawaiian means " Lord," " Master," while with the intensi- tive prefix Bo it becomes Boliahu, " a stone." But these stones of religious import, and symbolic of an indwelling God, were common in Arabia, Syria, and Greece. In the " Manual of Ancient History of the East,' by Le norm ant, vol. ii. p. 325, speaking of the religion of Yemen, the author says, that the ancient Arabs in some of their temples or high places, or on top of pyramids like those in Chaldea, worshipped the stars of heaven in pre- ference to idols, though these were also used ; and then vor . 1. d 50 THE POLYNESIAN RACE. adds : " They also in some temples adored, as natural images of the gods, or more accurately as objects in which resided the divine essence — in the same way as in Syro-Phcenician religions — certain stones believed to have fallen from heaven, and similar to the Boetylia of Phoenicia." Common to the Syrian as well as the Phoenician Baal worship, was the setting up of the Aster a " on every high hill and under every green tree," as we are told in i Kings xiv. 23, was the reprobated practice in Judah. In Harwood's " Grecian Antiquities," p. 1 39, I find that in ancient Greece the idol of the temples was but a rude stock or stone cut square, of a black colour, and called Zanis as well as Bactylia; and that Dcedalus is said to have been the first who made two feet to these stone idols. The emblem of Siwa, in Hindu mythology, is the double trident. On a hill called Kaulana-hoa, back of Kalae, island of Molokai, of the Hawaiian group, are a number of large, irregularly- shaped volcanic stones, stand- ing on the brow of the hill. One is shaped like a high- backed chair, and, judging from analogy to others like it in other parts of the group, may have served as a seat for the chief, or his priest, from which to look out over the ocean, or to watch the stars. On the east side of this, and near to, stands another large stone, marked with a double trident (4- 1 ) in two places. Who marked this stone, and what the import of the mark? Tradition is silent ; and in the absence of other marks of similar char- acter, or of corroborative nature, I forbear to offer a con- jecture. But if the name, attributes, and symbols of Siwa at one time were known to and obtained currency among the Polynesians, through their connection with the Cushite race during their residence in India or before, that fact is further strengthened by the Polynesian legend from Eaiatea, Society group, which states that the Deluge was THE POL \ 'N ESI AN RA CE. 5 1 occasioned by the wrath of Bua-Halcu (Rua, the Lord), the great Ocean God of that group. To those acquainted with the phonetic peculiarity — a peculiarity especially archaic — of the Polynesian dialects in the Pacific, and their aver- sion to double and terminal consonants, it is not difficult, nor will it be considered as an etymological crotchet, to connect Eua, in his character of an ocean god, with Muclra, one of the many names of the Hindu pre-Arian Siwa, whose permanent emblem, the trident, indicates his mari- time character, and that at some time, perhaps the earliest period of his culte, he was looked upon as lord of the sea. Whether this name, Budra, Rua, is of Arian or Cushite origin, I know not. If of Arian, it must be of highest antiquity, for in the Celtic-Irish mythology I find that Ruad was the name of the deity presiding over the waters. If of Cushite, it proves the immense extent of the influ- ence of that race, from the borders of the Atlantic to the heart of the Pacific. There is another Hindu god of pre-Vedic origin, and presumably of Cushite extraction, called Yama, or Dhcr- ■marajali, the Hindu Pluto, lord of the infernal regions ("Patala"), who finds his counterpart in attributes and emblems, if not in name, in Polynesia. In India he is represented with a snare in one hand, and a club in the other, looking out for the souls of the dying ; and one of his epithets is " the catcher of the souls of men." In the Hervey group, South Polynesia, Tangaroa, one of the principal gods in that place, is represented with a net in one hand, wherewith to catch the souls of the dying, and a spear in the other, wherewith to kill them. The Hawaiian tradition of Pele, the dreaded goddess of volcanic fires, analogous to the Samoan Fdc, is probably a local adaptation in aftertimes of an older myth, half for- gotten and much distorted. The contest, related in the legend, between "Pele" and Kama-puaa, the eight-eyed monster demi-god, indicates, however, a confused know- ledge of some ancient strife between religious sects, of 52 THE POL YNESIAN RA CE. which the former represented the worshippers of fire, and the latter those with whom water was the principal ele- ment worthy of adoration. Though the contest, accord- ing to the legend, ended in a compromise, and both sides claimed the victory, yet the worship of " Pele" held its ground to the latest times, while that of " Kama-puaa" disappeared, and the monster-god himself is said to have left for foreign lands. In the invocations by the respec- tive parties, addressed to their gods, superior and associate, the symbolism of the legend is clearly brought out. These two prayers, replete with archaic expressions, appeal di- rectly to the gods of fire and of water for assistance in the contest. In that of "Kama-puaa" reference is made to " the storm-clouds of Iku," ha punohu nui a Ihu ; in that of " Pele " reference is made to " the bright gods of night in Wawao, the gods clustering thick round Pele," "Liolio i Wawao na 'Kua o ha po, Ae-ae na ''Kua no Pele." The name itself, Pele, deserves some attention when considering the probable connection of the myth to the modes of thought, of speech, and of creed of those peoples in India or in Chaldea, from whom this myth was derived, alons? with so much other heterogeneous and unacknow- leglged lore among the Polynesians. In the Hawaiian, Pele is a personification of the forces of volcanic fires ; the fire goddess who dwells in the volca- noes. In Samoan, F4e is a personage with nearly similar functions. In Tahitian, Pcre is simply a volcano, the myth seemingly being unknown or forgotten there. But the Hawaiian, Samoan, Tahitian, " Pele," " Pee," " Pere," I consider, etymologically, as nearly allied to the general . Polynesian word Wera, Wela, which in different dialects signifies " fire, conflagration ; to be hot, as from fire or the sun ; to be on fire, to burn," &c. ; and this relation is made more evident from the pre-Malay dialects of the Indian Archipelago, where the My sol Pelah signifies " hot," the THE POLYNESIAN RACE. 53 Sunda Bclem "to burn," the Ceram (Gah) Woleh "the sun." But this word has evidently travelled further than from. Java to Tahiti. It meets us again in the far West, in the Celtic Bel or Belcn, " the sun-god ; " in the old Spartan Beta (BeXa), "the sun;" in the old Cretan A-helios (A(3eXio<>), "the sun;" and in the Phccnician and Syrian Bel, itself an offshoot and adaptation of the Babylonian Bel, the planet Jupiter, and the principal deity worshipped by the later Babylonians. I am not aware that the Polynesian word, or its cog- nates, with radical or derivative sense, occur in the San- skrit or Indo-European languages. It is true that the Hindus call the morning star Velle, and a legend in the Scanda Punana describes him as the leader of the Asuras in their war upon the Devas or gods. 1 But that very cir- cumstance induces me to consider his name as a foreign one, and the legend as having reference to the contention and separation of the Vedic and Zend speaking branches of the Arian race, incident to the reformation of Zoroaster. The Hindu and the Polynesian legends, however, if re- ferring to the same event, seem to be as much distorted the one as the other, with this difference, that the former relates the defeat of " Velle," the latter relates the victory of " Pele," thus showing the different streams on which the legend descended from the battle-field. National appellations are not common among the Poly- nesian tribes. They generally distinguish each other by the name of the island or group to which they belong. Kai Viti, Kai Tonga, " Fiji people," " Tonga people," &c, or they designate themselves as Maori, " indigenous, native," in contradistinction from foreigners, Pwpalangi, or Haolc, or Bapaa. The Marquesans and the Hawaiians, 1 Oriental Illustrations, by James Roberts, p. 411. London, 1835. 54 THE POLYNESIAN RACE. however, form two exceptions. The former have pre- served in their legends, and still retain for themselves, a national appellation which is te Take, " the Take.'"' Accord- ing to the legend they claim " Tane," one of the twelve sons of " Toho," or the original " Take," as their imme- diate progenitor, and the country of Take-hee-liee or AMe- take as their ancient home, the birth-place of their race. It is possible that at one time this national name was common to other tribes as well as the Marquesan, for in the Hervey group, which was confessedly settled by emi- grants from the Samoan and Society groups, we find an island bearing the name of Aitu-Talcc, and a place on the same island called Oni-Take. 1 .. Marquesan legends offer no explanation of whence this name was derived, or how it came to be adopted as a national designation, beyond the fact that " Take " appar- ently was a soubriquet of " Toho," the father of the famous twelve. With other Polynesian tribes the word has become obsolete and meaningless. This name, as well as the legend of " Toho," the first " Take," like so many other Polynesian legends, was pro- bably of Cnshite extraction, innltered by prolonged con- tact into the Polynesian mind, adopted and believed in, and retained as a national distinction long after its origin had been forgotten. I am not aware that this word " Take," under any dialectical variation, is or has been current among the Polynesian congeners in the Asiatic Archipelago, unless the Tagal Taga, " native, indigenous," be a relation or an adaptation of it. But among the 1 Dieffenbach, in his " Travels in tion of the title is correct, it corre- New Zealand," mentions that a title sponds to the Hawaiian Iiumu-honua, or appellation of the chiefs there was the name of the first man. The same "Tahi o te iccnua," and explains it author also mentions, p. 67, a place to mean " the root of the land." As where chiefs go after death, and says the New Zealanders also came from it is called Taki-wana. If the name the Samoan group, it seems as if does not refer to the old national what once was a national appellation, appellation of the Polynesians, or a in course of time became the title of portion of them, I do not see that it a chief. If Dieffenbach's interpreta- has any meaning at all. THE POL YNESIAN RA CE. 5 5 Cusliite peoples or tribes, of pure or mixed descent, who inhabited the Mesopotamian basin, and constituted the confederation of the Rot-u-nu in the time of Thotmes III. of Egypt, circa 1430 B.C., mention is made of the Takce, a people living in the neighbourhood of " Is," on the Euphrates. 1 Such a people or tribe has disappeared from history, and would probably never have been heard of but for the inscriptions on the great tablet of Karnak. Was it a namesake or an ancestor of the " Take," after whom the Polynesians called themselves ? At what period of their existence did they receive it ; and where were they dwelling at the time ? History and tradition are silent. But the fact is almost certain that the Polynesians brought the name with them into the Pacific from their former habitats in the Asiatic Archipel, or beyond. fTn the Hawaiian legend of " Kumuhonua " and his descendants, the Polynesians are distinguished by the appellation of ka poe Menehune, " the Menehune people," said to be descended from " Menehune," son of " Lua Nuu/1 and grandfather of the twelve sons of " Ivinilau-a- niano/^and thus in a measure, though with altered names, it conforms to the Marquesan legend. But this name, as a national appellation, was apparently dropped at a very early period. In Tahiti it became a distinctive name for the third class into which the people were divided, the labouring class, the commoners, the Manahune, and as such remains to this day. In Hawaii it disappeared as a national name so long agoj that subsequent legends have converted it into a term of reproach, representing the Menehune people sometimes as a separate race, some- times as a race of dwarfs, skilful labourers, but artful and cunning. I am inclined to consider the " Menehune " of the legend as a personification of " the people of Mene," for such is the literal signification of the word ; and then Mene alone becomes in reality the national appellation which still * Rawlinson's Herodotus, vol. ii. p. 502. 56 THE POL YNESIAN RA CE. lingers in Hawaiian legends and Tahitian usage. Our knowledge of the legendary lore of the pre-Malay Polyne- sian relations in the Asiatic Archipel, is too limited to enable us to say if any trace or remembrance there exists of either " Take " or " Mene " as national appellations ; but as the latter, like the former, was evidently an older appel- lation than Polynesian residence in the Pacific, we must A ook to the west for some former habitat or connection, (vhich may account for the adoption of the name. Though the Hawaiian legend makes the name-civer of the race the grandfather of the famous twelve, and the Marquesan legend makes him the father; yet the simi- larity of origin of both legends cannot well be doubted ; and that origin, as we shall see plainer as we proceed, was Cushite — Chaldean or Arabian. There, then, we must look for the name as well as the legend. Diodorus Sic. and Agatharcides relate 1 that in south- ern Arabia there lived a people called Mincei, whose capital or chief place was named Karana. When it is borne in mind that in the time of these writers the Him- garites, descendants of the Cushite Arabs, still ruled in that part of Arabia, the similarity, not only of the name of the people, but also of their chief place — which gives a clue to the name of the Hawaiian paradise, Kalana-i-Hau- ola— fbecomes of no small importance in ethnic inquiries. In the " Transactions of the Ethnological Society," London, vol. ii. p. 262, article " Ethnology of Egypt," by E. S. Poole, it is stated that the paintings of the tombs of the kings give four races, of which the first or Egyptian proper is called lien; the second Aamu, representing Asiatics generally but specially Arabs ; the third is called Nalisee, representing negroes; and the fourth Temhu, or the Libyans of north and north-west Africa. It appears, then, that this word Mene, Mincei, and Men, is an old-world national appellation, claimed by Polyne- sians, Arabians, and Egyptians. It may not prove the 1 Vide Antbon's "Classical Dictionary," sub voce. THE POLYNESIAN RACE. 57 ethnic descent of the former from either of the latter ; hut it forcibly indicates a once common possession of legends, traditions, and national existence, that could only have been obtained through an intimate and protracted inter- course, or from the political ascendancy of one over the other. As another coincidence of national appellations, it may he noted that the ancient Chaldean and Assyrian inscrip- tions call Arabia and upper Egypt by the name, as vari- ously read, of Mirukh, Miruki, and Milu-ka, as well as " Kusu " or Cush. In Hawaiian mythology, according to some traditions, Milu was a god who dwelt heneath the sea and ruled over the regions where departed spirits were said to dwell; and the direction in which this realm of " Milu " was situated was in the west. The spirits of those who died on the eastern shore of an island never started for " Milu's " abode in that direction, hut invari- ably crossed over to some one of the points of departure on the western shore. 1 In the far west, then, we must look for the origin or analogy of the myth. But it is a long way from Hawaii to the land of Cush. Let us see if any traces of the myth can he found on the road. The people of Pulo Nias, an island off Sumatra, like the Battas and Dyaks, a pre-Malay remnant of the Polyne- sian race, called the sky or heaven hy the name of Holi- Yaiva, and peopled it with an order of beings whom they 1 A similar legend seems to have Zealand, it is related, p. 67, that existed in Tahiti. According to when a chief dies he goes first to Rev. Mr. Ellis, Tour around Hawaii, Taki-ivana, where his left eye re- p. 205, I find that " the spirits of the mains and becomes a star. Then he Areois and priests of certain idols goes to Reinga and further. Spirits were not eaten by the gods after the sometimes leave the nether world death of their bodies, but went to and come back on earth and commu- Miru, where they lived much in the nicate with the living. For Hawaiian same way as the departed kings and legends to same effect, see p. 83, heroes of Hawaii were supposed to note 1 of present work. Reinga was do, or, joining hands, they formed a a place near the North Cape, New circle with those that had gone Zealand, where the spirits of the before, and danced in one eternal dead collected previous to their final round." departure. In Dieffenbach's Travels in New 53 THE POLYNESIAN RACE. called Baruhi, superior to mortals, gifted with wings, and invisible at their pleasure. And they relate that in olden time a king of these " Baruki," called Luo-mehu-hana, arrived from that " Holi-Yawa, and was the first who taught them arts and civilisation, and also how to speak.l This legend, doubtless, refers to their former intercourse with the early Cushite Arabians, who are said to have dwelt in, and to have come from, the sky, i.e., from beyond the visible horizon, and the name of whose abode they called " Holi-Yawa." Here again the word " Yawa " points to its Cushite source, " Zaba " or " Saba ; " and the name of the king, whose benefits the legend records, forcibly recalls the name of Nuu-(Lolo-i)-Mehani who, according to the Hawaiian legend, was the person that escaped from the Flaod and repeopled the earth. What kindred or connection, if any, except in name, there may be between the " Baruki " of the Pulo Mas legend and the Beloocliees on the Persian Gulf, I am not able to determine. Lieutenant Pottinger, in his " Travels," states that the Beloochees claim descent from Arabia, although they were not like the Arabs, and their language bore affinity to the Persian. Canon Eawlinson, in his Appendix, Book VII. of " Herodotus," vol. iv. p. 181, con- siders the Beloochees as consisting of two ethnic divisions : The " Beloochees " proper, inhabiting the interior and of Arian descent, and the " Brahuis " of the Cushite race, inhabiting the coast-land ; that the latter formerly pene- trated much further inland, but were driven out and con- fined to the sea-board by the former. The same author, in his " Five Great Monarchies," vol. i. p. 50, connects the name of the Beloochees with that of " Belus " or Bel, the great Cushite-Chaldean Divinity. However that may be, there is no misconstruing the import of the Pulo Mas " Yawa." That points unmistakably to the west and the Cushite race as the source of their knowledge and culture. 1 Memoirs of Sir Stamford Baffles, vol. ii. chap. 17. THE POLYNESIAN RACE. 59 The various readings of the cuneiform inscriptions, Milu-ha, 3Eru-ki, MiruJch, would seem to indicate that the word originally was a composite, and, judging from ana- logy, akin to the Egyptian kali, signifying " land," as in Cham-kali, " the land of Cham" (Egypt), San-clia in " San- cha-dwipa," " the land of San," and others. In Cushite parlance " Miruki " or " Miluka " would thus signify " the land of Milu," by which the Chaldeans understood Arabia and Upper Egypt. 1 Of this " Milu " with his realm of deep shadows 2 beneath the sea, in the west, the Polyne- sians retained the recollection in their folk-lore, and the Hawaiians converted him into a god of departed spirits ; while more rationalistic legends assert that he was a very wicked king who was thrust down into Po, the original chaos, for his many misdeeds on earth. Among the Dyaks on the north-west coast of Borneo, the future world, to which the dead are journeying, and where they hope to arrive, is called Sabayan ; and with remarkable fidelity to Chaldean imagery, this heaven of theirs is said to be " seven-storied." Here also the recol- lection of the ancient " Saba," " Sabai," " Sabagi," has sur- vived amid the wreck and debasement of subsequent ages. Creation. The symbolism of "Kane" and his compeers in the Hawaiian group, of " Atea" in the Marquesas group, and (^of " Tane " and " Hina " in the Society group, plainly indi- cate a former religious development among the Polynesian family, when Zabaism, planet-worship and the adoration of the forces of nature, was yet in its simplest form. 1 In Ilawlinson's Herodotus, Essay final ki being the mere affix of loca- i., Appendix, Book I., p. 518, note lity." It thus appears that ki is but 3, I find that in analysing one of the a variant of ka, and that I am correct ancient names of Babylon, " the old in construing Miruki as " the laud or Hamitic name, or, at any rate, one place of Mini." of the old Hamitic names of the city " In Hawaiian Milu, "shaded, of Babylon, must have been read grand, solemn." Din Tir ki, din, 'a city,' and the 60 THE POLYNESIAN RACE. Numerous prayers, invocations, and " meles " (religious poems), interspersed in the legends and handed down by tradition among the Hawaiians, with whom I am best acquainted, abundantly prove this. Throughout the grosser idolatry and the cruel practices springing from it in subsequent ages, these shreds of a purer culte were still preserved, soiled in appearance and obscured in sense by the contact, it may be, yet standing on the traditional records as heirlooms of the past, as witnesses of a better creed, 1 and as specimens of the archaic simplicity of the language, hardly intelligible to the present Hawaiians. This " Kane " creed, such as it has been preserved in Hawaiian traditions, obscured by time and defaced by interpolations, is still a most valuable relic of the mental status, religious notions, and historical recollections of the earlier Polynesians. No other group in Polynesia has preserved it so fully, so far as my inquiries have been able to ascertain ; yet I have met with parts of it on nearly all the groups, though more or less distorted, and in that case I hold that the universality of a legend among so widely scattered tribes proves its antiquity. This " Kane " legend, or rather series of legends, treating of Kane and the crea- tion of Kumuhonua or the first man, of Nuu or Kahinalii in the time of the deluge, of Lua Nuu or Kanelwalani and his descendants, from whom the Hawaiians and Tahitians are said to have sprung \ this legend, with accompany- ing prayers, -invocations/ genealogies, and half-forgotten " meles," has been partially and at different times reduced to writing in the Hawaiian language by David Malo, S. M. Kamakau, Kepelino, and other Hawaiian scholars, who obtained their information from the ancient chiefs and priests who flourished before the introduction of 1 " It could be shown that even practised them did not know the among the South-Sea Islanders, and origin or the meaning, and which -other tribes who have been driven clearly indicated their derivation farthest from the original settle- from an older, a more intelligible, ments of man, there were many reli- but a forgotten faith." — Primeval gious customs of which those who Man, by the Duke of Argyle, p. 196. THE POLYNESIAN RACE. 61 Christianity, some fifty years ago. Among the Hawaiian missionaries who endeavoured to preserve the ancient traditions from oblivion, the greatest credit is due to Eev. Messrs. S. Dibble and Lorin Andrews ; and my own col- lection of Hawaiian folk-lore, gathered from the lips of the old people, is both large and varied. Collating the different narratives thus preserved, I learn that the ancient Hawaiians at one time believed in and worshipped one god, comprising three beings, and respec- tively called Kane, Ku} and Lono, 2 equal in nature, but distinct in attributes ; the first, however, being considered as the superior of the other two, a primus inter pares ; that they formed a triad commonly referred to as Ku-kau-akahi, lit. " Ku stands alone," or " the one established," and were worshipped jointly under the grand and mysterious name 1 Surnamed Ka-Pao," the builder," "the architect." 2 Surnamed Noho i ka wai," dwell- ing on the water." Traces of the worship of these three deities are found throughout the Southern groups, though in later times it had fallen in abeyance in many places, or been entirely superseded by the wor- ship of other gods. The following remnant of what may be called the ancient Hawaiian Liturgy, lecited by the priest and the congregation at the time of the great festivals, has been preserved. It runs : — ■ " The Priest. — O Kane me Ku-Ka- Pao, e oia nei? The Congregation. — Hooia, e oia. The Priest. — O Lono-nui-noho-i-ka- wai, e oia nei? The Congregation. — Hooia, e oia. The Priest. — Ho-eu, kukupu, inana, ku iluna o ka moku, e oia nei? The Congregation. — Hooia, e oia, Hooia, e oia, Hooia, e oia, ke Akua oia. All together. — Kane-Po-Lani, O Lani-makua, me Ku-Ka-Pao i Kiki- laui, me Lono-nui-maka-oaka, he Akua, ke Akua i huila, malamalama paa ka Lani, Ku i ka honua, i ka honua a Kane-kumu-honua, he Akua. Hooia, e oia, Hooia, e oia ; Oia ke Akua oia." It may be translated thus : — " The Priest. — O Kane and Ku, the builder, is it true ? The Congregation. — If is true, it is so. The Priest. — O great Lono, dwell- ing on the water, is it true? The Congregation. — It is true, it is so. The Priest. — Quickened, increasing, moving. Raised up is the continent (island, division). Is it true ? The Congregation. — It is true, it is 30 ; it is true, it is so ; it is true, it is so ; the true god. All togetiicr. — Kane-Po-Lani, O heavenly father, with Ku the builder in the blazing heaven, with great Lono of the flashing eyes, a god, the god of lightning, the fixed light of heaven, standing on the earth, on the earth of Kane-kumu-honua, he is god. It is true, it is so ; it is true, it is so ; he is the true god." 62 THE POLYNESIAN RACE. of Hika ]po loci, while another ancient name was Oi-e, signifying "most excellent, supreme," sometimes used adjectively as Kane-oi-e. These gods existed from eter- nity, from and before chaos, or, as the Hawaiian term expresses it, " rnai ka Po mai " — from the time of night, darkness, chaos. By an act of their will these gods dissi- pated or broke into pieces the existing, surrounding, all- iontaining Po, night or chaos, by which act light entered .nto space. They then created the heavens — three in number — as a place for themselves to dwell in, and the earth to be their footstool, he Jeeehina honua-a-Kane. Next they created the sun, moon, stars, and a host of angels or spirits — i kini akua — to minister to them. Last of all they created man on the model or in the likeness of " Kane." The body of the first man was made of red earth — lepo via or ala-ea — and the spittle of the gods — wai-nao — and his head was made of a whitish clay — -palolo — which was brought from the four ends of the world by " Lono." When the earth-image of " Kane " was ready, the three gods breathed into its nose and called on it to rise, and it became a living being. Afterwards the first woman was created from one of the ribs — lalo pichaka — of the man while asleep, and these two were the progenitors of all mankind. They are called in the chants and in various legends by a large number of different names, but the most common for the man was Kumu-honua, and for the woman Ke Ola kit honua. Such is the general import of the Kumuhonua legend. Another legend of the series, that of Wela-ahi-lani, states that after " Kane " had destroyed the world by fire, 1 1 " Of the Egyptian theory of crea- been subject to several catastrophes, tion some notion may perhaps be 'not to one deluge only, but to obtained from the account given in many ; ' and believed in a variety of Ovid (Met. i. and xxv.), borrowed destructions 'that have been, and from the Pythagoreans ; as of their again will be, the greatest of these belief in the destruction of the earth arising from hre and water. " (Flat, by fire, adopted by the Stoics. (Ovid. Tim. pp. 466, 467). "The idea that Met. i. 256 ; Seneca, Nat. Qucest. iii. the world had successive creations 13 and 28 ; Plut. de Placit. Phil, and destructions is also expressly iv. 7). They even thought it had stated in the Indian Maim." — Hero- THE POL 1 'NESIAN RA CE. 63 on account of the wickedness of the people then living, lie organised it as it now is, and created the first man and the first woman, with the assistance of "Ku" and "Lonu," nearly in the same manner as above narrated. In that lesend the man is called Wela-ahi-lani, and the woman is called Owe} The Marquesas islanders have a legend called Te Vanana net Tanaoa (the prophecy or record of Tanaoa), which relates that in the beginning there was no life, light, or sound in the world, that a boundless night, Po, enveloped everything, over which Tanaoa, which means " darkness," and Mutu-lwi, which means " silence," ruled supreme. In course of time the god Atea, which means " light," evolved himself, sprang from or separated himself from " Tanaoa," made war on him, drove him away and confined him within limits. Light — Atea — having thus been evolved from darkness — Tanaoa — the god Ono or "sound" was evolved from "Atea," and he destroyed or broke . up " Mutuhei." But from the foregoing struggle between " Tanaoa " and "Atea," " Ono " and " Mutuhei," arose Ata- nua or the dawn. "Atea " then took "Atanua " for wife, and from them sprang their first-born Tu-mea. After that "Atea " created the host of inferior deities, fixed or created the heavens and earth, animals, man, &c. Another legend mentions "Atea " and his wife Owa as the progenitors of the Marquesans. 2 Although the Society islanders, like most of the South Polynesian tribes, held that the earth was fished up out of the ocean by Taaroa — Tangaloa — who with them was the source of all things, the father of gods and men, 3 /vet the remnant of a legend, collected by M. de Bovis, 4 for many years a resident on that group, bespeaks an older creed dotus, by G. Eawlinson, vol. ii. p. 1 Vide Appendix, No. I. 250, New York, 1875. Of the future 2 See Appendix, No. III. destruction of the world by fire, the 3 See Appendix, No. II. Scandinavian descendants of the 4 " Etat de la Societe Taitienne a Arian stock preserved a vivid impres- l'arrivee des Europcens," par M. sion" in their mythical poem, the de Bovis, Lieutenant de vaisseau, " Volu-spa." " Eevue Coloniale," annee 1855. 6 4 THE POLYNESIAN RACE. and a clearer conception, and harmonises more with the Hawaiian, Marquesan, and Samoan cosmogonies of more ancient date. The extract of the legend, preserved by M. de Bovis, reads thus : " In the beginning there was nothing but the god Ihoiho ; afterwards there was an expanse of waters which covered the abyss, and the god Tino Taata floated on the surface." It is to be regretted that no more of that interesting legend has been preserved. It has the ring of the true antique, ere the primal myth was shattered into fragments. M. de Bovis translates Ihoiho with " le vide," the empty space, as a better render- ing of sense than " image de soi meme." I know not if the Tahitian word Ihoiho has also the sense of " le vide " — a void, empty space — but it certainly has the meaning of the " manes, ghosts or remains of the dead," and in the legend was probably a trope expressive of a dead and perished world, the wreck of which was covered by water ; and the god Tino Taata, which I think M. de Bovis cor- rectly renders by " the divine type or source of mankind," floated on the waters. It is with some hesitation that I thus correct a writer whose article shows him to have been well-informed, exact, and cautious. But the expression, " le vide," seems to me misleading. Through all the Polynesian cosmogonies, even the wildest and most fanciful, there is a constant under- lying sense of a chaos, wreck, Po, containing all things, and existing previous to the first creative organisation; the chaos and wreck of a previous world, destroyed by fire according to the Hawaiian legend, destroyed by water according to the Samoan legend; a chaos, ruin or night, Po, in which the gods themselves had been involved, and, only in virtue of their divine nature, after continued struggle, extricated themselves and re-organised the world on its present pattern. The generally current tradition on the Society group is, that man was a descendant of Taaroa, through sundry demi-gods: but others, more in accordance with the THE POLYNESIAN RACE. 65 Hawaiian legends, make him a direct creation of Taaroa, who made him out of red clay — araea — and made the first woman from one of his bones, and hence she was called Iwi, lit. " the bone." In the Paumotu group, or, as the natives themselves call their group, the Tuamotu, the ancient tradition relates that " the earth was composed of three separate parts or strata, super-imposed one above the other. Each stratum had its particular heaven. The upper stratum was destined for fortunate souls or spirits ; the middle was inhabited by the living ; and the third was the place where spirits wandered in pain. Many restless spirits, however, sometimes escaped by hiding in the bodies of birds." ! As the Paumotu or Tuamotu group was doubtless originally peopled from the Society group, and also by occasional arrivals from the Marquesas, it is fair to infer that the same or a similar legend obtained in either of those groups in olden time, though forgotten in later ages. The allusion to the three heavens connects it with the Hawaiian legend, vide supra, p. 61. Of Samoan legends bearing on the creation but little has been published. One legend, 2 however, states that in the beginning the earth was covered with water, and the heaven alone was inhabited. Tangaloa, the great God, sent his daughter in the form of a bird called the Kuri, " the snipe," to look for dry land. She found a spot, and, as it was extending, she visited it frequently. At one time she brought down some earth and a creeping plant. The plant grew, decomposed and turned into worms, and the worms turned into men and women. Another legend states that man was formed from the vine of " Kuri " by a god called Ngai? In the Samoas and in Eotumah, the name of the first woman is given as Iwa, thus connecting 1 Annuaire des Etablissements - "Nineteen Years in Polynesia," Francjais de TOceanie, Papeete, 1863, by Eev. Mr. Turner. p. 95. 3 "United States Exploring Expe- dition, Ethnology," by Hor. Hall. VUl.i 1« -ri 66 THE POL \ 'NESIAN RA CE. itself with the Tahitian " Iwi," the Marquesan " Owa," the Hawaiian " Owe." Some of the New Zealand legends 1 ascribe the origin of all things to Rangi and Papa, "heaven and earth," but admit that Po, "night chaos," enveloped everything, heaven • and earth included. Still, under such unfavourable con- ditions, with " Eangi " and " Papa " packed close together, these latter generated six children. The first was Tu-mata- uenga? or the progenitor of man ; next, Tane-mahuta, the father of forests, &c. ; next, Ta-whiri-ma-tea, the father of winds ; next, Rongo-ma-tane, the father of cultivated food ; next, Tanga-roa, the father of fishes and reptiles ; and last Haumia-tikitiki, the father of wild-grown food. The close position of heaven and earth to each other, without interval or separation, greatly annoyed and inconvenienced their offspring, and alter several ineffectual attempts, '" Tane- mahuta," exerting his strength, succeeded in rending "Eangi" and "Papa" asunder, and, pushing the former up into space, let in light on the earth. " Ta-whiri-ma-tea disapproved of the separation of his parents, and followed his father " Eangi " up in the sky, but the other brethren remained with their mother, " Papa," and multiplied and developed indefinitely. War, however, soon arose between 1 "Polynesian Mythology," by Sir the Sanskrit Dhava, man, a word George Gray. which, in its compound form of Vi- - That the name of this god, and daiva — without man, a widow — has his character as the forefather of the survived in the principal Indo-Euro- human family, are older than the pean families of speech. Adolphe arrival of the Polynesians in the Pichet, in Origines Ind-Europ., vol. Pacific, is plainly shown in the fact ii. p. 342, refers this dhava to the that among some of the pre-Malay root DhU, agitare, " aux rapports dialects of the Indian Archipelago, as sexuels dea e"poux," and calls in con- in Saparua, Ceram, Salibabo, and firmation the other derivative DhiXta, Celebes, we fiud the words Tu-mata, " la femme comme agitata, nempe, in To-mata, Tau-mata, and Tau, as ex- concubitu." I believe, however, that pressing the sense of "man" espe- the Polynesian Tu, as the name of cially, and in a general way " man- one of their most powerful gods, and kind." the progenitor of mankind, if it has Professor Fr. Bopp, in his " Ueber any affinity to the Sanskrit, refers die Verwandtschaft der Mai. -Poly- itself rather to the Vedic Tu, "to be nes. Sprachen mit den Ind.-Europ. ," powerful," than to Dhava or any cor- refers this Tau or, contracted, Tu to rupted form of the same. THE POLYNESIAN RACE. 67 the brothers, and finally " Tu-mata-uenga " 1 subdued them all to his will and use, except " Ta-whiri-ma-tea." Another New Zealand legend, according to Nicholas,' 2 records three primitive gods, Maui-Rana-Rangi, the fore- most god, the New Zealand Jupiter ; Tipoko, god of anger and of death ; and Towdki or Tauraki, lord of the elements and god of tempests. The same work intimates that these three created the first man, and afterwards the first woman from one of the man's ribs. According to the foregoing versions of the creation of the world and of man, it is evident that the original Poly- nesian myth contemplated a pre-existing chaos, night, Po, containing within itself, according to the Hawaiian and Tahitian legends, the wreck and debris of a previously perished universe ; and that out of this chaos the first great gods evolved themselves, and then set about organ- ising and creating the world and man as now existing. The Samoan legend relates that in the beginning the world was covered with water: and the Tahitian legend, preserved by M. de Bovis, states in addition that " the god ' Tino Taata ' floated on the surface." This chaos idea among the Polynesian tribes bears a striking relationship to the old Babylonian and Hebrew accounts of the Genesis of the world. Every reader knows 1 This Tu-mata-ueufia, "Tu with is represented with four f;ices, Dour- the red face." This god, or demi-god, ga with eight arms, &c. I have no according to the legend, after having means of knowing whether this four- subdued his brothers, became known fold representation of divine faces or to his posterity by several other limbs was of Arian or Cushite con- names, amongst which was Tu-mata- ception. It does not appear, I believe, tcha-iti, " Tu with the four small in Egyptian mythology, nor generally faces." In Hawaiian mythology, among the Indo-European descend- Kama-puaa, the demi-god opponent ants of the Arian stock. The only of the goddess Pele, is described as analogy I can now remember, is that having eight eyes and eight feet; and of Odin's horse "Sleipner," said to in the chants and legends Maka-walu, have had eight feet. I note the co- " eight-eyed," is a frequent epithet of incidence, however, as bearing upon gods and chiefs. This specialty of the derivation of Polynesian myths four faces or heads, and of corre- and legends. spending limbs, is peculiar to some of 2 L'Univers-Oceanie, par G. L. D. the principal Indian deities. Brahma de Ilicnzi, vol. x. p. 161. 6S THE POLYNESIAN RACE. the second verse of " Genesis : " " And the earth was with- out form and void ; and darkness was upon the face of the deep. And the Spirit of God moved upon the face of the waters." The Tahitian " Tino Taata who floated on the surface " may be the original or the copy of the Hebrew legend. The Babylonian legend, according to Berosus, states that " there was a time in which there existed nothing but darkness and an abyss of waters ;" and accord- ing to the cuneiform inscriptions collected and translated by Mr. George Smith, 1 " Tiamat," the spirit of the sea and of chaos, "was self-existent and eternal, older even than the gods, for the birth or separation of the deities out of this chaos was the first step in the creation of the world." The Chaldean legend refers to a time " When above were not raised the heavens, And below on the earth a plant had not grown up ; The abyss also had not broken open their boundaries; The chaos (or water) Tiamat (the sea) was the producing mother of the whole of them. • ■• • • • • • -• When the gods had not sprung up, any one of them; A plant had not grown, and order did not exist." The Hebrew legend infers that the gods, Eloliim, existed contemporaneously with and apart from the chaos. The Marquesan legend makes the great god of all, Atea, the light, evolve himself from out of darkness, Tanaoa, the ruler of chaos, and from " Atea " sprung the next great god, Ono, or sound. The Hawaiian legend makes the three great gods, Kane, Ku, and Lono, light, stability, and sound, evolve themselves out of chaos, Po. The Babylonian legend makes the two gods Lahmu and Laliamu, the " male and female personifications of .motion and pro- duction, issue from chaos, followed by the gods Sar and Kisar, "representing the upper and lower expanse ;" which four deities, however, appear to be mere abstractions, and were followed by the first actual, personal gods, Anu, Elu 1 The Chaldean Account of Genesis, by George Smith, chap. v. THE POL YXESIAN RA CE. 69 or Bel, and Ilea, representing heaven, earth, and the sea, the Babylonian triad, corresponding to the Hawaiian triad, as the first real creators and organisers of the universe. The New Zealand legend, above referred to, of heaven and earth (" Bangi " and " Papa ") being shut up together, as it were, and enveloped in darkness, " Po ; " of their final separation, and the admission of light upon earth by one of their sons, who pushed the heavens (his father) far upward and away from the earth. This legend, though apparently forgotten or neglected in most of the groups, was still at one time common to the Polynesian family, and, as such, an heirloom brought with them from the West. Traces of it remain in the Samoan, where the legend tells us * that " of old the heavens fell down, and people had to crawl about. The plants grew, and pushed the heavens up a little from the earth. One day a man came along and offered to push the heavens up still higher for a drink of water from a woman's gourd. He did so, and they are now as he left them. The man's name was Tiitiir 2 In the Hawaiian group little remains of this legend but the old saying that at the hill of Kauwiki, not far from the eastern point of the island of Maui, the heaven was nearer to the earth than elsewhere ; in fact, so close that it could be reached by a good strong cast of a spear ; and the lani haaliaa, " the low-lying heavens," is a soubri- quet of the place to the present time. The Dyaks of Borneo — of whom the Bishop of Labuan says, 3 that " like many other uncivilised nations, they have legends of a better and loftier origin, something like the story of Ccelus and Terra" — still retain a legend that, up to the time of the birth of Tana-compta's daughter, the sky had been so 1 "Nineteen Tears in Polynesia," tively modern times, the idols or by Rev. Mr. Turner. images of the gods. 2 Tiitii or Tiki-Tiki or Kii is a 3 Transactions of Ethnological common expression throughout Poly- Society, vol. ii., London. "Wild nesia, signifying spiritual beings, de- Tribes of Borneo," by the Bishop of parted and deified ancestors, protec- Labuan. tors of boundaries, and in compara- 70 THE POL YNESIA N RA CE. near the earth that one could touch it with the hand ; hut she now raised it up and put it permanently on props. The legend further states that "in the beginning there was Solitude and Soutan, who could hear, see, speak, but had no limbs or body. This deity is supposed to have lived on a ball, and after some ages to have made the two great birds called Bullar and Erar, who flew round and round, and made the earth, sky, and rivers. Finding the earth greater than the sky, they collected the soil with their feet, and piled it up into mountains. Having tried to make man out of trees and out of rocks, and not succeeding, they took earth and mixed it with water, and so modelled a man of red clay. When they called to him, he an- swered; when they cut him, red blood came from his veins. This first man was Tana-compta, who afterwards brought to life a female child, who gave birth to off- spring. Then the succession of day and night began, and her progeny became numerous, sailing up and down the river." The juxtaposition of heaven and earth is not expressly stated in the Chaldean and Hebrew legends, though it doubtless is inferred, as both emerged from the primal chaos; but in some of the hymns of the Eig-Veda the idea is plainly held forth that at one time heaven and earth were close together, and the separating them is variously attributed to Varuna, to Vishnu, to Indra, and Soma, who " propped up the sky with supports, and spread out the earth, the mother." The Polynesian legend of the creation of man shows too remarkable an accord with the Hebrew account to be lightly passed over. The former says that " Kane," " Ku," and " Lono," formed man out of the red earth, and breathed into his nose, and he became a living being ; vide supra. The latter says, " the Lord God formed man of the dust of the ground, and breathed into his nostrils the breath of life, and man became a living soul " (Gen. ii. 7). The Polynesian account offers details — the mixing of the THE POLYNESIAN RACE. 71 earth with the spittle of the gods, and the forming the head of man out of white clay, which do not appear in the Hebrew. According to Mr. G. Smith, the Babylonian inscriptions, so far as yet discovered, are defective in that portion which treats of the creation of man ; but it appears that the race of human beings spoken of in line " 18," p. 82, chap. v. of " Chaldean Account of Genesis," is " the zcdmatqaquadi, or dark race, and in various other fragments of these legends they here are called Admi or Adami, which is exactly the name given to the first man in Genesis." And the author further says, that " it has already been pointed out by Sir Henry Eawlinson that the Babylonians recognised two principal races : the Adamu, or dark race, and the Sarku, or light race, probably in the same manner that two races are mentioned in Genesis, the sons of Adam and the sons of God. 1 It appears incidentally from the fragments of inscriptions that it was the race of Adam, or the dark race, which was believed to have fallen ; but there is at present no clue to the position of the other race in their system." The Hebrew word Adam signifies " Bed," and may thus help us to connect the Polynesian first man, made of " red earth," with the Babylonian " dark race ; " but the Polynesian reference to the head of man being made of " white clay," although a myth, may yet have a historical substratum, and indicate a lingering reminiscence of a mixed origin, in which the white element occupied a superior position. In regard to the creation of the first woman, the Polynesian and the Hebrew narra- tives coincide perfectly, even to the very name. Of the ancient Hawaiian chants, referring to the Crea- tion, and which were at once the rythmical, sacerdotal expression of the ancient creed and the crystallised form of the primal tradition — two only, so far as I know, have been preserved and reduced to writing. With the prac- 1 How other writers have handled see Primeval Man, by the Duke of this curious passage in Genesis vi. 2, Argyle (New York, 1874), p. 104, &o. 7 2 THE POL YNESIA N RA CE. tice and observance of the ancient culte, vanished also in great measure the ancient metrical formulas, though the prosaic tradition lingers still in the minds of some of the old people where the hymns have been forgotten. Of these two, which from intrinsic evidence I should judge to be older than the influx into the Hawaiian group of South Pacific emigrants about 800 vears ago, I offer the follow- ing translations. They are but portions of longer chants, and would probably in a few years more have been entirely forgotten ; but, ex pede Herculean, and they are extremely interesting. The first reads : — " O Kane, O Ku-ka-Pao 1 And the great Lono, dwelling on the water, Brought forth are Heaven and Earth, Quickened, increasing, moving, Raised up into Continents. 2 The great Ocean of Kane, The Ocean with dotted seas, 3 The Ocean with the large fishes, And the small fishes, Sharks and Niuhi, 4 Whales, And the large Hihimanu 5 of Kane. 1 Ka-Pao is an epithet of "Ku," 2 Moku, primarily "a division, the second deity, and is probably something cut off ; a land separated best rendered as " the architect, the from other land by water ; an island ; constructor, the builder." Pao, v. a district." signifies "to peck, as birds with the 3 Kai oo, "dotted, variegated sea," bill ; to dig out, as from a rock or seel, with islands or with coral other substance ; to dig down in the patches. Kai is used in speaking of ground, as in making a pit." Pao, s local seas, in contradistinction from signifies "the arch of abridge; the Moana, the great circumambient bridge itself ; a prop ; also a shallow ocean. pit, a place dug out." The original 4 Niuhi, a species of shark of the conception of the epithet ka Pao cor- largest kind. responds to the leading idea of the 6 Hihimrmu, a large, broad, soft Marquesan legend of Creation, te creature of the sea, one and a half or Pepcna, formerly referred to, where two feet in diameter. Both this and the great God " Atea" sets the infe- the foregoing "Niuhi" were forbid- rior deities to work to pick out or dig den, under the Kapu system, to be out the earth from the surrounding eaten by women, under penalty of c haos. death. THE POLYNESIAN RACE. The rows of stars of Kane, The stars in the firmament, The stars that have been fastened up, Fast, fast, on the surface of the heaven of Kane, And the -wandering stars, 1 The tabued stars of Kane, The moving stars of Kane ; Innumerable 2 are the stars ; The large stars, The little stars, The red stars of Kane. infinite space ! The great Moon of Kane,. The great Sun of Kane, Moving, floating, Set moving about in the great space of Kane. The great Earth of Kane, The Earth Kapakapaua 3 of Kane^ The Earth that Kane set in motion Moving are the stars, moving is the moon,. Moving is the great Earth of Kane." Subjoined is the Hawaiian text : — '■ Kane, Ku-ka-Pao Me Lono-nui noho i ka wai, Loaa ka Lani, Ilonua, Ho-eu, kukupu, inana, Ku iluna o ka moku. ka Moana nui a Kane, ka Moana i kai oo, 1 Kahakahakea, lit., "that have literal sense, one of the meanings of been hewn off, chipped off;" hence knpa in Hawaiian is "to gather up " set adrift, wandering." In Mar- in the hands and squeeze, as awa quesan Taha is "to go, march." In dregs," the line may be read "the Tahitian Tahataha is " to be declin- earth squeezed or strained dry by ing as the sun iu the afternoon, to be Kane," and in so far convey an an- wondering as the eye on account of cient conception of the mode of crea- some evil intended." tion analogous to the notion of the 2 I have rendered by "innumer- early Babylonians, who "evidently able," what in the text is expressed considered that the world was drawn by kini, ka lau, ka mano, which together out of the waters, and rested literally means "40,000, 400, 4000." or reposed upon a vast abyss of chao- 3 Kapakapaua, in Hawaiian legends tic ocean which filled the space below designates in a general way the first the world." — Chaldean account of land or country inhabited by man. Genesis, G. Smith, p. 74. If the word, however, is taken in its 7 A THE POLYNESIAN RACE. ka Moana i ka ia nui 1 ka ia iki, I ka memo, i ka niithi, I ke Kohola, I ka ia nui hihimanu a Kane^ na lalani hoku a Kane, na hoku i ka nuu paa, na hoku i kakia ia, 1 paa, i paa i ka Hi lani a Kane, na hoku i kahakahakea, na hoku kapu a Kane na hoku lewa a Kane, kini, o ka lau, o ka mano o ka hofci. ka hoku nui, ka hoku iki, na hoku ula a Kane. He lewa I ka Mahina nui a Kane, ka La nui a Kane, A hoolewa, a lewa, 1 hoolewa ia i ka lewa nui a Kane. ka Honua nui a Kane ka Honua i Kapakapaua l a Kane, ka Honua a Kane i hoolewa. lewa ka hoku, o lewa ka malama, lewa ka Honua nui a Kane." The second chant reads as follows : — " Kane of the great Night, Ku and Lono of the great Night, Hika-po-loa the King. The tabued Night that is set apart, The poisonous Night, The barren, desolate Night, The continual darkness of Midnight. The Night, the reviler. 2 O Kane, Ku-ka-Pao, And great Lono dwelling on the water, 3 1 See note 3, p. 73. f the third great god of the H» a Hai-amu, synon. with Ku-amu- waiians, forcibly recalls to mind the amu, " to revile sacred things, curse Tahitian legend referred to on p. 64, the gods, blaspheme." according to which " the god Tino- 3 " Dwelling on the waters," noho Taata floated on the surface of the * ka wai, as a characteristic or epithet waters." THE POLYNESIAN RACE. 75 Brought forth are Heaven (and) Earth, Quickened, increased, moving, Raised up into Continents. Kane, Lord of Night, Lord the father, Ku-ka-Pao, in the hot heavens, Great Lono with the finishing eyes, Lightning-like lights has the Lord Established in truth, Kane, master-worker ; The Lord Creator of mankind : Start, work, bring forth the Chief Ku-Honua. 1 And Ola-Ku-Honua 2 the woman ; Dwelling together are they two, Dwelling in marriage (is she) with the husband, the brother." The Hawaiian text reads : — ■ " Kane i ka Po-loa, Ku a me Lono i ka I'o-loa, Hika- Po-loa ke Lii. Ka Po-kapu i hoano e, ai-au ka Po, kekaha ka Po anoano, mail kulu ka Po-eleele, Ka Po ke haiamu. Kane, Ku-ka-Pao, Me Lono-nui noho i ka wai, Jjoaa ka Lavi, Honua, Ho-eu, Kukvpu, inana, Ku iluna ka Aloku. Kane-Po-Lani, Lani Makisia, Ku, ka Pao i kikilani, ' Lono-nui maka-oaka, Huila, malamalama, loaa ka Lani, Hooia i oia, Kane kumu-hana. ka Lani hookanaka, Hoi, liana, loaa ke Lii Ku-Honua, ke Ola-ku-Honua ka wahine, Konoho iho no laua, 1 hoi noho i ke Kane kaikunane. 7 It will thus be seen that the order of creation, according to Hawaiian folk-lore, was that after heaven and earth had been separated, and the ocean had been stocked with its 1 One of the many names of the 2 The corresponding name of the first man. first woman. 7 6 THE POLYNESIAN RACE. animals, the stars were created, then the moon, then the sun. In this order the Marquesan legend agrees with the Hawaiian, and both agree exactly with the Babylonian legend of the cuneiform inscriptions. Mr. G. Smith, 1. c. p. 75, says : " The Babylonian account of the creation gives the creation of the moon before that of the sun, in reverse order to that in Genesis ; and evidently the Babylonians considered the moon the principal body, while the book of Genesis makes the sun the greater light. Here it is evident that Genesis is truer to nature than the Chaldean text." Granted that it be truer; but may not that very fact indicate also that the Hebrew text is a later emendation of an older but once common tradition ? On the creation of animals these chants are silent ; but from the prose tradition it may be inferred that the earth,, at the time of its creation or emergence from the watery chaos, was stocked with vegetable and animal life. The- animals specially mentioned in the tradition as having been created by Kane were hogs, Puaa; dogs, Ilio; lizards- or reptiles, Moo. Puaa (in South Polynesian dialects Puaka) seems at one time to have been a general name for beasts and animals, but after the isolation of the Poly- nesians in the Pacific to have been limited to the hog species. Ilio was a general name for dog and his kindred, obsolete in the southern groups. 1 Besides the common domesticated dog, Ilio holo, the tradition speaks of Ilio nui niho oi, " the large dog with sharp teeth," and Ilio 'Hi a Kane, "the royal dog of Kane." The Moo or Moko men- tioned in tradition, reptiles and lizards, were of several kinds : the Moo with large, sharp, glistening teetli ; the talking Moo, moo-olelo ; the creeping Moo, moo kolo ; the roving, wandering Moo, moo-pelo; 2 the watchful Moo, moo- 1 The Hawaiian is the only group 2 The word Pelo is obsolete in the that has preserved the name Ilio for Hawaiian. It occurs, however, in the dog. In the southern groups the Tahitian Fero-pero, " to roam about, word Kuri is generally used, in the to wander." Mii.rfjuesan Nuhe. THE POL YNESIAN RA CE. 7 7 Tcaala ; l the prophesying Moo, moo-haula ; the deadly moo, moo make a fame. The Hawaiian legends frequently speak of moo of extraordinary size living in caverns, amphibious in their nature, and being the terror of the inhabitants. Now, when it is taken into consideration that throughout the Polynesian groups no reptiles are found much bigger than the common house lizard, it is evident that these tales of monster reptiles must have been an heirloom from the time when the people lived in other habitats where such large reptiles abounded. The Hawaiian traditions are eloquent upon the beauty and excellence of the particular land or place of residence of the two first created human beings. It had a number of names of various imports, though the most generally occurring, and said to be the oldest, was Kalana i Hau- ola, " Kalana with the living or life-giving dew." It was situated in a large country or continent, variously called in the legends by the names of Kahiki-honua-kele, Kahiki ku, Kapakapaua-a-Kane, and Mololani. Among the many other names for this primary homestead or paradise, re- tained in the chants and traditions, are Pali-uli, " the blue mountain ; " Aina i ka Kawpo Kane, " the land in or of the heart of Kane ; " Aina wai Akua a Kane, "the land of the divine water of Kane." The tradition says of " Pali- uli " that it was " a sacred, tabued land ; that a man must be righteous to attain it ; that he must prepare him- self exceedingly holy who wishes to attain it; if faulty or sinful, he will not get there : if he looks behind, he will not get there, if he prefers his family, he will not enter in Paliuli." Part of an ancient chant thus de- scribes it : — 1 This word in the present com- Its root is evidently Ala, " to wake, bination is not found in the Hawaiian, be watchful." 7 8 THE POL YNESIA N RA CE. " Pali-uli, hidden land of Kane, Land in Kalana i Hau-ola, In Kahiki-ku, in Kapakapa-ua a Kane, Land with springs of water, fat and moist, Land greatly enjoyed by the god." " Pali-uli, aina huna a Kane, ha aina i Kalana i Hau-ola, 1 Kahikiku, I Kapakapa-ua a Kane, ka aina i kumu, i lali, ka aina ai nui a he Akua." The prohibition referred to above, not to look back when starting on a sacred journey, under penalty of failure, curiously enough recalls to mind the Hebrew legend of Lot, vide Gen. xix. 1 7, &c, and the Greek legend of Orpheus and Eurydice. None of the three legends was in all pro- bability derived from or moulded by either of the others, yet the family likeness between them seems to bespeak a common origin in times anterior to the departure of Abraham from " Ur of the Chaldees," and among a people where superstition had already hardened into maxims and precepts. The Aina wai Akua a Kane, or, as it is more generally called in the legends, Aina wai-ola a Kane, " the livincr water of Kane," is frequently referred to in the Hawaiian folk-lore. According to traditions this spring of life, or living water, was a running stream or overflowing spring, attached to or enclosed in a pond. " It was beautifully transparent and clear. Its banks were splendid. It had three outlets ; one for Ku, one for Kane, and one for Lono and through these outlets the fish entered in the pond. If the fish of the pond were thrown on the ground or on the fire, they did not die ; and if a man had been killed and was afterwards sprinkled over with this water, he did soon come to life again." In the famous legend of " Aukele- nui-a-Iku " the hero visits " Kalana i Hau-ola " and, by the aid of his patron god, obtains water from this fountain of life, wherewith he resuscitated his brothers who had been killed a long time before. THE POLYNESIAN RACE. 79 The notion of a fountain of life is very old, and its origin and its raison d'etre are lost in the gloom of pre-historic times. The earliest allusion to it now known is found in the Izdubar legends of Chaldea, 1 where Ninkigal, the goddess of the regions of the dead, tells her attendant Simtar to pour " the water of life " over Ishtar and restore her to life and health and the company of the gods. I have not the means of ascertaining if the conception of a fountain of life or life-giving waters was common to the Arian family. The Indus and, specially, the Ganges, were sacred rivers with the Hindus, but in how far the sacredness attributed to the latter was local, and posterior to the Arian invasion of India, or older than Vedic times, and transferred from some equally sacred river, lake, or spring in more ancient habitats, I am unable to say, nor am I positive that "Mimer's well" in Scandinavian mytho- logy, where Odin sought wisdom, and pawned his eye to get it, or the " well of Vurdh," where the Nomas sat and watered the tree Ygdrasyl, — though somewhat analogous, are of kindred origin with the Chaldeo-Polynesian foun- tain of life or water of life. The Chaldeans placed their waters of life in the realm of the dead; the Polynesians placed theirs in paradise. Winch is the older form of the conception ? Among other adornments of the Polynesian paradise, the " Kalana i Hau-ola," there grew the Ulu kapu a Kane, "the tabued bread-fruit tree," and the Ohio, Hcmolele, " the sacred apple-tree." The priests of the olden time are said to have held that the tabued fruit of these trees were in some manner connected with the trouble and death of Kumu-honua and Lalo-honua, the first man and woman, and hence in the ancient chants the former was called " Kane Laa-uli, Kumu-uli, Kulu-ipo," the fallen chief, he who fell from, by, or on account of the tree, the mourner, &c, or names of similar import. I have only been able to obtain a portion of a Hawaiian 1 Chaldean Account of Genesis, by G. Smith, pp. 234, 245. 80 THE POLYNESIAN RACE. chant which bears upon the subject of those trees and the fall of man, as connected with the eating of their fruit ; and I am inclined to think it far more ancient than the comments of the priests on the occurrence therein referred to. It may be rendered in English, viz. : — "0 Kane-Laa-'uli, uli, uli, Dead by the feast, feast, feast, Dead by the oath, by the law, law, law, Truly, thus indeed, dead, dead, dead. O ! vanish the stars ! ! vanish the light ! In company with The moon, moon, moon, And cursed be my hand, Cut off be my course ! •E Kane-Laa-'uli, uli, uli, E Kane-Laa-huli, huli, huli, E Kane-Laa-make, make, make, Dead are you, you, you, By Kane thy god, god, god, Dead by the law, law, law, Truly, thus indeed, dead, dead, dead, O Kane-Laa-'uli, uli, uli, O Kane disobeying the gods, gods, gods, Kane (returned) to dust, dust, dust." The text reads : — Kane Laa- uli, uli, uli, 1 make i ahaina, 'ina, 'ina, I make i hoohiki, i kanauni, wax, wai 7 Oia nae, no ka make, make, make. hele ka hoku, hele ka malama, Ka kakai pu ae no, Me ka Mahina, 'hina, 'hina I A Laa kuu lima la ! Kaapahu kuu hele e ! E Kane- Laa- uli, uli, uli, E Kane-Laa-huli, huli, huli, E Kane-Laa-make, make, make, make oe, oe, oe, la Kane Kou Akua, 'kua, 'kua, THE POLYNESIAN RACE. 8 1 / make Jcanawai, 'wai, 'wai, Oia nae no ke make, make, make, Kane-Laa-uli, 'uli, 'uli. Kane aaia, ia, ia, Kane i ka wai-lepo-leoo, lepo." 1 The tradition of the creation above referred to, and the enumeration of various animals of the reptile kind, speaks of the Moopelo as an astute and lying animal, and that he was also known in the ancient chants by the name of Ilioha. In the very chant quoted on pp. 74, 75, after relating the creation of the first man and woman, and giving some eight different names or appellatives whereby they were known, and all referring to their happy and powerful state before the fall, occurs the following allusion to some catastrophe in which the said reptile or " Moo " was concerned, and after which the previous names of the first human pair, expressive of joy and power, were changed to names expressive of misfortune and remorse or grief. The lines of the chant referred to read : " Ka Ilioha kupu-ino ku iluna oka moku } Loaa na Lii Ku-Honua, Polo-Haina ka wahine la e, He mau A lii kapu a Kane. Polo-Haina, ka wahine Ulia-wale, ke kane, Laa'i ka wahine, Laa'-hee-wale ke kane, Zaa'-make ka wahine, Laa'-uli, ke kane, Kanikau, ka wahine Kani-kuo, ke kane, JYoho-ti, lea wahine, Koho-mihi, ke kane, Huki-ku, ka wahine, 1 This is one of the expressions used had died—" he has gone to the moist to designate the moist earth, from earth, or to the muddy water, " scil : which man was made. Hence Ua from which he was made ; or, as we hele i ka wai lepo-lepo became one sa) r , returned to dust, from which he of the many poetical or sacerdotal sprang, phrases to design ate that a person VOL. I. P S2 THE POLYNESIAN RACE. Piliwale, ke kane la e, i Piliwale laua la e." Which may be rendered in English : — " The Ilioha, mischief-maker, stands on the land ; He has caught the chief Ku-Honua, And Polo-Haina, the woman, The Tabu chiefs of Kane," &c, &c Here follow the new names of " Fallen," " Tree-eater," "Tree-upset," "Mourner," "Lamentation," "Kepenting," &c; and it is, moreover, curious to observe that, whereas in enumerating the names of the first pair before their mis- fortune, the chant places the husband's name before that of the wife, in the list of names after the fall the names of the wife precede those of the husband, who becomes, as it were, an intensified echo of the former. The tradition adds, that the first pair lived in " Kalana i Hauola," until they were driven out from there by Ka-aaia-nuhea-nui a Kane, " the large white bird of Kane." This is all that Hawaiian folk-lore, so far as I have been able to collect it, tells us of the forbidden fruit in Paradise, and of the disobedience and fall of the first of mankind. It is but little, but is remarkable for its agreement with the Hebrew legend of the same event, and with the Chaldean allusions thereto, as collected by Mr. George Smith. I know not how far any reference of similar import may have been preserved among the traditions and -chants of the south-westerly groups of Polynesia ; but in one of the sacrificial hymns of the Marquesas, when human a ictims were offered, frequent allusions were made to " the red apples eaten in Vavau," Kciha hua JcaiJcai ia i Vevau, and to " the tabued apples of Atea," te Jeeika tatpu no Atea, as the cause of death, wars, pestilence, famine, and other calamities, only to be averted or atoned for by the sacrifice of human victims. The close connection between the Hawaiian and Marquesan legends indicates a common THE POL YNESIA N RA CE. 83 origin, and that that origin can be no other than that from which the Chaldean and Hebrew legends of sacred trees, disobedience, and fall also sprang. There are still two other Hawaiian legends that also bear upon the subject of the fall of man and the introduc- tion of death in the world : the legend of " Wela-ahi-lani," and that of " Kumu-honua." According to those legends, using one to supplement the other, at the time when the gods created the stars, they also created a multitude of angels or spirits, i Kini Akua, who were not created like man, but were made from the spittle of the gods, i Jcuha ia, to be their servants or messengers. These spirits, or a number of them, disobeyed and revolted because they were denied the Awa, which means that they were not permitted to be worshipped ; Awa being a sacrificial offering and sign of worship. These evil spirits did not prevail, however, but were conquered by Kane and thrust down into uttermost darkness, ilalo-loa i ka Po ; and the chief of these spirits was called by some Kanaloa, by others, Milu, the ruler of " Po," Akua ino, kwpu ino, the evil spirit. 1 The legend further tells that when Kane, Ku, and Lono were creating the first man from the earth, Kanaloa also was present, and, in imitation of Kane, attempted to make another man out of the earth. When his clay model was ready, he 1 Other legends, however, state that was fire. Like the classical Tartarus, the veritable and primordial lord it could be visited by gifted mortals, of the Hawaiian Inferno was called and the spirits of the dead could be Jin una. The Inferno itself bore a brought away from there to the light number of names, such as Po-pau-ole, and life of the upper world. Haw- Po-'kua-kini, Po-kini-kini, Po-papa- aiian legends relate several instances ia-cnva, Po-ia-Milu. "Milu," ac- of such descents and returns to and cording to the legends, was but a from"Po." Mokulehua brought his chief of superior wickedness on earth wife Pueo back from there by the who was thrust down into " l'o," but help of his god "Kanikaniaula." vrho was really both inferior and pos- Maluae brought his son Kaalii from terior to "Manila." This Inferno, there with the help of "Kane "and this"Po" with many names, one of "Kanaloa." Hiku or Iku brought which, remarkably enough, was he up the spirit of the woman Kawelu Po-lua-ahi, "the pit of fire," was and restored her to life ; and several not an entirely dark place. There other instances, was light of some kind, and there 84 THE POL YNESIA N RA CE. called to it to become alive, but no life came to it. Then Kanaloa became very angry, and said to Kane, " I will take your man, and he shall die ; " and so it happened ; and hence the first man got another name, Kumu-uli, which means " a fallen chief," he Lii kaliuli. That the Marquesan Tanaoa and the Hawaiian Kanaloa embody the same original conception of evil, I consider pretty evident. "With the Marquesans the idea is treated in the abstract. With them " Tanaoa " is the primary con- dition of darkness, chaos, confusion, elevated into a divinity battling with Atea, the god of light and order. With the Hawaiians " Kanaloa " is the same idea in the concrete, a personified spirit of evil, the origin of death, the prince of " Po," the Hawaiian chaos, and yet a revolted, disobedient spirit, who was conquered and punished by Kane. 1 In most of the southern groups of Polynesia, though nearly defaced and greatly distorted, the original idea still shines out in the fact that they consider Tangaloa, or, contracted, Taaroa, as the demiurgos of the world, and the father of gods and men, and is there worshipped as the supreme God, taking precedence of Tane, Tu, Oro, Boo, or Lono, and others. 2 That this perversion of the original idea among the southern groups was subsequent in time to the separation of the Hawaiians and Marquesans from the rest of the Polynesian family in the Pacific, I infer from the fact that the introduction and worship of "Kanaloa," as one of the great gods in the Hawaiian group, can only be traced back to the time of the immigration from the southern groups 1 In " Voyage aux Isles du Grand supreme consideration in which this Ocean," par M. Moerenhout, vol. i. p. god was held on Tahiti from the fact, 568, he says, " On ne trouve, nulle as M. de Bovis relates, that no wor- part, de vestiges des deux principes, ship was offered to him, and, with the ni de ces combats entre les tenebres et single exception of the small island of la lumiere,lavieet la mort." If lost Tapuenanu, there was not in all the among Tahitiau legends, the Marque- group a single morae erected in his san and Hawaiian have plainly re- honour. Having created gods and tained those " vestiges." demigods, he was above the considera- 2 Some idea may be formed of the tion of the concerns of mortals. THE POL YNESIAN RA CE. 85 some eight hundred years ago, and that in the more ancient chants he is never mentioned in conjunction with " Kane," " Ku," and " Lono," and that, even in later Hawaiian worship and mythology, he never took precedence of " Kane." This Hawaiian myth of " Kanaloa " as a fallen angel antagonistic to the great gods, and the spirit of evil and death in the world, bears a wonderful relation to the Chal- dean myth of the seven spirits which rebelled against Ann, and spread consternation in heaven and destruction on earth, but were finally conquered by Bel, the son of Hea. See " Chaldean Account of Genesis," by G. Smith, p. 107. The Hebrew legends are more vague and indefinite as to the existence of an evil principle. The serpent of Genesis, the Satan of Job, the Hillel of Isaiah, the dragon of the Revelations — all point, however, to the same underlying idea, that the first cause of sin, death, evil, and calamities was to be found in disobedience and revolt from God. They appear as disconnected scenes of a once grand drama, that in olden times riveted the attention of mankind, and of which, strange to say, the clearest synopsis and the most coherent recollection are, so far, to be found in Poly- nesian traditions. It is probably in vain to inquire with ■whom the legend of an evil spirit and his operations in heaven and on earth had its origin. Notwithstanding the apparent unity of design and remarkable coincidences in many points, yet the differences in detail, colouring, and presentation are too great to suppose the legend borrowed by one from either of the others. It probably descended to the Chaldeans, Polynesians, and Hebrews alike from some source or people anterior to themselves, of whom his- tory now is silent. Gn the events in the world and the generations of man- kind, from the creation of the first man to the time of the 86 THE POLYNESIAN RACE. Flood, Polynesian legends are almost as barren as those of the Chaldees or Hebrews. The latter counted ten genera- tions or dynasties from the first man to, and inclusive of, Noah or Xisuthrus, and in this corresponded with the ten Egyptian reigns of the dynasty of gods, from Ptha to Hor II. The Hebrew account mentions three sons of Adam, of which the first killed the second, and mankind was propagated from the first and third up to the time of the Flood ; but while the line of Seth, from Adam to Noah, counts ten generations, the line of Cain stops short at the eighth generation. The Polynesian legends, as we have seen, both in the Tongas and New Zealand, make reference to the killing of the younger by the older of the sons of the first man. The Hawaiian legend is fuller, and, while referring to the same fratricide, gives a complete genealogy of both the older and youngest branches from the first man to the time of the Flood. I have three different Hawaiian genealogies, going back, with more or less agreement among themselves, to the first created man. One is the genealogy of Kumuhonua, con- nected with the legend frequently referred to. This gives thirteen generations from " Kumuhonua," the first man, to " Nuu " or " Kahinalii," both inclusive, on the line of Laha, the oldest son of " Kumuhonua." It also gives thirteen generations, during the same period, on the line of Ka Pili, the youngest and third son of " Kumuhonua." The second genealogy is called that of Kumu-uli, and was of greatest authority among the highest chiefs down to the latest times, and it was tabu to teach it to common people. This genealogy counts fourteen generations from Huli- honua, the first man, to " Nuu " or " Nana Nuu," both in- clusive, on the line of " Laka," the son of the first man. The third genealogy, which, properly speaking, is that of Facto, the high-priest who came with Pili from Tahiti about twenty-five generations ago, and was a reformer of the Hawaiian priesthood, and among whose descendants it has THE POLYNESIAN RACE. 87 been preserved, counts only twelve generations from " Kuinuhonua " to " aSTuu," on the line of " Ka-Pili," the youngest son. These three genealogies were from ancient times consi- dered as of equal authority and independent of each other, the " Kuinuhonua " and " Paao " genealogies obtaining principally among the priests and chiefs on Hawaii, and the " Kumuuli " genealogy being specially claimed by the chiefs of Kauai and Oahu as their authority ; yet during this early period, from the first man to the Flood, the names of the different generations on the " Kumuhonua " and " Kumuuli" in the line of "Laka" are identical, ex- cept where the latter exceeds the former by one ; and the names on the "Kumuhonua" and "Paao" in the line of " Ka-Pili " are also identical, except where the former exceeds the latter by one. It is fair to infer, therefore, that there was some common bond of union, some sacred deposit from primeval times, which kept the record of these names intact from the interpolations, changes, and varia- tions which in subsequent times more or less affected the number and order of names of generations in post-diluvian periods. But though the Polynesian differ from the Chaldeo- Hebraico-Egyptian account of the number of antediluvian gods, kings, and patriarchs, it coincides w T ith the Hebrew in the number of sons of the first man ; in the murder of the second son by the first ; in the enormous length of days attributed to each generation ; in the translation to heaven of not only one, as the Hebrew text gives it, but of two worthy individuals, whose pious lives had merited such favour ; and, finally, in the very name of the hero of the Flood and that of his wife — Nutc and Lili-Noe — which evidently point to the Hebrew and Arabic Null or Noah. 88 THE POLYNESIAN RACE. In many of the Polynesian groups, there still exist legends of a flood in which the majority of mankind perished, while only a few escaped. Many of these legends are apparently only later editions and corrupted versions of a once common theme, or attempts to localise the cata- strophe and its incidents on this or that group. In the Fiji group, where so much of Polynesian ancient lore was deposited during their sejour on that group, several versions of an ancient tradition of the Flood have been collected by the Eev. Thomas Williams, of which he gives the following synopsis in his work called " Fiji and the Fijians," p. 196. He says of the Fijians— " They speak of a deluge which, according to some of their accounts, was partial, but in others is stated to have been universal. The cause of this great flood was the killing of Turukawa — a favourite bird belonging to Ndengei — by two mischievous lads, the grandsons of the god. These, instead of apologising for their offence, added insolent language to the outrage, and fortifying, with the assistance of their friends, the town in which they lived, defied Ndengei to do his worst. It is said that although the angry god took three months to collect his forces, he was unable to subdue the rebels, and, disbanding his army, re- solved on more efficient revenge. At his command the dark clouds gathered and burst, pouring streams on the devoted earth. Towns, hills, mountains were successively sub- merged ; but the rebels, secure in the superior height of their own dwelling-place, looked on without concern. But when, at last, the terrible surges invaded their fortress, they cried for direction to a god who, according to one account, instructed them to form a float of the fruit of the shaddock ; according to another, sent two canoes for their use ; or, says a third, taught them how to build a canoe, and thus secure their own safety. All agree that the highest places were covered, and the remnant of the human race saved in some kind of vessel, which was at last left by the subsiding waters on Mbenga ; hence the Mbengans THE POLYNESIAN RACE. 89 draw their claim to stand first in Fijian rank. The num- ber saved — eight — exactly accords with the " few " of the Scripture record. By this flood it is said that two tribes of the human family became extinct. One consisted en- tirely of women, and the other were distinguished by the appendage of a tail like that of a dog. The highest point of the island of Koro is associated with the history of the Flood. Its name is Ngginggi-tangithi-Koro, which conveys the idea of a little bird sitting there and lamenting the drowned island. In this bird the Christians recognise Noah's dove on its second flight from the ark. I have heard a native, after listening to the incident as given by Moses, chant " Na qiqi sa tagici Koro ni yali," — " The Qiqui laments over Koro, because it is lost." At Eaiatea, Society group, the legend runs that one day Rua-Haku, the Lord Eua, the Ocean God, was asleep at the bottom of the sea, when a fisherman came along that way with his hook and line. The hook got entangled in the hair of the god, and the fisherman, thinking he had caught a fine fish on his hook, pulled up so vigorously as to bring the god to the surface. Enraged at being thus disturbed in his sleep, the god threatened instant destruc- tion to the unlucky fisherman ; but the latter, having implored the god's pardon, was told to repair to a coral bank or islet called Toa-marama for shelter, while the god vented his displeasure on the rest of the world. The fisher- man did as he was told, and took a friend, a hog, a dog, and a couple of hens with him to the islet. After that the ocean commenced rising, and continued rising until all the land was covered with water and all the people had perished. Then the waters retired, and the fisherman returned to his former home. Other versions of the event exist at Tahiti, but equally distorted. 1 1 M. Moerenhout, in his " Voyage mountains, " sans que, nulle part, il aux Isles de Grand Ocean," vol. i. p. soit question des eaux pluviales." 571, says that the Polynesian legends M. Moerenhout apparently did not represent the ocean as overflowing know the Marquesan and Hawaiian its bed, and rising up to the highest legends, to which I will refer directly. 9 o THE POLYNESIAN RACE. Of the Marquesan legends bearing on this subject, I have only had access to the " Chant of the Deluge," te-tai- toko, in Mr. Lawson's collection. 1 It takes higher ground than the half- remembered and corrupted versions current among the southern groups, and is a remarkable specimen of native poetry, as well as of strict fidelity to the original narrative, so far as that may be ascertained from the Chal- dean and Hebrew accounts. Mr. Lawson has given an English translation, but it is so very literal and rugged, that I prefer to give a prose synopsis of the chant in order to convey its contents to the reader. The chant opens by saying that the Lord Ocean, Fatu- Moana, was going to overflow and pass over the dry earth, but that a respite of seven days was granted. It then speaks of the animals who were to be reserved from the Flood. Then speaks of a house to be built high above the waters ; a house with stories, with chambers, with openings for light, stored with provisions for the preservation of the various animals. The animals then are' fastened with ropes, tied up in couples, and, with one man before and one behind, marched off to this big, deep house of wood. Then the family enter, consisting of four women and four men. The men's names are given " Fetu-moana" apparently the father and master of the family, Fetu-tau-ani, Fetu-amo- amo, and Ia-fetu-tini. A turtle is then sacrificed; the family retires to rest amidst the din, confusion, and crowd- ing of the confined animals. Then the storm bursts over them ; the rain is pouring fearfully, and gloom prevails ; all on earth is displaced and mixed up by the waters. The second part opens with a description of the waters retreating, and mountain summits and ridges reappearing, the grounding of the house, and the command of the Lord Ocean for the dry land to appear. The head of the family, encouraged by the sight, promises to sacrifice to the Lord Ocean seven holy and precious things and seven sucklings. Then a bird, called te teetina o Tanaoa — from 1 See Appendix, No. IV. THE POLYNESIAN RACE. 91 its name apparently of a dark colour — is sent out over the sea of Hawaii, but after a while returns to the vessel. The wind sets in from the north. On a second attempt the same bird alights on the sand of the shore, hut is recalled to the vessel. Then another bird, called te Teetina Moepo, is sent out over the sea of Hawaii. It lands on the dry land, and returns with young shoots or branches it had gathered. The land is now dry, and the great ridges of LTaicaii and of Matalwu are fit to dwell on. In the third part reference is made to the debarkation of men and animals. In the Hawaiian group there are several versions of the Flood. Some indicate the decay and corruption of the original legend in a similar manner to the Fiji and Eaiatea legends above referred to ; but one legend approaches nearly to the Marquesan, though greatly shortened in details as I obtained it. It relates that in the time of Kuu or Nanc^-Nuu, as he is also called, the Flood — Kai-a-Kahinalii — came upon the earth and destroyed all living beings ; that " Nuu," by command of his god, built a large vessel with a house on top of it, which was called and is referred to in the chants as He Waa- Halau-Alii oka Moku, "the royal vessel," in which he and his family, consisting of his wife Lili-Nae, Ms three sons, and their wives, were saved. When the Flood subsided, " Kane," " Ku," and " Lono " entered the " Waa-Halau " of " jSTuu," and told him to go out. He did so, and found himself on the top of Mauna-kea (the highest mountain on the island of Hawaii), and he called a cave there after the name of his wife, and the cave remains there to this day, as the legend says, in testimony of the fact. Other ver- sions of the legend sav that " Nuu " landed and dwelt in Kahiki-honua-kclc, a large and extensive country. I have already given the remainder of the legend on page 44, tell- ing how " Nuu," by mistake, after debarking, offered his 1 Also pronounced Lana, I and n being interchangeable. It means "floatin E ala I e ka ua ! E ka la, E ka ohu-kolo i uka, E ka ohu-kolo i kai ! Kai nuu — kai ee, Kai pipili 5 a Iku ; 1 Ai, "food," when addressed to vian personages. Conjecture in that a god, means the sacrifice, the offer- case might mislead. ing, the gods being supposed to con- 4 Lepo-Akulu, also one of the many sume what was offered them and de- names of the first man. light in it. 5 Pipili, properly means " topsy- 2 Kahuli, one of the many names turvy," "helter-skelter." I have of the first created man. rendered it "boisterous," as appli- 3 I am unable at present to refer cable to the sea or ocean, these names to any known antedilu- THE POLYNESIAN RACE. 93 La I e ua Puni ! huahua kai, ka ale i, ka ale moe, ka ale hakoikoi, 1 kahiki, A hiki a ola, No nei make ia oe la e Lone. E kaukau nou e Lono, E Lono i-ka-Po, 1 E Lono i-ka-Hekili, E Lono i-ka- Uwila, E Lono i-ka- Ua-loko, E Lono i-ka-Oili maka akua nei la, E Lono, e Lono, maka-hia-lele ; A lele oe i ke kai uli, 2 A lele oe i ke kai kona, I kai koolau, 1 One-uli, i One-kea, I mahina, uli, i mahina-kea. Pipipi, Unauna, Alealea ; 3 o hee ; N~aka, Kualakai,* Kama, Opihi-kau-pali, O Kulele poo, helelei ke oho. 5 Waa-Halau-Alii ka moku, Kahi i waiho ai na hua, olelo a Pii, 5 Kama a Poepoe ka wahine i ka ipu-wai! 1 1 I-ka-Po, "from the time of uni- 5 A poetical expression. Oho means versal night, chaos," one of the the hair of the head, and also the ancient titles of the three great gods, leaves of the cocoa-nut tree. The " Kaue," " Ku," and " Lono." The phrase " hina ke oho," or " helelei following lines give the various epi- ke oho," conveys the sense of a severe thets of Lono, and plainly enough storm which tears the leaves from the indicate that in the ancient Hawai- cocoa-nut trees. ian creed he was the god of the 6 I am unable, from any legend, atmosphere and its phenomena. chant, or tradition that has come to - Kai- uli. I have rendered it "the my knowledge, to explain what this Northern Sea," because the following line refers to beyond what the words antithesis of "the Southern Sea" themselves convey. It would seem, and "the Eastern Sea" required it however, from this and the next line, so; and also in view of the designa- that some account of the antedilu- tion of the North mentioned on page vian world was supposed to have been 16. deposited in the Ark, or the Waa- 3 These three names designate dif- Malau, at the time of the Flood, ferent species of shell-fish. 7 Probably contracted, or an ancient 4 Two other kinds of fishes. form of Ipu-icaiau-au, "an epithet 94 THE POLYNESIAN RACE. The above text may be rendered in English as follows : " Here is the food, God, O Kahuli, O Kahela, O the woman sleeping face upward, O Mae-a-Hanuna, O Milikaa. ka Lepo-Ahulu, O Pahu-kini, O Pahu-lau, O Kulana-a-Pahu, O Ola-ka-hua-nui, O Kapapai a Laka, 1 O Manuu, the mischievous, O the great supporter, awaken the world. O wake up. wake up, here is the rain, Here is daylight, Here the mists driving inland, Here the mists driving seaward, The swelling sea, the rising sea, The boisterous sea of Iku. It lias enclosed (us). O the foaming sea, O the rising billows, the falling billows, O the overwhelming billows, In Kahiki. Salvation comes From this death by you, Lono. An altar for you, Lono. O Lono of the night, O Lono of the thunder, O Lono of the lightning, O Lono of tbe beavy rain, O Lono of the terrible, divine face, O Lono. Lono with the restless eyes, Ah, fly to the northern sea, Ab, fly to the southern sea, applied to those who kept the gene- 1 Laka was the oldest son of Kumu- alogies of the chiefs, because they honua, the first man, and the phrase managed to wash the characters of " Ka jmpai a Laka" is a poetical the chiefs so far as their pedigree was expression equivalent to " the de- concerned." — Vide "Andrews's Ha- seendants, the family of Laka." waiian Dictionary," sub voce. THE POLYNESIAN RACE. 95 To the eastern sea, To the dark shore, to the white shore, To the dark moon, to the bright moon. O Pipipi, Unauna, O Alealea ; O glide away ; O Naka, Kualakai, O Kama, O Opihi, sticking to the rock3, O fly beneath the sand, The leaves are falling. O the Waa-Halau-Alii o ka Moku Where were deposited the words of Pii, O Kama-a-Poepoe, the woman of the water-howl." Were the original legend of the Flood to be recon- structed from Polynesian sources alone, it will be seen at a glance how striking its conformity would be to the Hebrew version of said legend, as well as to the Chaldean in parts. Beside the general correspondence in outline, however, there are minor touches of conformity, such as the truce or respite of seven days before the Flood should come ; the fastening of the animals in couples, to be stowed away in the ark ; the sending forth the raven, or dark- coloured bird at the first, instead of the dove, as in the Chaldee account ; the setting in of the north wind to assist in drying up the earth, not mentioned in the Chal- dee ; the reference to " the words of Pii," corresponding to the writings of Xisythrus in the Chaldee account of Berasus, deposited in the city of Sippara, but not referred to in Genesis ; sacrifice offered before entering the ark or vessel, not referred to in Scripture, but probably indicated in the Chaldee account and the Izdubar legends, 1 which make it extremely improbable that the Polynesian legend was borrowed or copied from either the one or the other. And thus, though closely akin, I think it may justly be ranked as another independent version of that great cata- clysm which at some remote period spread desolation over the present Mesopotamian basin. The Chaldean Account of Genesis, by George Smith, p. 266, kc. 96 THE POL YNESIA N RA CE. We know that the story of the Flood spread from " Ur of the Chaldees" to the shores of the Mediterranean, and doubtless different versions of it obtained among the in- tervening nations of Aramians and Hittites, though their accounts of it are now lost to us. It is, therefore, ex- tremely probable that similar versions, variously coloured, found their way southward to Arabia, and eastward to Persia and the early homes of the Arian nations ; the more so, as from the earliest times the ancient Chaldea was designated as the Kiprat-Arbat, " the four nations," or Arba-lisun, " the four tongues," which Mr. G. Eawlin- son, in his " Five Great Monarchies of the Ancient Eastern World," vol. i. p. 55, intimates to have consisted of the Cushite, Turanian, Semitic, and Arian elements, among whom the Cushites preponderated in influence. I have just shown that the Polynesian version of the Flood was probably not derived from either Chaldee or Hebrew originals, at least such as we now have them ; nor, view- ing the state of the Arian legends relating to the Flood, is there the slightest likelihood that it was derived from that quarter. Unfortunately, we have no well-preserved ac- count of the Flood from the Cushite-Arabian quarter ; but I am inclined to consider the Polynesian version as origi- nally representing the early traditions on this subject among the Cushite-pre-Joklanite Arabs, whose sway and whose culture extended over India and the Archipelago, and in so far concurrent in time, equal in authenticity, and equally deserving of consideration, with the Chaldee and Hebrew accounts. Of the Hebrew legend of the Tower of Babel I have found no trace among the traditions of the Polynesian tribes, properly so called, in the East Pacific ; but in the Fiji group, where so many shreds of Polynesian folk-lore THE POLYNESIAN RACE. 97 have been stowed away that have been forgotten else- where, there is a legend that — " Near Na-Savu, Veima Levu, the natives point out the site where, in former ages, men built a vast tower, being eager for astronomic information, and especially anxious to decide the difficult question as to whether the moon was inhabited. To effect their purpose they cast up a high mound, and erected thereon a great building of tim- ber. The tower had already risen far skyward, and the ambitious hopes of its industrious builders seemed near fulfilment, when the lower fastenings suddenly broke asunder, and scattered the workmen over every part of Fiji." 1 Except the genealogical record, Hawaiian traditions give but a small account of the worthies who flourished immediately after the Flood. "We are told that Nuu's three sons were Nalu-Akea, Nalu-Hoohua, and Nalu- Manamana, and that in the tenth generation from Nuu arose one Lua-Nuu, or " the second Nuu," known also in the legends as Kanehoalani, Ku-Pule, and other names. The legend adds that by command of his God he was the first to introduce circumcision to be practised among all his descendants. He left his native home, and moved a long way off until he reached a land called Honua-ilalo, " the southern country," and hence he got the name Lalo- Kona, and his wife was called Honua-Po-ilalo. He was the father of Ku-Nawao by his slave- woman Aim, and of Xalani Menehune by his wife Mec-Hiwa. Another legend says that the God " Kane " ordered Lua-Nuu to go up on a mountain and perform a sacrifice there. Lua-Nuu looked among the mountains of Kahiki-hu, but none of them appeared suitable for the purpose. Then Lua-Nuu in- quired of God where he might find a proper place, and God replied to him : " Go travel to the eastward, and where you find a sharp, peaked hill projecting precipitously into the ocean, that is the hill for the sacrifice." Then Lua- 1 Fiji and the Fijians, by Eev. Mr. Williams, p. 199. 8 G 98 THE POLYNESIAN RACE. Nuu and his son, Kupulu-pulu-a-Nuu, and his servant, Pili-Lua-Nuu, started off in their boat to the eastward ; and in remembrance of this event the Hawaiians called the mountain back of Kualoa, Koolau, Oahu, after one of Lua-Nuu's names — " Kanehoalani " — and the smaller hills in front of it were named after " Kupulu-pulu " and " Pili-Lua-Nuu." By a strange coincidence, Lua-Nuu is the tenth descendant from " Nuu," by both the oldest and youngest of Nuu's sons, " Nalu-Akea " and " Nalu-Mana- mana," of whom the former is represented to have been the progenitor of the Kouaka maoli, the people living on the mainland of Kane — Aina Kumu-paa a Kane — and the latter to have been the progenitor of the white people — he poe keoko maoli. Here again the national conscious- ness of a mixed origin of race reveals itself in a legendary, half-mythical form, similar to the creation legend, where the body of the first man was made of red earth and the head of white clay. This tenth descendant from the hero of the Flood, this " Lua-Nuu " or " Kane-hoa-lani," does again forcibly recall the Hebrew legend of the tenth from Noah — the Abram who travelled into Egypt ; the Abraham of the promise, the originator, by Divine command, of the practice of cir- cumcision ; the father of the slave-woman's child Ishmael, as well as of the legitimate Isaac — the man who in blind obedience would have sacrificed his own child. To make the correspondence more complete, this Lua-Nuu, through his grandson Kini-lau-a-mano, became the ancestor of the twelve children of the latter, and the original founder of the Mene-hune people, from whom this legend makes the Polynesian family descend. Here again the Marquesan legends come to the support of the Hawaiian traditions. They tell us that Toko the Take, the first of that national name, was the grandson of Apana, to whom the introduction of circumcision is ascribed ; that " Toho " was the younger of the twins born to I-aaJca, the son of " Apana ; " and the Marquesan account THE POLYNESIAN RACE. 99 of the children of " Toho " is even more conformable to the Hebrew legend than the Hawaiian account of the children of " Kini-lau-a-niano," inasmuch as the latter enumerates only the twelve sons, whereas the former mentions not only the twelve sons, but also the thirteenth child, the daughter. After this period of " Kini-lau-a-mano's " or " Toho's " twelve sons, the similarity between Polynesian and Hebrew- Chaldean legends becomes very scarce and not well defined. There are references to Kana-loa and Kane-Apua, his brother, a pair of prophets or high-priests, who overthrew the power of King Waha-nui, and who walked about the world causing water to flow from rocks, and similar won- derful exploits, which, in the light of the foregoing resem- blances, may bear reference to Moses and Aaron. 1 In the 1 S. M. Kamakau, the Hawaiian archaeologist, to whom we are indebted for the preservation of so many of the ancient legends, relates the fol- lowing as part of an ancient legend : — " Keahi-Wahanui, king of the coun- try called Honua-i-lalo, oppressed the Lahui-Menehune (the Menehune people). Their God, Kane, seutKane- Apua and Kanaloa, the elder brother, to bring the people away, and take them to the land which Kane had given them, and which was called Ka aina Momona-a-Kanc, or, with an- other name, Ka One Laueita a Kane, and also Ka Ama i Ka Haupo a Kane. The people were then told to observe the four Ku days in the be- ginning of the month as Kapu Houno (sacred or holy days), in remembrance of this event, because they thus ' arose ' — Ku — to depart from that land. Their offerings on the occasion were swine and goats. The narrator of the legend explains that formerly there were goats without horns, called Malailua, on the slopes of the Mauna- loa mountain in Hawaii, and that they were found there up to the time of Kamehameha I. The legend fur- ther relates that, after leaving the land of Honua-i-lalo, the people came to the Kai-ula-a-Kane (the Red Sea of Kane) ; that they were pursued by Ke Alii Wahanui ; that Kane-Apua and Kanaloa prayed to Lono, and that they then waded across the sea, travelled through desert lands, and finally reached the Aina-Lauena-a- Kane." On first receiving this legend, I was inclined to doubt its genuineness, and to consider it as a paraphrase or adaptation of the Biblical account by some semi-civilised or semi-Christian- ised Hawaiian, after the discovery of this group by Captain Cook. But a larger and better acquaintance with Hawaiian folklore has shown me that, though the details of the le- gend, as narrated by the Christian and civilised Kamakau, may possibly in some degree, and unconsciously to him, perhaps, have received a Biblical colouring, yet the main facts of the legend, with the identical names of places and persons, are referred to more or less distinctly in other legends of undoubted antiquity. I am com- pelled, therefore, to class the legend among the other Chaldoeo-Arabico- Hebraic mementoes which the Poly- ioo THE POLYNESIAN RACE. famous Hawaiian legend of Hiaka-i-ka-poli-o-Pele, it is said that when " Hiaka " went to the island of Kauai to recover and restore to life the body of Lohiau, the lover of her sister " Pele," she arrived at the foot of the Kalalau mountain shortly before sunset, and being told by her friends at Haena that there would not be daylight suffi- cient to climb the Pali (mountain) and get the body out of the cave in which it was hidden, she prayed to her gods to keep the sun stationary, i ha muli o ITea, " over the brook, pool, or estuary of Hea," until she had accomplished her object. The prayer was heard, the sun stood still, the mountain was climbed, the guardians of the cave van- quished, and the body recovered. What previous legend, if any, had been culled and applied to furnish this episode of the Hiaka legend, I cannot say. If the Hebrew legend of Joshua, or a Cushite version, gave rise to it, it only brings down the community of legends a little later in time. And so would the allusion in the legend of Naula- a-Maihea, the Oahu prophet who left Oahu for Kauai, was upset in his canoe, was swallowed by a whale, and thrown up alive on the beach of Wailua, Kanai, unless the legend of Jonah, with which it corresponds in a measure, as well as the previous legend of Joshua and the sun, were Hebrew anachronisms, compiled and adapted in later times from long antecedent materials, of which the Polynesian refer- ences are but broken and distorted echoes, bits of legendary mosaic, displaced from their original surroundings, and made to fit with later associations. In the legend of " Aukele-nui-a-Iku," previously re- ferred to, especially in the opening parts of it — being the youngest but one of twelve children ; being the pet and liesians brought with them from their viz., that no other gods are referred ancient homesteads in the West. And to than to those primordial ones of it is possible that the legend was pre- Hawaiian theogony — Kane, Ku,, and served in after times by the priest- Lono, the latter of whom is clearly hood, as offering a rational explana- recognised as the god of the atmo- tion of the institution of the Kapu- sphere, of air and of water, the days of Ku. Another feature attests Lononoho-i-ka-wai of the Creatio the genuine antiquity of the legend, chants. THE POLYNESIAN RACE. 101 favourite of his father, and consequently bitterly hated by his brothers ; being thrown into a pit by them, and left to die; being delivered from the pit by his next eldest brother; his adventures and successes in foreign lands; and, finally, his journey to the place where " the water of life," Ka wai ola-loa a Kane, was kept ; his obtaining it, and therewith resuscitating his brothers, who had been killed and drowned some years before — there is a most striking resemblance to the Hebrew legend of Joseph and his brethren. In the beginning of the Hawaiian legend the scene is laid in that ancient, well-remembered, and often-quoted home of the Polynesians, the Kua-i- helani of song and saga, situated in Kahiki-lcu, and bor- dering on the ocean. This is about all that I have been able to collect of the most striking coincidences and similarity between Polyne- sian and Hebrew-Chaldean legends. The correspondence seems almost too great to be ascribed to the accidental development of the same train of ideas in the minds of people apparently so widely separated in time and dis- tance as the Hawaiians and the Israelites. Two hypo- theses may with some plausibility be suggested to account for this remarkable resemblance of folklore. One is, that during the time of the Spanish galleon trade, in the six- teenth and seventeenth centuries A.c., between the Spanish main and Manilla, some shipwrecked people (Spaniards or Portuguese) — of whose arrival at these Hawaiian islands there can now be no doubt — had obtained sufficient influ- ence to introduce these scraps of Bible history into the legendary lore of this people. The other hypothesis is, that at some remote period a body of the scattered Israelites had arrived either at these islands direct, or in Malaysia, before the exodus of the Polynesian family, and thus imparted a knowledge of their doctrines, of the early life of their ancestors, and of some of their peculiar cus- toms, and that, having been absorbed by the people among whom they found a refuge, this is all that remains to at- 102 THE POLYNESIAN RACE. test their presence — intellectual tombstones over a lost and forgotten race, yet sufficient, after twenty-six cen- turies of silence, to solve in some measure the ethnic puzzle of " the lost tribes of Israel." On the first hypothesis I would remark, that if the ship- wrecked foreigners were educated men, or only possessed of such scriptural knowledge as was then imparted to the commonalty of laymen, it is morally impossible to con- ceive that a Spaniard of the sixteenth century should con- fine his instruction to some of the leading events of the Old Testament, and be totally silent upon the Christian dispensation, and the cruciolatry, mariolatry, and hagi- olatry of that day. And it is equally impossible to con- ceive that the Hawaiian listeners, chiefs, priests, or com- moners, should have retained and incorporated so much of the former in their own folklore, and yet have utterly rejected or forgotten every item bearing upon the latter. Besides, even were this possible, it fails to account for the fact that so many of these legends, wholly or in part, more or less distorted, are to be met with among the southern groups of Polynesia, where the Spaniards never went, with the exception of Mendana's voyage, when, however, no men were left at the Marquesas to propagate bits of Bible his- tory from either Old or New Testament. In regard to the other hypothesis, the Israelitish impact on Polynesian folklore, it is certainly more plausible, and cannot be so curtly disposed of as the Spanish theory. The assertors may not be able to prove that any portion of the "lost tribes of Israel" ever arrived at the Hawaiian group, or at the Asiatic Archipelago, during the occupa- tion and before the exodus of the Polynesian family ; but they may boldly stand on the established facts, and logically infer the cause from the results, and thus throw the onus upon us to show that the results do not warrant the infer- ence, and to account in some other way for their appearance. I have already shown, in the foregoing pages, in what the Polynesian and Hebrew and Chaldean legends differ THE POLYNESIAN RACE. 103 from and agree with each other, and have ventured my opinion that, so far from being copies the one from the other, they are, in fact, independent and original versions of a once common legend, or series of legends, held alike by Cushite, Semite, Turanian, and Arian, up to a certain time, when the divergences of national life and other causes brought other subjects, peculiar to each, promi- nently in the foreground ; and that as these divergences hardened into system and creed, that grand old heirloom of a common past became overlaid and coloured by the particular social and religious atmosphere through which it has passed up to the surface of the present time. But beside this general reason for refusing to adopt the Israelitish theory, that the Polynesian legends were intro- duced by fugitive or emigrant Hebrews from the sub- verted kingdoms of Israel or Judah, there is the more special reason to be added, that on those grand episodes of Hebrew national life — the Egyptian bondage, the exodus, 1 the law on Sinai, the conquest of Canaan, the organisation and splendour of Solomon's empire, his temple and his wisdom, become proverbial among the nations of the East subsequent to his time, and the dismemberment and fall of that empire — that on all these, to a Hebrew momen- tous and never-to-be-forgotten subjects, the Polynesian legends are absolutely silent. Had the former legends, whose correspondence I have noted, been derived from Hebrew sources, it is perfectly inconceivable that the latter legends should not also have been imparted, and some traces of them remained in the Polynesian folklore. Among the customs, usages, rites of worship, modes of thought, prevalent among the Polynesians, much may be found that still further indicates their connection, ethnic and social, with the races who met and mingled at the early dawn of history in the Mesopotamian basin. 1 See note to p. 99, touching the legend of Kane-Apua and his brother Kaiialua. 104 THE POL YNESIAN RA CE. Circumcision. The custom of circumcising every male child was an almost universal custom among the Polynesian tribes. It was generally performed by the priests, and was accompanied with religious ceremonies. In some places, however, as in New Zealand and the Southern Mar- quesas, the practice had become obsolete, or, for rea- sons now unknown and forgotten, prohibited. In the Hawaiian group its origin was ascribed to " Lua Nun," the tenth in descent from the period of the Flood {vide p. 128), thus not only indicating its extreme antiquity, but also its correlation to the Hebrew legend of Abraham. Another Hawaiian legend ascribes its introduction to Paumahua, a famous navigator, and noted ancestor of Hawaiian chiefs, who flourished about twenty-eight generations ago, or in the early part of the eleventh century of our era. But "Paumakua" belonged to the period of the South Polyne- sian incursions in the Hawaiian group, and probably only renewed or enforced the ancient practice. In tracing back the custom of circumcision, we find it practised by the Tagals and other pre-Malay tribes in the Asiatic Archipelago. It was the custom in Egypt \ from the earliest times ; also of Cushite-Arabia,2 and Phoenicia, or rather Palestine. 3 It does not appear to have obtained 1 In Sir Gardner Wilkinson's notes men from remotest antiquity." — to Mr. Eawlinson's Herodotus, book "Manual of Ancient History of the ii. chap. 37, I read that " its (oircum- East," by Lenormant and Chevallier, cision) institution in Egypt reaches vol. ii. p. 318. to the most remote antiquity ; we find 3 Herodotus, book ii. chap. 104, it existing at the earliest period of says : " Phoenicians and the Syrians which any monuments remain, more of Palestine ;" but in the notes to Mr. than 2400 years before our era, and Eawlinson's edition signed " G. W.,'» there is no reason to doubt that it it is shown that " circumcision was dated still earlier. ... It was a dis- not practised by the Philistines . . . tinctive mark between the Egyptians nor by the generality of the Phoeni- and their enemies ; and in later times, cians." In Egypt the custom was when Egypt contained many foreign common, "at least as early as the settlers, it was looked upon as a dis- fourth dynasty, and probably earlier. " tinctive sign between the orthodox It was also observed in Ethiopia and Egyptian and the stranger." Abyssinia. 2 " Circumcision established in Ye- THE POL YNESIA N RA CE. 1 05 among the Chaldeans, the Arian nations, or the Hebrew conveners of the Semitic stock. In the transmission of O customs, however, from one people to another, whose origin and purposes are lost and forgotten through the lapse of a^es, the observance of circumcision among the Kaffirs of South Africa may supply a link to establish the extension of Cushite-Arabian influence, through commerce and colo- nisation, in that direction. Sir G. Wilkinson, in his notes to Eawlinson's edition of Herodotus, loc. cit., argues that the Hebrews did not borrow circumcision from Egypt after the exodus, because its institution with them dated back to Abraham, and, having fallen into desuetude, was merely renewed or re- inforced by Moses. But the remarkable parallelism of the Hawaiian legend of Lua-Nuu with the Hebrew legend -of Abraham, and the institution of circumcision connected with each, doubtless indicate a common origin for both legends — a Cushite-Arabian origin, in a land where cir- cumcision was practised "from remotest antiquity," as well as in Egypt. Considering, moreover, that Abraham did not adopt circumcision until after his visit to, and return from, Egypt, on the occasion of the birth of his son Ishmael by the Egyptian woman Hagar, there is certainly some ground for holding that a custom — unknown to the Semitic tribes of whose lineage Abraham claimed to be, unknown or not practised by the Chaldean branch of the Cushite family where he and his father before him were born and bred — was borrowed or adopted by him from the Arabian or Egyptian Cushites, with whom he came in contact after leaving the uncircumcised Chaldeans of Ur and the uncircumcised Semites of Haran. Taken together with the numerous other instances of correlation of Poly- nesian and Cushite folklore, this custom and accompanying legend is but another argument for the long and intimate connection between the Cushite Arabs and the Polynesian ancestors, while the latter were still living on the shores of the Erythnean, or, somewhat later, occupying the Sunda io6 THE POLYNESIAN RACE. isles, the Saba-ii, Sava-ii, " the volcanic Saba " of Poly- nesian cosmogony. Manner of Burial. Two modes of burial, if so they may be called, ob- tained among the Polynesians. In the Marquesan and Tahiti groups, deceased people of consequence were ex- posed on raised platforms until natural decomposition and the action of the air had reduced the corpse to a skeleton. This custom was also practised by some of the Dyakh tribes in Borneo, and at the island of Pulo Mas, and may at some time have obtained greater prevalence in the Asiatic Archipelago, but I have found no traces of it in Hindostan or beyond. In the Hawaiian group this mode of burial was not practised. There the older and more general manner of disposing of the dead was to em- balm the body, or rather cover it with a glutinous wash made from the Ti-root, 1 which effectually sealed up the pores of the skin and excluded the air. The body was then deposited in a sitting posture in a cave on the moun- tain-side, or on some natural shelf or niche on the side of a precipice. These burial-caves seem to have been cither private family property, or the property of the commune living on the land where they were situated. Offerings were frequently carried there, and prayers performed by the relatives of the deceased. Tradition says that the first man, Kumuhonua, was buried on the top of a high moun- tain, and his descendants were all buried around him until the place was filled. In analogy with the above custom we find that the burial-places in the Hedjaz, in Southern Syria, in Egypt, and Nubia were generally on or near the summit of moun- tains, and in natural caves on their sides ; and among the various races which crowded each other on the plains of ancient Chaldea, where mountain-sides or natural caves were not available, the preservation of the dead by inter- ment in artificial tombs was equally practised. In no 1 Dracccna terminalis. THE POLYNESIAN RACE. 107 part of Polynesia, so far as I can learn, is there any indication that cremation was ever practised. Interment, no doubt, was the earliest custom. It preceded cremation in Greece and in Europe generally, 1 and the Hindoo custom of burning the corpses did probably not spread either west or east until after the Polynesians had left the mainland of Asia, nor did it obtain in the Indian Archi- 1 According to the researches of J. Grimm (" Ueber das Verbrennen der Leichen "), all the Arian peoples, with one exception, practised incre- mation at their funerals from time immemorial, in place of interment. The Indians, Greeks, Romans, Gauls, ancient Germans, Lithuanians, and Slavs, during heathen times, burned their dead with ceremonies which present evident traces of resemblance, notwithstanding their diversity. The Iranians alone at an early time aban- doned this ancient custom on account of the radical difference which arose in their religious creed. The Hebrews and Arabs never practised increma- tion. ( Vide A. Pictet, "Orig. Ind. Europ.," vol. ii. p. 504.) Is it fully established that the schism of the Iranians was subsequent to the sepa- ration and migrations of the Greeks, Romans. Gauls, &c, &c, from the Arian sto3k? M. Pictet adds, on p. 505, loc. cit.: " Cette coutume, comme l'observe Grimm, a du prendre naissance aux temps primitifs de la vie pastorale, avant l'etablissement de demeures fixes, parce qu'elle per- metfcait d'emporter avec soi la cendre vencre'e des morts." If so, why did it not obtain among the Turanian peoples, than whom none were more pastoral or nomadic ? But on page 529, vol. ii. , of the work just quoted, M. Pictet says : " D'apres -tous le developpements qui precedent, il semble Evident que l'usage de bruler les morts doit avoir existe deja chez les Aryas primitifs ; mais il est a presumer que la coutume plus simple de l'inhumation a tenu chez eux une certaine place, comme chez la plupart de leurs descendants. On la voit meme prescrite, dans quelques cas, par les lois de plusieurs peuples. Ainsi, d'apres Maim (v. 68) un enfant au-dessous de deux an3 doit etre in- hume, et il en etait de meme chez les Romains (Juven. Sat. 15, v. 139), sui- vant 1'line (7, 16), avant la dentition. Au temps de Cecrops, i'incineration etait peu pratiquee, et rinhumation parait avoir predomine chez les Ro- mains les plus anciens (Cicer. Leg. 22, 26 ; Plin. 7, 54). Numa defendit de bruler son corps, ce qui indique la simultaneite des deux usages, con- firmee 300 ans plus tard par laloi des Douze tables. Dans toute l'Europe du Nord, on trouve l'inhumation comme la coutume la plus ancienne, celle qui appartenait a ce qu'on appelle lage de la pierre, et ce n'est qu'a l'age du bronze que les urnes cineraires font leur apparition dans les tombeaux. On en conclut, non sans vraisemblance, qu'elles sont l'iudice de Farrivce en Europe des premieres immigrations ariennes, se m6tant a une race anterieure que nous ne connaissons plus que par les restes de 1 age de la pierre. Ce que Ton peut conjecturer, deja pour les anciens Aryas, c'est que I'incinera- tion, qui exigeait toujours un certain appareil, etait reservee pour les chefs et les hommes considerables, tandis que rinhumation dtait le lot de la multitude." This but confirms what I said above, that interment was of older practice than cremation, even among the Arians, and may have descended to the Polynesians from them as well as from the Cushites. 108 THE POLYNESIAN RACE. pelago while the Polynesians yet were the masters there. It is practised to some extent among the Dyaks, but only ex- ceptionally, whereas interment is the most prevalent mode. In connection with the funeral rites of the Polynesians, it may be observed that the practice of immolating one or more of the wives of a deceased chief, which obtained in Tonga and the Fiji islands, was not adhered to in the Hawaiian group, nor generally among the other groups ; but it was de rvjucur that more or less of the friends of the deceased should accompany him or her in death, in manifestation of their love and attachment. Those who thus died were called Moe-pu, " companions in sleep." The act, however, was purely voluntary, and generally performed by starving, sometimes by strangling. In Hawaii, when a chief died, according to rank and circum- stances, from one to forty human victims were in later times sacrificed at the Heiau (temple) in honour to the deceased. In some places, as in New Zealand, slaves were killed to accompany and attend on their masters. A somewhat similar custom obtains also among the Dyaks and Battas, where slaves are slain on the graves of the deceased. I have no means at hand to ascertain if this custom of immolating wives, friends, and slaves in honour of the dead was ever practised among the Cushite peoples, and transmitted by them to the forefathers of the Battas, Dyaks, and their Polynesian cousins ; but it certainly obtained among the Arian branches before their disper- sion.i it prevailed among the Gauls, Scandinavians, Lithuanians, and Slavs. With some the sacrifice of the wife was voluntary and optional, but with all the sacrifice of clients and slaves was compulsory. Among the Vedic Hindoos the immolation of the wife was not compulsory, though she was expected to make a semblance of accom- panying her husband on the funeral pyre. 2 Thus, if the 1 "Origiues Ind. Europ.," par A. Pictet, vol. ii. p. 527. a Ibid., p. 526. THE POL YNESIAN RA CE. 1 09 custom did not come to the Polynesians from their Cushite teachers and civilisers, it was one of those early national traits of Arian descent which no subsequent Cushite training could efface. Of Castes. It is undeniable that a system of caste, a peculiar and exclusive division of society, obtained throughout Polynesia at the time when its groups were first visited by Europeans. Though the arrangement of these classes of society differed somewhat in different groups, yet a threefold division may be considered as the most ancient — chiefs, commouers or freemen, and slaves, or, as expressed in Hawaiian, na Lii, na Makaainana, and na Kauwa. In Tahiti, Aril, Raa- tira, Manahune. In Tonga, Eihi, Mat abide, Mua, and Tua. In Samoa, Alii, Tulafale, Songa. In Earotonga, Arihi, Rangatira, Unga. In New Zealand there were but two classes, the Rangatira, the freemen, and the Taurekareka, or slaves. In Marquesas, Hakaiki, Tuwpoi. ' The priest- hood does not seem originally to have been a separate class or caste among the Polynesians, but to have been a prerogative, right, or duty of the chiefs and heads of families. In course of time it became hereditary in certain families, as in Hawaii. In New Zealand, where the peculiarly distinctive title and functions of chieftain- ship had become extinct, yet the priests were styled Arihi, indicating the former connection between the chiefship and the priesthood. In Tahiti the priests were generally relations of the chief families, but socially never more than the delegates of the presiding chief for religious purposes. 1 1 See " Etat de laSociete Taitienne swarms of immigrants; that the a l'arrivee des Europeens," par M. de Manahune, being the first, was con- Bovis, in "Revue Coloniale," 1855. quered by the E,aatira, who, in their In this, in many regards, thoughtful turn were subdued by the Arii. JIo and well-written essay, M. de Eovis does not think that the Arii and considers that the Tahitian group Baatira arrived together at the group, was peopled by at least three different and that the distance between the 1 10 THE POL YNESIAN RA CE. That this division of society was older than the arrival of the Polynesians in the Pacific, and was brought with them from their former habitats in the west, may safely be inferred from the universality of the custom among all the principal groups ; and on each the institution is as old as the people, and goes back to the earliest times of their remembrance. In the west, then, among the Polynesian congeners in the Asiatic Archipelago, and beyond, let us look for the origin of this political organisation of society, Throughout the Indian Archipelago, whatever modifica- tions may have supervened from conquest, change of religion, and other causes, the essential groundwork of pre-Malay society was a division into chiefs, landholders, subjects by tenure, but free in persons, and slaves, whom war or other social causes reduced to that condition. With the intro- duction of Brahmanism in the Archipelago, its elaborate system of caste does not seem to have fallen in congenial soil or to have materially modified the ancient division of society. That division is then older than the Hindoo or Malay supremacy in that Archipelago. It is one of the remnants of the lod Cushite, Chaldeo- Arabian training and civilisation, which, twenty to thirty centuries after its power had vanished, elsewhere and its very name been forgotten, has so strangely been preserved in Polynesian folklore and in Polynesian customs. At first sight it may appear so, to judge from the condition of society in the latter centuries of Polynesian life, when ages of oppression and deepening barbarism had succeeded in sharply defining and cruelly two castes was too great for them to tainly brought with them the same have had a common origin (pp. 240, orders of society which prevailed in &c. ) their former homes. It is very probable that the Tahitian M. de Bovis further intimates that group was peopled at different times the crisp, frizzled hair, and lean, by the arrival of Polynesian emigrants, lank bodies, found in some of the But whether those emigrants came Polynesian tribes, as well as the from Fiji via Samoa, or from Fiji number of Malay words, derive from direct, on the expulsion of the Poly- a later immigration of Malays into nesians from that group, they cer- Polynesia. THE POL YNESIA N RACE. m enforcing the lines of separation between different classes of society. In this way, on some of the groups, as on Hawaii, the priestly order obtained exclusive privileges, and became a tabued caste, whose dicta even the highest chiefs only disregarded at their peril. But behind this later corruption and degradation the national legends give us glimpses of the earlier condition of society, when, as above stated, it exhibited a less artificial and more primi- tive form. We look in vain to the older Cushite-Sabrean or Cushite-Chaldean systems of caste ; we look in vain to the later Brahminical system for a prototype of the original Polynesian classification of men. 1 It is certainly older than the latter, and, if not older than the former, it is dif- ferent in principle and origin, though somewhat modified perhaps by contact with it. Failing in these directions to find an analogy or an original of Polynesian classification, I find it in the early Arian condition of society, previous to the irruption in India, previous to the migrations of the Indo-European branches, when, having already become aggressive, the nation or its various tribes naturally enough were divided into the warrior class, subsequently the nobles, those who fought the battles of the tribe or nation, and the cultivators, herdsmen, artisans, and general mass of the people who provided for the wants of the former. 2 To these two primary classes became in time added the slave class, whom the fortune of war or social laws had reduced to slavery. And that such was the early Arian condition of society may be inferred also from the classification obtain- ing among the Scandinavian branch of the Arian stock at its first appearance in historic light, 3 which was that of chiefs, 1 " The basis of the social organisa- preceded them in the basins of the tion of the Sabasan kingdom was the Indus and Ganges, and whom they system of caste, unknown to the conquered, namely, the Sudras and Shemites, an essentially Cushite in- Kausikas." — Manual of the Ancient stitution, which, wherever it is found, History of the East, by Lenorniant is easily proved to have originated and Chevallier, vol. ii. p. 317. with that race. We have seen it a Vide Max Midler's Essay on flourishing at Babylon. The Arians "Caste" in " Chips from a German of India, who adopted it, borrowed it Workshop," vol. ii. from the Cushite populations who 3 About 400 B.C. 1 1 2 THE POL YNESIA N RA CE. freemen, and slaves ; the chiefs performing the functions of priesthood down to Christian times, 1 and the freemen consisting of the husbandmen and the body of the people generally, who were free to follow this or that chief, and whose consent was necessary to all public enterprises. This branch, moreover, taking a northern route in its migration through the w T ilds of Scythia and Eussia, was less, if at all, affected by contact with the Cushite civilisa- tion, wdtich so deeply tinged the Indo-European branches who took a more southerly route. As the Arian organisation was then, before the disper T sion of the race, north, west, and south, so the Polynesian has remained with slight modifications until compara- tively modern times. Like interment of the dead, it w T as an Arian heirloom from a pristine, pre-vedic age, which Cushite culture and contact did not eradicate. If the seeds of stringent exclusiveness and priestly supremacy - were sown by Cushite . intercourse, they took long ages to develop, and in most of the tribes never bore fruit at all. It is true that (a Hawaiian legend relates that Kahiko, an ancestor of the people, had three sons, Wakea, Lihau-ula, and Mdkuu ; that the chiefs, Alii, sprang from the first ; the priests, Kahuna, from the second ; and the husband- men, Makaainana, from the last, thus indicating a pos- sible origin of the classification of the people. But this legend, besides being contradicted by other legends of probably older date, which mention only two sons of " Kahiko," and that " Lihau-ula " was older brother of " Wakea," and was not a priest, but a warrior chief w T hom " Wakea " conquered in battle, is evidently a composition of later date, when the priesthood had become a tabued institution and caste, and sought a sanction for itself, and a raison (litre in the ancient folklore. 2 1 "The early kings of the various 2 In later Hawaiian times the priest- Grecian states, like those of Rome, hood, 'Oihaanu Kahuna, consisted were uniformly priests likewise." — of ten branches or colleges. He who G. Rawliuson's Herodotus, vol. iii. was master of or proficient in all 161, n. 2. was called a high priest, Kahuna THE POLYNESIAN RACE. "3 The Tabu. Throughout Polynesia the Tapu or Kapu system of pro- mulgating and enforcing religious or political laws, was equally known, equally developed, and equally practised. It was a body of negative commandments — " Thou shalt not " do this, that, or the other thing under penalty, bind- ing on the consciences of the people. The meaning of the word is " sacred, prohibited, set apart," whether referring to religious or civil matters. The religious tabus relating; to rites, observances, public worship, and the maintenance of the gods and their priests, were well known, compara- tively fixed in their character, and the people brought up from childhood in the knowledge and observance of them. But the civil tabus were as uncertain and capricious as the mind of the chief, priest, or individual who imposed them on others, or on himself and his family. However much the Kapu system may in after ages have been abused, it no doubt was originally a common law of the entire Polynesian family for the protection of Nui. The names of these branches of learning or colleges were — 1st, Ana- ana ; 2d, Hoopiopio ; 3d, Hoouna- una ; these three were connected with the practice of sorcery, by- prayer and signs, &c, for the death or injury of another; 4th, Hookomo- komo ; 5th, Poi-Uhane ; connected with divination by causing spirits of the dead to enter the body of a person and possess it ; 6th, Lapaau- maoli, medicine and surgery gene- rally; 7th, Kuhikuhi-iwuone, per- taining to the building of temples, dwellings, &c, their location, pro- priety of time, and favourable or un- favourable conditions, materials, &c. ; 8th, Oneone-i-honua ; 9th, Kilo-kilo; 10th, JYana-uli, different degrees and classes of soothsayers, diviners, and prophets. Each one of these ten was again subdivided in classes and occu- pations of detail counected with the religious rites and sacrifices. Vol. 1. The priesthood was governed by rules and regulations of its own stringent oaths were exacted before admission, and severe penalties upon infraction. A number of gods were invoked by the different classes and subdivisions of the priesthood; but the principal god, who seems to have been the presiding and tutelar deity of the entire body of priests, was called TJli. As I have found no god in the archaic Hawaiian theogonies, nor those of the other Polynesian groups, bearing the name of Uli, I am in- clined to believe that it was at first a sacerdotal epithet, degenerating into a soubriquet, and finally becoming a distinct personification ; its first sense being equivalent to that of Hiica, " sacred, dark - coloured, blue or black," and as such applied to one of the great principal gods. H 1 14 THE POL YNESIAN RA CE. persons and things, an appeal to the gods for punishment of offenders, where human vigilance failed to detect them, or human power fell short of reaching them. The universality of the " Kapu " within the Polynesian area, without referring to the positive declarations of par- ticular legends, makes it beyond a doubt that the Polyne- sians brought it with them from their former abodes in the west, and there traces are yet found of it. In Timor a system of interdict, called Pomali, 1 was practised, which by competent travellers is said to very strongly resemble the Polynesian " Kapu ; " and among the Dyaks of Borneo a similar custom of interdict is said to have obtained, and was there called Pamali} Among the Cingalese and southern Hindoos, the word Kapu, which is the name of the scarlet string worn round the arm or wrist, to indicate that the wearer is engaged in a sacred cause, and should not be interrupted, 2 — singularly enough, though with somewhat altered sense, recalls the name and purpose of the Polynesian interdict. When we consider that the Ceylonese never adopted Brahmanism, and that their ear- liest civilisation and religious notions were moulded, if not created, by the Cushite Arabs, whose intercourse with, and hold over, the Dravidian and other peoples in southern India and the islands, was long and intimate, it is reason- able, in conjunction with other facts, to seek a common origin for the Polynesian and the Cingalese word in some Cushite term of religious import, now forgotten and as yet undiscovered. Tattooing. This custom has been so widely diffused among the vari- ous nations of antiquity as to afford no reliable guide in ethnic inquiries ; yet now and then certain traits connected with it challenge attention by their striking similarity to those of other peoples, and their apparent conformity to a 1 Malay Archipelago, by E. A. 2 Oriental Illustrations, by Joseph Wallace, p. 203. Roberts, p. 133. THE POL YNESIAN RA CE. 115 once common rule. It is well known to be universally practised among the Polynesian tribes, varying only in style, in pattern, and fulness ; and variations occur even . anions different subdivisions of the same tribe. This also was a custom brought with them from the west. According to M. Domeny de Eienzi, it is practised on the Island of Savu, south-east of Timor, and among the Dyaks of Borneo and other tribes in the Asiatic Archipel. 1 In Leitch Eitchie's " History of the Indian Empire," vol. ii. p. 428, Art. "New Zealand," occurs the following paragraph: — " Tattooing is fast going out of fashion with cannibalism but it appears to have been but little practised at any time by the females, who have merely three short lines drawn from the under-lip. This is precisely the case with the Coptic women." Holy Waters. Among the many Polynesian customs which they brought with them on entering the Pacific, and which serve as links long lost or overlooked in the ethnic chain that binds them to the Cushite and Arian races, may be mentioned the preparation and use of sacred or holy waters. From New Zealand to Hawaii the custom prevailed, and its efficacy was believed in. The origin and explanation of the custom is thus given in the Hawaiian " Kumuhonua " legend : — " The Ocean, lea moana nui a Kane, which surrounds the earth, was made salt by Kane, so that its waters should not stink, and to keep it thus in a healthy and uninfected state is the special occupation of Kane. In imitation of Kane, therefore, the priests prepared waters of purification, prayer, and sanctification, Wai-liui-lcala, Wai-lupa-lupa, and he kai-olena, for the public ceremonials, for private consola- tion, and to drive away demons and diseases. Such holy waters were called by the general name of ha wai kapu a Kane." 2 Prom the sprinkling of a new-born child to the 1 Oceanie, par M. M. G. L. Domeny - In some ancient prayers in my de Eienzi, vol. i. p. 65. possession, these waters are also calk' ' 1 16 THE POL YNESIAN RA CE. washing of the dying, its application was constant and multifarious. The baptismal ceremony — E fiiri — of the New Zealanders, related by Dieffenbach, 1 with the accom- panying prayers invoking the gods Tu and Rongo (the Hawaiian Ku and Lono), is a valuable and remarkable remnant of the ancient culte. 2 The use of these holy waters was of the highest antiquity, and universal through- out Polynesia. It was a necessary adjunct in private and public worship, a vade mecum in life, a viaticum in death ; and even now, fifty years after the introduction of Christi- anity in these Hawaiian islands, there are few of the older people who would forego its use to alleviate pain and remove disease. A custom so universal, so deeprooted, must have existed previous to the arrival of the Polynesians in the Pacific. I have not the means of knowing to what extent, if at all, its use has been retained among the Polynesian cousins in the Malay Archipelago, but it certainly had its origin farther west. That holy water — the water of the Ganges, and, perhaps, previously of the Indus — was employed by the Hindoos for almost the same purposes as by the Polynesians is well known, and would at first sight seem to claim priority of consideration when looking for prototypes or analogues of the Polynesian custom. But there is a radical difference between the two, which makes it little likely that the latter owed its origin to the former. The holy water of the Hindoos — the Ganges water — is holy per se, and re- quires no mixture, preparation, or prayers to make it so. The holy water of the Polynesians was expressly prepared and consecrated with prayers in order to obtain that par- ticular efficiency for religious and medical purposes which it was believed to possess. The Polynesian holy water Wai-oha. The word Oka in this sense 1 Travels in New Zealand, by Dief- is obsolete in the Hawaiian, but is fenbach, p. 28. still retained in the Marquesan dia- ~ See Appendix, No. V. lect, where it means "sacred, ador- able." THE POL I 'NESIAN RA CE. 117 represented the great world-ocean and its purifying pro- cesses ; the Ganges water represented nothing but itself. Looking beyond the Indus and the Chaldean Empire, of whose customs in this respect I have seen no detailed information, I find in Ancient Greece a striking corre- spondence with the Polynesian custom. Holy water, and sprinkling and washing with it, was an indispensable element of the old Greek ritual. In the preparation of the Greek holy water, as well as in the Hawaiian, salt was a necessary ingredient. With the former, sea-water was preferred, when attainable, on account of its saltness ; otherwise salt was invariably mixed with the fresh water, and sometimes brimstone added. At the entrance of the Greek temples stood the " Perirrantoerion " or vessel con- taining the holy water, and no person was permitted to pass beyond or assist at the sacrifices who had not previ- ously washed his hands in it, or been sprinkled with the water it contained. The Greek custom of lustral waters was probably of Arian origin, 1 but the peculiar manner of its preparation, unknown to, or, so far as I can learn, not practised by, their Arian congeners, may possibly be a modification brought about by their connection with the Cushite civilisation, of which the early Phoenicians were such remarkable propagandists ; or, perhaps earlier still, 1 In Les Origines Indo-Europeen- iens, les eaux crees par Ormuzd nes, par Ad. Picbet, vol. ii. p. 681, I etaient aussi le principal moyen de read : purification, surtout apres avoir ete "Les traits essentiels d'une culte, consacrees par la ceremonie du Zao- elementaire des eaux se retrouvent thra, ce qui rappelle singulierement encore presque inaltcres chez les l'eau benite du catbolicism (Spiegel, principaux peuples de race Arienne. Avesta, ii., xcii.) L'emploi des eaux Dans le Rigveda, comme dans l'Aves- lustrales dans l'antiquite classique est ta, elles sont encore invoquees sous sufh'sammeut connu. Les Scandi- leur nom propre, dpas, au pluriel et naves consideraient les eaux du ciel collectivement. On les appelle les comme sacrees ; l'Edda les appelle meres, les divines ; on dit d'elles qu' heil'dg votn, et heilaicdc du moyen age elles renferrnent Yamrta, l'ambroisie, germanique, c'est a dire l'eau de et tous les remedes salutaires ; on source puisee a minuit, ou avant le leur demande, non-seulement la saute lever du soleil, devenait un remede du corps, mais la purification de puissant, et acquerait de proprietes 1'a.me de tout peche. Pour les Iran- magiques (Grimm, Deut. Myth. 327). 1 1 8 THE POL YNESIAN RA CE. while skirting the upper borders of Chaldea on their migra- tion through Asia Minor. At any rate, this similarity in a matter of detail of preparation cannot well be considered as a coincidence under pressure of similar circumstances, but was more likely an engraftment in different directions from a common source and a once common religious idea. The idea of holy water as a co-efficient in religious ceremonies was common to the Oriental nations. The Jews only borrowed their Laver from others. And the metaphysical explanation of the Hawaiians is perhaps as ancient a conception of the action of the sun on the ocean as any on record. Cities of Refuge. Some stress has been laid on the peculiar institution called " cities of refuge," Hawaiian, Puu-honua, which was found to have obtained among the Polynesians, especially the Hawaiians, and which has been quoted as another in- stance of Hebraic influence upon the customs and culture of the Hawaiians. * Cities of refuge, however, were not an institution peculiar to the Hebrews, and originating with them. They existed in the time of ancient Greece. We read of the temple of Ceres at Hermione, in Argolis, which was a similar institution ; and there were numerous others, both there and elsewhere, where Cushite influence had modified the customs and moulded the culte of the people on its own pattern. The Division of the Year, &c. The Polynesians divided the year into seasons, months, and clays. The seasons — Tau or Kan — of the year were generally two, the rainy or winter season, and the dry or summer season, varying according to the particular situa- tion of the group, either north or south of the equator. The commencement of the seasons, however, were regu- lated by the rising of the Makarii stars, the Pleiades, at the time of the setting of the sun. Thus, in the Society THE POL YNESIAN RA CE. \ i g group, the year was divided in Makarii-i-nia, Pleiades above the horizon, and Makarii-i-raro, Pleiades below — the first from November to' May, the latter from May to November. In the Hawaiian group the year was also divided in two seasons — Hooilo, the rainy season, from about 20th November to 20th May; and Kau, the dry season, from 20th May to 20th November. 1 In the Samoan, Tau or Tau-sanga meant originally a period of six months, and afterwards was employed to express the full year, or twelvemonth, as in the Tonga group. There are traces also on the Society group of the year having been divided in three seasons, as at one time was done by the ancient Egyptians, Arabs, and Greeks, though the arrangement of the months within each season seems to me to have been arbitrary, and probably local. In regard to the division of the year by months, the Polynesians counted by twelve and by thirteen months, the former obtaining in the Tonga, Samoan, and Hawaiian groups, the latter in the Marquesas and Society groups. Each month consisted of thirty days. It is known that the Hawaiians, who counted twelve months of thirty days each, intercalated five days at the end of the month Welehu, about the 20th December, which were tabu-days, dedi- cated to the festival of the god Lono ; after which the 1 Mr. R. G. Haliburton, of Halifax, Pyramid, 1865, by C. Piazzi Smyth, N.S., has shown that the primitive vol. i. chap. xii. p. 330. year of the Pleiades was a pre-histo- In the Hawaiian group the red star rical tradition, spread amongst almost in the constellation is called Kao — the all races of mankind in both the new star Autares, in the horns of the Bull and old hemispheres, and alike in the — was also called Makalii. That the north and south. The leading cha- ancient Hawaiians should have called racteristics of that year being that it the constellation of the Bull— Taurus began on the 19th day of Athyr, or — by the very name which was one of November, when the Pleiades, or the earliest appellations for that ani- their containing constellation, the mal, while the Arian stock was yet Bull — the great Tau of the Egyptians, unsundered, is one of those quiet but the Taurus of the Latins, the Thor surprising witnesses to the Western of the Scandinavians, and the Atliir origin and Arian connection of the or Arthur of the ancient Britons — Polynesian family, which rise in judg- was on the meridian at midnight, ment against modern theorists of l';i- Vide Life and Work at the Great puan, Malay, Mexican, or other pro- clivities. 1 20 THE POL YNESIA N RA CE. New Year began with the first day of the month Mahalii which day, being the first of the year, was called Maka- hiki (equivalent to " commencement-day"), and afterwards became the conventional term for a year in the Hawaiian, Marquesas, and Society groups. 1 There is evidence that the Marquesans at one time counted the year by ten lunar months, and called it a Picni — a circle, a round, a revolu- tion — but how they managed either this or the year of thirteen months to correspond with the division by sea- sons, or with the sidereal year, I am not informed. It is probable that in Tahiti the month Te-eri was occasionally, perhaps alternately, dropped from the calendar. 2 That a computation by lunar months preceded the other is evident from the very names given to different days in the month, but both computations were certainly far older than the arrival of the Polynesians in the Pacific. They brought those names and those computations with them. The absolute Hawaiian expression of Ana-hulu indicates a primary but subsequently disused division of the month 1 Rev. S. Dibble, in his History numbered twenty-nine, and some- of the Sandwich Islands, Lahaina- times thirty days in a month." Mr. luna, 1843, p. 108, says : "Those who Dibble omits to mention that the took the most care in measuring time "correction " of their reckoning " by measured it by means both of the the stars " was made by the intercala- moon and the fixed stars. They di- tion I have referred to. It thus ap- vided the year into twelve months, pears that the Hawaiians employed and each month into thirty days, two modes of reckoning— by lunar They had a distinct name for each of cycles, whereby the monthly feasts, the days of the month, as has been or Kapu-days, were regulated ; and shown on a former page, and com- the sidereal cycle, by which the close menced their numbering on the first of the year, and the annual feast of day that the new moon appeared in Lono, was regulated, ihe west. This course made it neces- 2 The alternation of twenty-nine sary to drop a day about once in two and thirty days in the Hawaiian months, and thus reduce their year months, referred to by Mr. Dibble in into twelve lunations instead of three the above note, though certainly not in hundred and sixty days. This being general usage among the Hawaiians, about eleven days less than the side- yet, as one of the several modes of real year, they discovered the discre- computing time which they brought pancy, and corrected their reckoning with them from their primitive by the stars. In practice, therefore, abodes, forcibly reminds one of the the year varied, being sometimes Hebrew and Assyrian division in twelve, sometimes thirteen lunar months of alternate twenty-nine and months. So also they sometimes thirty days. THE POL YNESIA N RACE. 121 into periods of ten days, corresponding to the increase, the full, and the decline of the moon, analogous to the Greek Dechcemera and the Egyptian Se-sic ; and the institution of the Hawaiian Kapu or sacred-days at intervals of ten days seems to favour such a conclusion, for I look upon the fourth monthly Kapu-day, — that of Kane on the 27th of the month, — as of subsequent introduction, following so closely, as it does, upon the Kapu-day of Kaloa-ku-kahi, the 24th. 1 Though obsolete now in common parlance, the term Ana-hulu is of frequent occurrence in the ancient legends and songs as a measure of time comprising ten days. fThe Hawaiian day was divided in three general parts, like that of the early Greeks and Latins, 2 — morning, noon, and afternoon — Kakahi-aka, breaking the shadows, scil. of night ; Awakea, for Ao-akea, the plain, full day ; and Auina-la, the decline of the day. The lapse of the night, however, was noted by five stations, if I may say so, and four intervals of time, viz. : (1.) Kihi, at 6 p.m., or about sunset ; (2.) Pili, between sunset and midnight ; (3.) Kau, indicating midnight ; (4.) Pilijmka, between midnight and 1 S. M. Kamakau, in one of his the other Polynesians, so far as I articles on ancient Hawaiian beliefs, know, never had a week of seven refers to an old legend, according to days. On comparing the Tahitian which " the creation commenced on and Hawaiian calendars, and finding the 26th (27th ?) of the month, on the the Kanaloa (Taaroa and Kaloa) day called Kane, and was continued days in the same position on both, I during the days called Lono, Mauli, am strongly inclined to believe that Muku, Hilo, and Hoaka. In six days when in after ages the South Poly- the creation was done. The seventh nesian element obtained ascendancy day, the day called Ku, became the in Hawaii, its principal god Taaroa, first Kapu-day— La-Kapu. The first Tangaroa, and the days dedicated to and the last of these seven days in him, were interpolated on the Ha- every month have been kept Kapu waiian calendar, and the A'a»e-day ever since by all generations of Ha- and its festival or Kapu was made to •waiians." The seven days of creation follow after that of Taaroa, a being and rest (Kapu) may be a dim re- whom the Hawaiians did not recog- collection of the Hebrew-Chaldean nise as a divinity in their earlier version of the creation ; but the ap- creed, nor until after that invasion of plication of the first day as a Kapu- South Polynesians, of which I shall day to Kane is evidently a priestly have more to say hereafter, commentary, and of later origin. Prac- 2 Hws, Mecroe-'H/ias, and AaXij ; tically the Hawaiians, and none of Mane, Meridies, Suprema. 1 22 THE POL YNESIAN RA CE. sunrise, or about 3 a.m. ; (5.) Kihvpuha, corresponding to sunrise, or about 6 a.m. According to M. D. de Eienzi a similar division of day and night seems to have been current of old in Jawa. To this may be added that the Polynesians also counted time by the nights — Po. " To-morrow " was A-po-po, lit. the night's night. "Yesterday" was Po-i-nehe-nei, the past night. Po-akahi, Po-alua, the first, the second day. Po was the collective term for the twenty -four hours, and Ao or daylight was but the complement of the full Po. This method of reckoning by nights ascends to the hoariest antiquity, j The unbroken Arians counted by nights, and the custom prevailed late into historic times among the Hindoos, the Iranians, the Gauls, the Cymri, the Saxons and Scandinavians. 1 The Hebrews commenced time with the evening of the first day : whether the idea came to them from Chaldea or from Egypt, I cannot say. The Babylonians believed that the world had been created at the autumnal equinox. 2 There has been so little light thrown upon the ancient computations of time among the pre-Malay inhabitants of the Indian Archipelago — those blood-relations of the Polynesians — that I am again unable to refer to them as a connecting link between the latter and their more western congeners ; but the lunar computations of both Arians and Cushites ; the division of seasons by both ; the method of determining the sidereal year by the Pleiades ; the method of intercalating the twelve months of thirty days with five days, which obtained in Egypt and in Persia and among the Vedic Arians, though the latter intercalated an entire month of thirty days after every quinquennial cycle ; 3 the division of the month in thirds of ten days each, as in Egypt and ancient Greece, 1 Origines Ind.-Europ., par A. East, by Lenorniant and Chevallier, Pictet, vol. ii. p. 588. vol. i. p. 451. 2 Manual of Ancient History of the 3 Origines Ind.-Europ., par A. Pictet, vol. ii. p. 608. THE POL YNESIAN RA CE. 1 23 and which possibly was the basis of the ancient Javanese subdivision of the week into five days, before Brahrnanisni introduced the week of seven clays ; x the division of the day into three portions and the night into four ; the count- ing the length of time by nights and not by days ; all these cumulative parallelisms, I think, will go far to con- firm the western origin of the Polynesians, and their intimate connection in pre-historic times with the Arian and Cushite peoples. They cannot all be fortuitous coin- cidences, and must, therefore, justly be considered as remnants of a once common civilisation, which the isola- tion of two thousand years or more has not been able entirely to efface, though partially obscured. 2 I think it proper in connection with this subject to refer to an article in the " Ethnological Society's (London) Transactions," vol. ii. p. 173, "On the Antiquity of Man from the Evidence of Language," by J. Crawfurd, a gentle- man whose researches and knowledge regarding the Indian Archipelago and its various peoples were undoubtedly great and valuable. He says that " the terms employed in 1 Speaking of the rural calendar of waiian, who, some twenty-five years Jawa, which was in vogue before the ago, composed a work on "Hawaiian introduction of Brahmanism, M. D. Antiquities," mentions that the an- de Rienzi, in Oceanie, vol. i. p. 167, cient year closed with the mouth of says : " Le calendrier rural est de 360 Ikuwa, about 20th of November, jours. 11 se divise en douze mois ou whereas Kamakau gives it as ending douze saisons, d'une longueur inegale, with the month of Welehu, or about et est termine par des jours inter- 20th December, and the new year calaires." commencing witli the month Makalii. 2 It appears that there was con- Both of these authorities agree, how- siderable diversity between the dif- ever, that the public sacrifices and ferent sections of the Hawaiian group Kapu-days were only observed during in counting the months and the days, eight months of the year, and discon- In several respects the Kauai and tinned during the months of Ikuwa, Oahu calendar differed from that Welehu, Makalii, and Kaelo, when, which was generally followed on in the mouth of Kaulua, they re- Hawaii and Maui. I obtained my commenced again. information from Hon. S. M. Kama- It is probable that D. Malo refers kau, an intelligent and educated to the sidereal year regulated by the Hawaiian, born and brought up while Pleiades, and according to which the the heathen regime still prevailed, seasons were divided, and that Kama- On the other hand David Malo, an kau refers to the solar year regulated equally intelligent and educated Ha- by the winter or December suhsiiee. 124 THE POLYNESIAN RACE. the computation of time, according to their poverty or maturity, afford material evidence of the antiquity of man." He quotes Australians who have no terms for solar day, month, or year. He refers to the same poverty in the ruder languages of Africa and America, and then says: " The principal nations of the Phillipine Islands had made considerable progress when first seen by Europeans ; yet their languages have no native name for solar day, month, or year, for these have been taken from the more advanced Malayan nations. The language of Madagascar has no names for month or for year, and has taken both from the Malay; and it is remarkable that the Malayan term for year has even reached the rude inhabitants of the islands of the Pacific." " Interdum dormitat Homerus ; " it will be seen from the foregoing pages that, as regards the " rude inhabitants of the islands of the Pacific," Mr. Crawfurd's remarks are not borne out by actual facts, nor yet by philological evidence. The principal Polynesian groups, as above shown, had not only names for year, month, and seasons, but had also dis- tinct names for every month and every day in a month. Nor are these names of Malay origin. The Malay word for year is Taun or Tahun. In all the Polynesian dialects the primary and original meaning of Tom is " a season," " a period of time." In the Tonga group it has the further sense of " the produce of a season," and, derivatively, " a year." In the Samoan group, beside the primary sense of " season," it has the definite meaning of " a period of six months," and conventionally that of " a year." In the Society group it simply means "a season." In the Hawaiian group, when not applied to the summer season, it retains the original sense of an indefinite "period of time," "a lifetime," "an age," and is never applied to a year; its duration may be more or less than a year, according to circumstances and the context. Thus in all the Polynesian dialects it retains the primary abstract sense, whereas in the Malay it has only the derivative concrete meaning. THE POL YNESIAN RACE. . 125 The Polynesian also retains what I consider the original form of the word, while the suffix n in the Malay betrays a later corruption. Had the Polynesians received the word from the Malays, its form, by the invariable rules of the former language, would have been Tauna or Kahuna. The Polynesian names for month are Masina or Mahina, Malama, and Avae, and there are certainly no trace of Malay in them ; they are identical with the current names of the moon, and indicate the early computation by lunar months. In proof that the Polynesians were not beholden to the Malays for the names of year, season, month, or days, but had a nomenclature particularly their own, the following tables may suffice : — Names of Months. Hawaiian. Sainoan. Tonga. Society Islands, Iluahine. Marquesas, Futuhiwa. 1 Makalii Utuwa-mua Liha-niua Avareku Kuliua 2 Kaelo Utuwa-muli Liha-mui Faaahu Katuua 3 Kaulua Faaafu Wai-mua Pipiri Eliuo 4 Nana Lo Wai-mui Taaoa Nanaua 5 Welo Aunuau Hilinga-gele-gele Aununu Oaumanu 6 Ikiiki Oloamanu Tanu-manga Apaapa Awea 7 Kaaona Palolo-mua Uluenga Paroro-mua Ehua 8 Hinaieleele Palolo-muli Hilinga-mea Paroro-muri Weo 9 Hilinehu ' Mulifa Fuca-afu-mate Muriaha Uaoa 10 Hilinama 1 Lotuanga Fuca-afu-moui Hiaia Ualiaameau 1 1 Ikuwa Taumafamua Uluagi-mate Tema Pohe 12 Welehu Toe-tauafa Te-eri Napea 13 Te-tae Makau 1 Also called " Mahoe-mua " and " Mahoe-hope.'' 126 THE POLYNESIAN RACE. Names of the Days in the Month. Hawaiian. i Hilo 2 Hoaka 3 Kukahi 4 Ku-lua 5 Ku-kolu 6 Ku-pau 7 Ole-ku-kahi 8 Ole-ku-lua 9 Ole-ku-koiu io Ole-ku-pau 1 1 Huna 1 2 Mohalu 1 3 Hua 14 Akua 1 5 Hoku 1 6 Mahealani 1 7 Kulu 1 8 Laau-ku-kahi 19 Laau-ku-lua 20 Laau-pau 21 Ole-ku-kaki 22 Ole-ku-lua 23 Ole-pau 24 Kaloa-ku-kalii 25 Kaloa-ku-lua 26 Kaloa-pau 27 Kane 28 Lono 29 Mauli 30 Muku Society Islands. Hiro-hiti Hoata Hami-ami-mua Hami-ami-roto Hami-ami-muri Ore-ore-mua Ole-ore-muri Tamatea Huna Ari Maharu Hua Maitu Hotu Mara'i Turu-tea Raau-mua Raau-roto Raau-rouri Ore-ore-mua Ore-ore-roto Ore-ore-muri Taaroa-mua Taaroa-roto Taaroa-muri Tane Ro'o-nui Ro'o-maori Mutu Teriere Marquesan. Ku-nui Ku-hawa Hoaka Maheamakahi Maheaniawaena Koekoe-kaki Koekoe-waena Poipoi-haapao Huna A'i Huka Meha'u Hua Akua Huku-nui Huku-manae Ku'u Aniwa Makahi Kaau Kaekae-kaki Waena Haapao Hanaokalii Wawena Haapaa Puhiwa Kane Ona-nui Onamate THE POLYNESIAN RACE. 127 The names of the Tahitian and Hawaiian seasons have been mentioned. The solstices were observed and named in Tahiti. The December solstice was called Eua-maoro or Eua-roa ; the June solstice, Eua-poto. The Hawaiians called the northern limit of the sun in the ecliptic ke A Ia- mbi polokhva a Kane, " the black shining road of Kane ; " and the southern limit was called ke Ala-nui polohiwa a Kanaloa, " the black shining road of Kanaloa ; " and the equator was named ke A la-ula a ke kuukuu, " the bright road of the spider ; " and also ke Ala i ka piko Wakea, " the road to the navel of Wakea," equivalent to " the centre of the world." l Whatever the origin of these names, and the knowledge which underlies them, they certainly owe nothing to Malay instruction. Superstition. As a matter of course, and not otherwise to be expected, the Polynesian folklore abounds with superstitious notions and usages. Their belief in, and reverent and affectionate regard for, their deceased ancestors, 2 the Au-makua of Hawaii, the benevolent ones who protected their descen- dants, the Oro-matua of Tahiti, the malevolent ones who were to be propitiated by prayer and offerings. 3 Their 1 The ancient Hawaiians knew and one more travelling star, but he had named the five earliest known planets, never recognised it, and was acquaint- which were called collectively na, ed with only these five. The more Hoku aea, "the wandering stars," in distinguished fixed stars and clusters contradistinction to the "fixed stars," had their distinct names, and the na Hoku paa. Their names were ac- people were in the habit of observing cording to Hoapili — them so much, that they judged of Mercury, Kaaivela, also Hoku-ula. the hour of the night about as accu- Mars, Holoholopinau. rately as of the hour of the day." Venus, Naholoholo. 2 The ancient inhabitants of Ye- Jupiter, Hoomanalonalo, also Iao men, Arabia, canonised and wor- and Ikaika. shipped their ancestors.— Manual of Saturn, Makulu. the Ancient History of the East, by Mr. Dibble (History of Sandwich Lenormant and Chevallier, vol. ii. Islands, p. 107) says that " Hoapili 3 j n \V a jou, the principal territory was so much in the habit of observing of the Buguis in Celebes, the presi- these that he could at any moment dent or head of the chiefs is styled tell the position of each ; and that he the Arumatua. had heard from others that there was i 28 THE POL YNESIAN RA CE. belief in ghosts and apparitions, whether benignant or malicious ; their implicit faith in all manner of witchcraft, enchantments, sortilege, signs, and omens ; i their practices of incantation, objurgation, and divination — all these vari- ous modes of superstition, more or less current among every race of people, appear to have been a very early inheritance of the Polynesians; but the difficulty is to trace them to their proper sources, either Cushite or Arian, for so many of them seem to have been shared by both alike. Some of them, however, can philologically be re- ferred to an Arian root, and among them I would men- tion — Haw. Lapu, "ghost, apparition, spectre." They were good or bad according to the known personal character of the deceased, whom they were supposed to represent. Lapu-ia, " to be possessed of a spirit, to be haunted." I have only found this word with this application in the Hawaiian. Etymologically it refers to the verb Lapu- lapu (Haw.), " to collect together in little heaps, to pick up, as small sticks for fuel ; " to Rapu (1ST. Zeal.), " to search for ; " to Rapu (Tahit.), " to scratch, squeeze, pinch, stir up, be in confusion ; " Napu, " to be confused, nonplussed ; " to Bavu (Fiji), " to smite, smash, kill." This Lapu and its cognate verbs refer themselves without much difficulty to the Ebhu or Rablm of Vedic mythology, spirits or inferior deities of a benevolent character, and to the Sanskrit root, verb Rabh, " to seize, to take," with its Indo-European congeners, such as Rabies (Lat.), " rage, madness ; " Rhaib (Welsh), " fascination ; " Rheibes, " sorcerer," et al. 1 Another word of the same class is the Hawaiian Mana, " supernatural power, an attribute of the gods, glory, might, intelligence, worship ; " Hoomana, " to worship, adore ; " Lloo-mana-mana, " to bewitch, enchant." Samoan, Mana, " supernatural power ; " Tuu-mana, " to curse, to rejoice in another's misfortunes." Tahitian, Mana, " strength, power, 1 See Appendix, No. VI. Pritet, vol. ii. p. 637, and Burfey's 2 Vide Orig. Ind.-Europ., par A. Sanskrit Dictionary, sub voce. THE POLYNESIAN RACE. 129 influence." Tonga, mana, " thunder, omen, sign." Fiji- mana, "an omen, wonder, miracle," a word used at the closing of a prayer or address to the gods, equivalent to " Amen, so be it." The word connects itself etymolo- gically with the Sanskrit mantra, " prayer, magic formula, incantation, charm ; " the Zend manthra, " incantation against diseases ; " the Greek pavr^, " a prophet, sooth- sayer ; " the Latin moneo, " to remind, instruct, predict ; " monstrum, " whatever is strange and unnatural, a prodigy, a marvel ; " the Irish manadh, " incantation, divination, omen ; " manai, " sorceries, juggling," all which primarily refer themselves to the old Arian root man, " to think, to wish, to mind, to know," and also indicate the early and common application of the word to designate what was marvellous or supernatural, an application dating back to the Arian unity." x Human Sacrifices. The custom of sacrificing human victims goes back to so old a date among the Polynesian family, that it is almost in vain to attempt to define a time for its introduc- tion. However accustomed and callous the Polynesian may have become to human sacrifices, as a means to avert public danger, to appease the gods, or to satisfy the caprices of priests and rulers, yet there may be found in the ancient legends palpable indications that there was a time before that, when human sacrifices were not only not of common occurrence, and an established rule, but were absolutely prohibited. Kapu he kanaka na Kane, " sacred is the man to Kane," — for whom and by whom he was made — was the oldest Polynesian doctrine handed down by tradition, however much it may have been disregarded in after times. The universality of the custom among the leading Polynesian tribes would seem to indicate that it was one of many evil practices brought with them from the Indian Archipelago. But wherever the custom came from, the 1 Vide Orig. Ind.-Europ., par A. Pictet, he. cit. Vol. i. 1 •f 1 30 THE POL YNESIAN RA CE. maxim above referred to remained in the national folklore as a standing, though ineffectual, protest against it, show- ing plainly that at one time, in the far indistinct past, the cruel practice was contrary to the national creed and na- tional mode of thought. With this maxim, then, pointing to a condition of purer creed and simpler manners, how far must we ascend in the past ages and past connections of Polynesian life ere we reach that state of society in which human sacrifices were prohibited, abhorred, and unusual ? We do not find it among the Cushite instruc- tors of the Polynesian ancestors, at least not at the period when the latter may be considered as the pupils of the former. We must then ascend the Ariau line of connec- tion to find the source whence the maxim originated. Speaking on the subject of sacrifices among the Arian people before their separation, M. A. Pictet says : " La comparaison des termes qui se rapportent aux sacrifices semble montrer qu'ils consistaient surtout en libations, mais que Ton immolait aussi certains animaux. Eien n'indique, par contre, que l'effroyable coutume des sacri- fices humains, pratiquee plus tard aux temps de barbarie, ait attriste le culte des ancetres de notre race." x In ancient Greece sacrifices were of the fruits of the earth, and it was originally forbidden to immolate victims, under pains of death. Afterwards animals were sacrificed. Human sacrifice was accounted so barbarous an act by the ancient Greeks, that Lycaon was feigned by the poets to have been turned into a wolf for offering such a sacri- fice to Jupiter. In latter times this custom became more common. 2 Cannibalism. Though the custom of cannibalism among some of the Polynesian tribes, as well as among the Battas of Sumatra and the eastern Dyaks of Borneo, would seem to justify 1 Origines Ind.-Europ., vol. ii. p. 2 Harwood's Grecian Antiquities, 702. pp. 146-51. THE POL YNESIAN RA CE. 131 the inference of a community of origin ; yet I am strongly inclined to believe that it was not a common national trait at the time of the unity of the Polynesian race in the Indian Archipelago, and as such brought with the wandering- Polynesians into the Pacific, but rather of comparatively later date, originating during their sejour among and con- tact with the man eating Fijians, on their first arrival in the Pacific. It is true that among the Marquesans and New Zealanders the custom prevailed extensively, and they claimed a fabulous antiquity for its origin ; but among the Tongans the practice was exceptional with some of their warriors in war-time, and then as a matter of bravado, and avowedly in imitation of their Fijian neigh- bours ; and among the Society Islanders and the Hawaiians the custom never obtained. With these two branches of the Polynesian family, though the practice of cannibalism was not unknown, yet it was looked upon with horror as an exceptional depravity of a few wicked and outlawed persons, who in the ancient legends were loaded with in- famy and exterminated as monsters. 1 There is no refer- ence in the legends of either the Hawaiian or the Society group to a time when cannibalism was a national practice that was afterwards discontinued. The very fact that on occasions of human sacrifices, both on Hawaii and at Tahiti, the left eye of the victim was offered to the presid- ing chief, who made a semblance to eat it, out did not, was, a virtual protest against the custom as being neither ori- ginal nor universal with the Polynesian family, and was merely a seeming concession to a horrid fashion that had 1 Rev. Mr. Dibble, in his History due to the Hawaiians to say that those of the Sandwich Islands, 1843, p. few instances that did exist were 133, &c, refers to the tradition of the looked upon by most of the people cannibal chief Kalo-aikanaka at Hale- with horror and detestation." The mam;, Oahu, Hawaiian group. The extermination of the Halemanu chief tradition was well known in Dibble's and his accomplices, and the infamy time, and is also referred to by to which his memory was consigned Messrs. Tyerman and Bennett, agents in the national traditions, ought to be of the London Mission Society. Dut sufficient refutation of the accusation Dibble expressly states that "the brought against the Hawaiians. practice was not common, and it is 132 THE POLYNESIAN RACE. been adopted by other tribes of the family during their contact and combats with the Fijians. As regards cannibalism among the Battas and Dyaks, the practice among the two peoples differs so widely in occasion, mode of procedure, and other matters, as to pre- clude the idea of a community of origin, and therefore probably arose separately in each of them from local causes and circumstances now unknown ; the more so, as it is not known to have been practised by the other pre- Malay tribes of kindred blood, either on the same islands or in more isolated places. I think it fair, therefore, to conclude that this horrible practice was not an original heirloom brought with the Polynesians from their primitive homes in the Far West, but was adopted subsequently by a few of their tribes under conditions and circumstances now unknown ; and the non-observance and indignant reprobation of the same by two such leading members of the Polynesian family, as the Tahitian and Hawaiian, and, I would fain believe, the Samoan, ought in justice and equity redeem the race from the stigma so lavishly thrown upon it as a whole. 1 On a previous page I have referred to a Hawaiian legend which narrates that a certain chief, called Hawa-ii- loa or Kekowa-o-Hawa-ii, was the first who discovered and settled on the island of Hawaii of the Hawaiian group, and that he called it after his own name. The legend further adds that at that time this group consisted of only the two larger islands, Hawaii and Maui, the other islands having as yet not emerged from the ocean. I referred to the legend to show, by the traditional evidence of the Polynesians, that they came into the Pacific from the Far West, and that by whatever route they came they claimed some continent or large islands in the West as their birth- place. But though the legend unequivocably confirms the pro- position of the Western origin of the Polynesians, it is by 1 Vide Appendix, No. VII. THE POLYNESIAN RACE. 133 no means conclusive that either Hawaii, Sawaii, or Habai in the Pacific was the terminus of that celebrated voyage of discovery ; and for reasons to be found in the legend itself, I am induced to believe that the departure of that voyage was from some part of the coast of the Erythraean Sea, in Southern Arabia, and that its terminus was at Ja"\va in the Sunda group. First, the legend expressly makes Hawaii-loa the seven- teenth generation after the Flood, and the fourth only after that mythical and legendary Twelveship with which the traditional life of so many Eastern peoples begins. He is said to have been born and lived on the eastern coast of Kapa-kapa-ua-a-Kane, "the land where his forefathers dwelt before him," which land was also called ka Aina-kai-Mele- mele-a-Kane, or " the land of the yellow, or handsome sea." But this land bore also the name Hawa-ii-kua-uli- kai-oo, or " Hawaii with the verdant hills and the dotted sea." Under previous considerations, does not the latter epithet coincide with the Erythraean Sea, called by the older Greeks " the sacred wave," and " the coralled bed ? " From analogy and the general idiomacy of the Polynesian language, it becomes highly probable that Kapa-kapa-ua is an old intensitive, duplicated form of the Cushite Zciba, and this derivation would harmonise the old Arabian tradi- tions, which place Paradise in the south-west part of Arabia, with the Hawaiian tradition, which states that after the expulsion from Kalana-i-Hau-ola, the descendants of the first man went eastward and occupied the coasts of Kapa-kapa-ua. Now, from numerous other parts of this and other legends, we learn that Kapa-kapa-ua was a sub- division of the large continent generally called Kaliiki-ku, or Eastern Kahiki, and from other references we infer that it was situated in the western part of that continent, and that to the south of it was a large land or continent called Ku-i-lalo or Honua-ku-i-lalo, " the southern land," renowned ^for its warlike and savage people, while to the west was mother large land or continent called Kahlki-moe, " the 134 THE POLYNESIAN RACE. Western Kahiki." If we now refer to some of the ancient and obsolete Hawaiian names for the North, we find two that arrest our attention in connection with this inquiry. The reader may remember that among the Hawaiian names for the North were Ulu-nui and Mele-mele, and that origin- ally they were names of lands situated to the northward of some former habitat of the Polynesian family, or of those from whom they received their culture, their myths, and a goodly portion of their legends. Now the land of Mele- mele forcibly connects itself with " the Sea of Mele-mele " above referred to, and indicates another land or country or kingdom situated on the shores of the same sea, but to the north of the birthplace of Hawaii-loa. Viewed under that light, and assuming the south-eastern coast of Arabia to be the Kapa-ka/pa-ua of the legend, the name of the other northern land, Ulu-nui, cannot possibly have any other explanation than that of Ur, the city and kingdom of Uruch in ancient Chaldea, at the head of the Persian Gulf. From this coast Hawaii-loa set sail, and steered to the eastward, crossing the ocean called 3foana-7cai-maoJcioki, or "the spotted, many-coloured sea," and also called Moana-kai-popolo, " the blue or dark-green sea." Consider- ing his point of departure, that ocean must have been the Indian Ocean, and the two large islands which he dis- covered can be no other than Sumatra and Java. He called the one after his own name, Haiva-ii, and the other after that of his son Maui. But I have previously shown that the Polynesian Hawa-ii, Sawa-ii, Habai, and the Malaysian Jawa, Djaiva, Ciaiva and Zapa-ge are all referable for their protonom to the Arabian Zaba, the centre and pride of the Cushite Empire, whose commerce, colonies, and con- quests extended from Madagascar to the Moluccas. With these premises it is difficult to conceive that these two islands could have been any of the Polynesian groups, or that whatever might have been the western site of that original Kapa-kapa-ua, the navigator of those days could THE POL YNESIAN RA CE. 135 have crossed the Pacific Ocean in an easterly direction within the belt of the trade-winds, and not have encountered any of its numerous islands and Atoll groups before reach- ing either of the three groups bearing the name of Hawaii. And if this, by some singular combination of fortuitous circumstances, could have been done once, it is hardly credible that it could have been repeated often. Yet the legend makes no mention of any such landfall, and Hawaii- loa is represented as having made several voyages after- wards between Kapa-kapa-ua and Hawa-ii, as well as other voyages to " the extreme south," i Tea mole ha honua, and to some other Western land, not Kapa-kapa-ua, where dwelt a " people with turned-up eyes," Lahui maka-lilio, and travelling over this land to the northward and west- ward, he came to the country called Kua-heiva-lieiua ; a very large country or continent. Eeturning from this country, he is said to have brought with him two white men, poe keokeo kane, whom he married on his return home to Hawaiian women. It would be interesting to know who these people witli turned-up eyes or drawn-up eyes, living on a continent to the west of the Sunda Isles, may have been. They certainly were not Chinese, Japanese, or any of the Mongol families of men. At first view the legend would seem to give strength to the opinion that Hawaii-loa actually had discovered and settled on the Hawaiian group ; for, know- ing no other oblique-eyed people than the Chinese and their varieties, they could not have been reached by a westerly voyage unless the point of departure had been somewhere in the Pacific. But it is fair to question whether the Chinese and their varieties were the only oblique-eyed people in the world. With the Sunda Isles as a point of departure, and a westerly course, the coast of Africa is the natural landfall. What people inhabited that coast in times so far back as those of Hawaii-loa ? Denon, in his " Voyage en Egypte," describing from the ancient paintings in the temples and tombs of Egypt, 136 THE POL YNESIAN RA CE. says : " The female forms resembled figures of beautiful women of the present day ; round and voluptuous ; a small nose, the eyes long, half-shut, and turned, up at the outer angle, 1 like those of all persons whose sight is habi- tually fatigued by the burning heat of the sun, or the dazzling whiteness of snow ; the cheeks round and rather thick, lips full, mouth large, but cheerful and smiling." With the exception of the " turned-up eyes," a Polynesian beauty might have sat for the picture. In some of Dr. Livingstone's letters, published in the Proceedings of the Eoyal Geographical Society, November 8, 1869, speaking of the people of Rua, on the west side of Lake Tanganyika, he says that they are said to live in rock-excavations, that " the people are very dark, well made, and outer angle of eyes slanting upwards." Here, then, we have the testimony of the ancient Egyptians themselves, that " turned-up eyes " was not an individual specialty or deformity, but so common a char- acteristic of their people — the women at least — as to be preserved and faithfully copied on their pictorial records. As the women of ancient Egypt enjoyed a degree of social and political consideration, of which their modern Oriental sisters are sadly deprived, it is improbable that any painter of that time would have dared to thus caricature his countrywomen. The trait, then, was a national one, and it is so designated in the Hawaiian legend. As regards the other people " with outer angle of eyes slanting upward," though now living in the centre of Africa, no man acquainted with the migrations of peoples and the changes of empire would venture, without positive proof in confirmation, to assert that at some period in the past unrecorded history of Africa they did not occupy the eastern coast from Abyssinia southward, and thus have been the people of whom the legend speaks. Lahui maka- lilio, " people with eyes turned up or drawn up," were, then, anciently found on the African coast, and could be 1 The exact Hawaiian expression, Maka-lilio. THE POL YNESIAN RA CE. 137 reached by a westerly voyage from the Sunda Isles : and thus the Hawaiian legend becomes consistent with itself, and with historical facts independent of it. Historically considered, I am inclined to think that the legend of Hawaii-loa represents the adventures and achieve- ments of several persons, partly pure Cushites, partly Cushite- Polynesians, which, as ages elapsed, and the indi- viduality of the actor -retreated in the background, while the echo of his deeds was caught up by successive genera- tions, were finally ascribed to some central figure who thus became the traditional hero not only of his own time, but also of times anterior as well as posterior to his actual existence. While the one set of legends shows the voyages and intercourse of the early Cushites with the countries and archipels about the Indian Ocean, the other set of legends shows the intercourse and voyages of the earlier Polynesians between the groups of the Pacific. But to find the former set of legends in the possession of the latter race of people argues a connection, political and social, if not ethnic, and to some extent probably both, so intimate, yet so far antecedent, that the latter had really come to identify themselves with the former, and appropriate to their own proper heroes the legends brought them by the others. In much later times the same process was re- peated, when the Hawaiian group was overrun by princely adventurers from the South Polynesian groups, who incor- porated their own legends and their own version of common legends on the Hawaiian folklore, and interpolated their own heroes on the Hawaiian genealogies. While, therefore, the presence of the Cushite element in the Polynesian race, in its legends, culture, and creeds, and to no inconsiderable extent in its blood, and at all times an element of supremacy, cannot be ignored in estimating the origin of this race ; yet the Arian, mostly pre-Vedic, affinities of the language, with certain Zend proclivities in the phonetic values of some letters in the majority of the Polynesian dialects, are a stubborn fact, pointing either to an absolute ethnic relationship, or to a period of subjection 1 38 THE POL YNESIAN RA CE. sufficiently prolonged to completely change or materially modify the older tongue of the Polynesian progenitors, whatever that may have been. But still a third element bespeaks its presence with potent and incontrovertible force — the dark colour and the black eyes — and points to an intimate connection, or rather complete fusion with the brown or Dravidian race ; whether in or out of India history and tradition are equally silent. The order in which these elements contributed to the formation of the Polynesian race is not so patent, nor yet at this distance of time, hardly possible to define. Neither history nor tradition records any invasion or immigration into India by the Arian race before the Vedic-Arians crossed the Sarasvati and came in immediate and perma- nent contact with the brown Dravidian race. And yet the occupation of Hapta-Hindu was an invasion of India, and must have been effected by an entire displacement of the pre-existing Dravidians between the Sutlej and the Indus. The echoes of that event had long died out before the Vedic hymns were composed or the laws of Menu compiled. May not then some other hordes of the Arian family, and more akin to the Iranian division, have passed into India along its western coast from Beluchistan, through Cutch and Guzzerat, and mingling earlier with the Dravidian tribes on the way, and with less religious venom, have changed their colour while they retained their language, as their Vedic brethren did after they commenced settling on the banks of the Ganges ? How long they remained in these new habitats, when or by whom displaced, how or by what route arrived at the Indian Archipelago, their struggles with, their conquest and expulsion of the Papuans, are all blank leaves in their history. A few local names, common to the west of India and some parts of Polynesia, would seem to indicate that the stream of migration had set southward through Western Deccan. To what extent, if any, they had come in contact with Cushite culture, com- merce, and colonies, before they arrived in the Sunda Isles, no traces are left to indicate with any reliable precision. THE POL YNESIAN RA CE. 1 39 But here, if not before, they became subject to the direct enterprise and influence of the Arabian Cushites, who here as elsewhere established their supremacy, introduced their institutions, customs, and creed, and mixing freely with the subject race, identified themselves so thoroughly with it as to substitute their own cosmogony, genealogies, and legends for whatever memories and whatever notions may have previously been entertained. In some such way alone, it seems to me, can we explain the composite character of the Polynesian race. We find the remains of Cushite culture and creed; we find the Cushite-Arabian type abundantly cropping out ; 1 we find the dark colour of the Dravidian race, varying from lightest olive to darkest brown ; we find the language fundamentally Arian, but of a form far older than the oldest written remains. These phenomena must have a natural solution, based upon his- torical sequences, and, under correction from those of greater knowledge, I have presented mine. I will now refer to the opinion entertained by many authors, that the Polynesian language is an offshoot from and a corrupted dialect of the now widespread Malay, and that the Polynesian tribes are simply colonists from the Malay stock. The first explorers in this direction of ethnic inquiry, finding that a large number of Polynesian words bore a family likeness to the Malay language, naturally enough concluded that the former was a descendant or an importation from the latter ; and the marked contrast in which the Papuan race stood to both Malay and Polynesian probably strengthened the illusion of an ethnic connection between the two latter peoples. When, afterwards, such philologists as Fr. Bopp discovered and established the connection between what he called " the Malayo- 1 The Asiatic Ethiopians or Cushites Gallas and Abyssinians, as well as the were, according to Herodotus (vii. Cha'b and Montefik Arabs and the 70), of a dark complexion, but with Belooches. — The Five Great Mon- straight hair, not curly like the archies of the Ancient Eastern World, African Ethiopians. Kawlinson com- G. Rawlinson, vol. i. p. 52. Loudon, pares them in tint- dark red, brown, 1871. or copper colour— to the modern 140 THE POLYNESIAN RACE. Polynesian and the Indo-European" tongues, the conclu- sion seemed unanswerable that the Malays had received their Sanskrit words from the Hindus, and the Polyne- sians from the Malays. In Bopp's time the knowledge of the Polynesian was too imperfect to venture" a demurrer, and none was made. All honour to Bopp, however, for establishing the kindred of the Polynesian, as well as the Malay, with the Indo-European family of speech ; but the degree of kindred of the two former to the latter, and their relationship between themselves, as intimated by Bopp, may well be called in question at this day. I think that few people who are competent to form an opinion on the subject, and have seen and observed the Malays and Poly- nesians — not by individual exceptions, but by masses, and at their homes — will now maintain that they are ethni- cally descended from the same race. They have both, doubtless, passed through the Dravidian crucible, but the Polynesian entered it from the north-west, and the Malay, I think it will yet be found, entered it from the north- east. The Polynesians brought some primordial form of the Arian language with them when they came to India, as their mother-tongue ; the Malays, whatever may have been their aboriginal tongue, acquired their connection with, and knowledge of, the Arian in its Sanskrit or Sanskritoid form, during their sejour in India, before they followed the Polynesians into the Indian Archipelago. Thus, while a large number of words are common to both, yet the form of these words in the Polynesian is by far the most archaic, the most conformable to what we may conceive to have been the earlier form of the Arian lan^uao-e, ere it acquired the Vedic- Sanskrit or Zend developments. 1 Malay and Javanese historians and traditions give themselves no higher origin in the Sunda Isles than the first century of the present era. One account represents them as coming from India, under the leadership of a son 1 The original elements of the Arian Miiller, Lectures on the Science of language consisted of open syllables Language, 2d series, p. 192. Lon- of one consonant, followed by one don, 1864. vowel, or of a single vowel. — M. THE P OL I 'NEST A N RA CE. 141 of the Rajah Souren, the founder of Bisnagour, and estab- lishing their empire at Palembang, Sumatra. 1 Another account says that they came from the Telinga country, in north-east Deccan, with one Tritestra, or Aji-Saka, and established their empire in Jawa, at the foot of Mount Semiru. He introduced Brahmin culture, customs, and creed. He found the country inhabited by Eakshasas, which in a Brahmin mouth meant the infidel, non-Brah- min, Cusliitised inhabitants of India and the islands. With these people several and bitter wars were had before the new settlers established themselves. That these Eak- shasas were the Polynesians, who at some previous period had displaced the Papuans, and were now displaced in their turn by the rising Malay power, I think there can be little doubt. It is manifestly wrong, therefore, to class the Polynesians as colonists, or their language as a dialect, of the Malay. There is nothing in the Polynesian lan- guage to show that they are later than the Vedic develop- ment of the Arian race, probably much older ; there is nothing in the Malay to show that they are older than the Sanskrit development of that same race. The one is the genuine stock, the other an engraftment upon it. In connection with this subject of relation, ethnic or otherwise, between the Malay and Polynesian peoples, I ought not to pass over in silence an essay by Eev. S. J. Whitmee, an English missionary of many years' residence in the Samoan group, and published in the " Contemporary Eeview," February 1873, pp. 389-407. In this essay Mr. Whitmee, with much justice and correctness, refutes the assertion of so eminent a naturalist and traveller as Mr. A. R. Wallace, who in his " Malay Archipelago," chap, xl., declares that the Polynesian race is merely " a modifica- tion of the Papuan race, superinduced by an admixture of Malay or some light-coloured Mongol element, the Papuan, however, largely predominating physically, mentally, and 1 "Monde Maritime," par G. A. extended their conquests to the pen- Walkenaer. About 1159-60 A.c. the insula of Malacca, thus called Oud- Palembang chiefs invaded Jawa, and jong-tanah. — Ibid. 1 42 THE POL YNESIA N RA CE. morally ; but that such admixture probably occurred at such a remote period as, through the lapse of ages, to have become a permanent type." But in avoiding the Scylla of Mr. "Wallace and the Papuan theory, Mr. Whitmee has fallen upon the Charybdis of Bopp and the Malay theorists. Mr. Whitmee refuses credence to Mr. Wallace's classifica- tion of the Polynesians, but he believes him explicitly in his classification of the Malays. Mr. Wallace sweeps the almost entire population of the Indian Archipelago into one ethnic net, and calls his catch the Malay race. The Malays proper, the Javanese, the Battas, the Dyaks, the Buguis, the Suluans, the Moluccans and Buruans, the coast inhabitants and the mountaineers, are all Malays in his generalisation, and their different languages are but diffe- rent dialects of the Malay. Such generalisation is hasty, and cannot but be misleading. As well class the Magyar and the Sclave, the Finn and the Swede, in the same ethnic compartments as to crowd the Batta and the Dyak,the Pulo- Nias man, the Bugui, the Buruan, Saouan, and numerous other original tribes still surviving in the Archipelago, into the comparatively modern Malay box. The Malays them- selves, with instinctive national consciousness, and, I am tempted to say, conscientiousness, consider the above tribes as ethnically different from themselves, and express that sentiment or consciousness by calling the others Orang- Benua, " men of the country," i.e., " aborigines." Mr. Wal- lace makes no distinction, except that of degree of civilisa- tion, between these latter, whom the Polynesians resemble ethnically, and the Malays, whom, if they (the Polyne- sians) resemble them at all, it is by accident, as a Jew might resemble a Eornan. Misled by Mr. Wallace's gene- ralisation, Mr. Whitmee fails also in making a distinction between the Malays proper and the pre-Malay inhabitants of the Archipel. Hence the summary of Malay words, quoted by Mr. Whitmee as having " Polynesian equivalents very closely resembling them," is not a correct showing in the sense that Mr. Whitmee intends it. Leaving out the numerals, of the thirty-six remaining words in the list of THE POLYNESIAN RACE. 143 Mr. Whitmee, only eleven are Malay words proper, while twenty-five belong to the language of those very pre-Malay tribes, with which and of winch the Polynesians formed an integral part before the intrusion of the Malays ; and of those other eleven Malay words with " Polynesian equi- valents," nine are such equivalents for the very reason that they are Arian words adopted by the Malay. Mr. "Wallace, in the work above quoted, asserts that the Polynesian has a greater physical, mental, and moral re- resemblance to the Papuan than to the Malay. Mr. Whitmee, in the essay referred to, asserts the greater phy- sical, mental, and moral resemblance of the Polynesian to the Malay. Each one argues a kindred of race according to his views. Both cannot possibly be right ; and, from my reading of Polynesian characteristics and Polynesian folklore, they are both wrong. Great similarity of char- acter may be found in peoples of widely different origin, and great dissimilarity of appearance in peoples of known community of race. What can be more similar in char- acter than the Papuan, as described by Mr. Wallace, and the Marquesan of to-day ? And yet the latter is probably a fairer and nearer approach to the primitive Arian type among its descendants in the East than the Celt among its descendants in the West. What can be more dissi- milar in appearance than the light-haired, blue-eyed, high- statured Scandinavian, and the black-haired, dark-eyed, low-statured Bas-Breton ? And yet their community of race is now undisputed. There is, then, something deeper, older, more expressive as a criterion of the consanguinity of peoples than the character, which may be the result of social conditions during long previous ages — tyranny, op- pression, debasement, or the reverse; or of appearance, which may be the result of physical and local conditions and surroundings, operating for ages in a given direction, inland or maritime, alpine or lowland, forest or prairie, fertile or desert, sunny or cold, border people subject to outside influences and mixtures, or people in the interior 144 THE POLYNESIAN RACE. or in isolated situations. This something, this criterion, I hold to be the language of a people. ' It is true that languages perish and languages decay ; but the decay and the extinction are always pari passu with the decay and extinction of the particular people to whom such or such language was the aboriginal mother- tongue. The Gauls, whom Brennus led and whom Csesar conquered, perished with their language during the five centuries of Eoman occupation. The Cushite race and the Cushite language disappeared together from Arabia and the Mesopotamian basin, and Semites filled their places. But the languages preserved of both these peoples prove that the former were of kindred race with the Eomans, and that the latter w r ere of alien race to the Semites. And so, when the English language shall have super- seded the Polynesian in the Pacific groups, the pure Polynesian, as an ethnic branch of the Arian tree, will have vanished from the scene, whether civilised or savage, and be succeeded by an Anglo-Polynesian people, drawing largely at first from the vanishing stock, but gradually becoming quite distinct in character and appearance. And yet the Polynesian language, preserved in books and more or less interspersed in the vernacular of the new people, will attest the originally ethnic kindred of the old stock to the new, and tell the future inquirer how the Polynesian, after various Dravidian and Cushite detours, returned to the primitive Arian type through a fusion with one or more of its Indo-European descendants. 1 TJie Numeral System. In further confirmation of the Polynesian relation to the Arian stock, at a very early period, the numeral systems of both will furnish rather decisive testimony. It is now pretty well established that the more ancient and rude a people is or was, the more limited is or was its 1 See Appendix, No. VIII. THE POLYNESIAN RACE. 145 numeral system. The Australian aborigines to this day do not count beyond three or four. The Dravidian lan- guages exhibit signs, by the composition of their higher numbers, that at one time the range of their numerals was equally limited. The Polynesian language gives undoubted evidence that at one time the people who spoke it did not count beyond four, and that its ideas of higher numbers were expressed by multiples of four. Judging from the Hawaiian dialect, which has preserved so much of the archaic idiomacy of the language, the Polynesians and their pre-Malay congeners in the Indian Archipel evidently counted "one, two, three, four," and that amount called kau- na : was their tally, when the process was repeated again. That the same quaternary system obtained in the Arian family, in early times, is evident not only from the marked relationship between the four first Arian and Polynesian numbers, but the method of counting by fours as a tally still obtains among some of the Arian descendants in Europe. To the personal knowledge of the writer, on the Baltic coast of Sweden small fish, specially herring, are counted by fours, each four being called a hast. The following table will show the relation I am seeking to establish. It is selected from Arian and Polynesian branches ; but there is this to be observed, that while the latter in all probability exhibit the archaic form of the language, the former exhibit a comparatively later and more or less modified form of the same. 1 Professor Fr. Bopp, in his excel- it a substantive ; and the root-word lent work, " Ueher die Verwaudt- kau remains in the Tonga dialect as schaft der Malayisch-Polynesischen a collective noun, indicating "a col- Sprachen mit den Indisch-Europiii- lection," " a gathering together," "a schen " (Berlin, 1841), proposes that body of somethings," and in that the Hawaiian kau-na is derived sense is also used as a plural article from, or abbreviation of, the Java- and a prefix. It i3 more likely, nese Basa-Krima form sa-kavan or therefore, that this Polynesian kaic simply kavan, meaning "four," and and kau-na is the root of, or closely suggests it to be a corruption (Vers- related to, another Javanese word tummelung) of the Sanskrit catvar, kavan, meaning "a herd, a flock," '"four." The Polynesian kau-na, than that it is a descendant from the however, is a contraction of kau-ana, Sanskrit cat-var, through the Java- the participial ending ana rendering nese kavan, four. VOL. I. K be- es 3 g tO g _S S3 S3 S3' - - S3 S3 o „p tc 60 O tcC" S3 P O SI'S O J, 1 O ri Sri ^ri P ce "3 a'S'MSP'E'i S3 31 S3 r- « C S3 OC p,,-j p, Pr-J a jr 1 p ^PO^ ce ' te ce ce es ce ce s3 . P J h .3 • a . e: ^ 4J m 5 -£2 ~ -r* -w m Si5 o 3 jS !> t> |> > t> > ? ^ 7- Id C 2 13 l."" S "3 ce,.-* ce .5 C d-g-da. rs, S3 cerd ^sscecedceces SSB«lBOiH S w S W a •< a s ,s S f P S S ffi S E i S h S S f^ h h h wp fin i-H o eh WB c • OJ • rS <-' Cw P t- c <~ . feO Sj feO dooos3cDdooocoo^ogregpcS3ceceCCO r pO«7g ooooa-<5o6o6ooooao-5o^P-<-<^^-5i4oSc5PH . . ce". . . • to - ,jd; o ce ce' .cicJMc3cec«sdcj - cefi o: c;[ecSccOcs<*oc'S<5cecstecs* .oJ ce* • *^-1 E, ^ eU • ~ -a a ce a a S §*es ce . ag ce ce ^ -a £ a. S S, %£ S 4 "g 3 etS stJrS J pSa^feacecpOw- L HWBHfH<;HHHHBHHHEHHHHHBHHBrHHHHHE-iOs r-. cece' . ce ce ^ : : £ • ta» e ce _ b • .« m -2 „ o • -=e"5 g § e ® C3 fc C o ~ O -c ss, £ s s s ►« e>. o co .^ * — i e o S>a s d CO .gl .H . H "53 -h En •«! CO M JT mmm Butanga. Lep-wonan. Ne. Nena. Ennoi. 1 »nam. Una. Eno. Anim. Anam. Anam. One. 14 Anam. • „S OjN— . 0) O M — .2 MM . i "Sg'SS^.ssaassss rd . ® . ®* o< (0 ft o bl bo :o" r- *. ^ <*- P-* P^-t P- P-t P- CP Ph ^ f=H -|3 i2 :"S . ci . .« c* ^ £. = — ..•ci'TP'cici^g § » eg * a 43 ft." S ft p«g £ 5 c"q o a '"3 "3 "3 "5 "V a"o.£. ** u ■; E ; c !■' i o M e-i as o ©"eh Pi p^ ^ . a »-a -a HiZEHHrHHH&HH . . c3 •' • . • • >SPPOS>»5>> a . g -2 o =f 2 -r . ? -a , . j 2 - d 'SS , 9jS'3doSee > | s ft. s a 1 1 s o H e '■Cl ot o a . yl >^A-^ "^fl* <» «s ™ CO - - r -3 -• = =2 ci > S ro !^ **i C^-^ = 2 b"M o ^ 5 ci tn ^ g O cf a to p;.s^ s H ;t • . » g^i Pi o gs.toj. to H OS ^> g £P ■*■ a t- - a W a^ CO to p^ o a ,3 co-— o ^S g"J O C-Q-2 O 5 J< o o a o Q H O a to a a 1 a p — i a ci - i— > H ni Z to id fa o a — o c3 O — OJ U i: )-. 0) o a r' cTu-. a o "5 '-!■ >. ID o -t- z — ' Q c CI CI - ■_ kI 6 ^f. -' -^ o o "5 c -|a o 1-1 a> ci Pi "o ■+3 o > •r. > s '— 1 i> 3 a^ S o ^> c Efl - 00 s : od .Z o — a ^ o ^ Pm DO — '^ 6 a - c: 43 tv H -g z - o3 a . zz ^_ o - 7t -- s ■.' 7i a c vn ') -1 ^ ST* H 'A a ■"• V- .v •> . V- — I 1 48 THE POL YNESIAN RA CE. Believing, with Professor Bopp, that the origin as well as the cause of resemblance, or of difference, in the first numeral of the different branches of a kindred tongue, must be sought for in those oldest remnants of a once common speech, the definite and indefinite articles and pronouns ; and recognising that within the Arian family at least, with which I have some acquaintance, the primary- expressions "the," "this," "that," "a," "an," either simply or combined, were the original equivalents of " one," as a number, the early connection between the Indo-European and Polynesian is not very difficult to establish. 1. The Polynesian ha, he, ta, te, ha, he, refer imme- diately to a common origin with the Sanskrit ha and sa, the Greek to, ttj, 69, -q, the Gothic tlia, the Latin hi-c, hce-c, ho-c, and evince the same dialectical tendencies. Bopp considers the Sanskrit eha as composed of e, a demonstra- tive pronoun, and ha, an interrogative. But this eha, " one," has a near relation, and one applied to the same uses in the Tagal ica, ca, caca, meaning "the," "this," "that," and also "one." The Tagal and Malgasse isa, " one," refers itself to the Sanskrit esa, " this ; " and even in Sanskrit the single sa remains as a demonstrative article, and the numeral " one " in compound words, as in sa-hart, " ein-mal," " once." To this sa refers itself, without doubt, the second constituent of the Polynesian taha, " one," and is evidently akin to the Gothic obsolete form of ha in compound words, such as ha-ihs, " one-eyed," ha-nfs, " one- handed," et al. That sa was in Archaic times a numeral design atincr " one," and common to the various branches of the Arian stock, is further illustrated by the Javanese siji, " one," contracted from ancient sa-vigi, which itself is an Arian compound of sa, " one," and viga (Sanskrit), " a kernel, a seed ; " also by the Malay sa-tu, contracted from sa-latu, " one stone," and of which the North Celebes (Bolanghitan) dialect so-boto is a better preserved form. The Basa-Krima points to the same archaism in its sa-havan, "four," lit., "one- THE POLYNESIAN RACE. 149 four " or " one-kau," and in sa-dhasa, " ten," lit., " one-ten." The softening of the broad a in sa to i in si of the Polynesian ta-si, is perhaps as old as any of the branches of the Arian tongue. It is found in Buru (Cajeli) si-lei, " one," Buru (Waiapo) um-si-un, " one," in the South Celebes (Salayer) se-dri, " one," in the Salibabo se-mbao, " one," in the Sula Islands hi-a, " one." Instead, therefore, of considering with Professor Bopp that the Hawaiian Jcu-hi and Samoan ta-si are attenuated derivations of the Tonga ta-ha or an older ta-sa, " one," I look upon them as coeval varieties of a once common expression for the numeral " one." It is possible that the Brahui (Beluchistan) numeral asit, " one," in which the t is either a Cushite or Dravidian postfix, concreted afterwards, may have a linguistic affinity to the Polynesian tasi. If so, the initial t must have been dr6pped; and I have no means of knowing the ancient form of the Brahui word. But this same Brahui asit singularly 1 enough recalls to mind the ancient Latin as, the generic name for a unit, and the Greek et?, ais, " one." 2. The relation of the Polynesian number two to the Indo-European branches is self-evident by a glance at the foregoing table. The same dialectical tendency to change d into I or r is manifest in the Polynesian branch as in the Indo-European. 3. The correspondence between the Indo-European tri and the Polynesian torn, toht, is not so palpable on the face of it, thoutrh the radicals of both are identical. The absence of this form in the Persian, and its substitution by sih, would lead to infer that other synonyms may have existed also in the other branches in early times, but are now lost. The Latin form of ter in ternus indicates clearly the ancient connection. 4. The numeral four, however, corresponds remarkably well in all the varieties of both the Indo-European and Polynesian branches. The chat, quat, chit, tct, pet, and fid forms of the former family bespeak unmistakeable relation with the pat, fat, pa, fa, and ha forms of the latter family. 1 50 THE POL YNESIAN RA CE. Professor Bopp, in the work above referred to, seems to consider the Sanskrit form of the numeral four as the original, archaic form, from which all the others, whether in the Indo-European or Polynesian family, have de- scended or deteriorated. And not only this number, but also all the preceding ones, and most of the subsequent ones. Not content with establishing their family relation, he exalts the Sanskrit to the position of a parent and an architype, by which the others are to be measured. He says that the Malayan dialects are less corrupted or muti- lated (verstummelt) from this architype, this mother-form, than the Polynesian. He may be right in one sense ; for whatever there is in common between the Malay and the Arian, the Malay probably derived from the Sanskrit or the Sanskritoid dialects of upper India ; whereas the Poly- nesian owes nothing to the Sanskrit, properly so called, having left the homestead of the Arian race loncc ases before the Sanskrit, Zend, or other European sisters had assumed to so large extent those trappings of inflections and those habits of elision, by which these younger branches of the ancient stock now mutually recognise each other, however far apart their lot in after-life was thrown. But to return to the numeral under consideration. Is it at all credible, simply on philological grounds, that the first conception of the number four by the ancient Arians should have found expression in so long and complicated a word as chatvar ? Disclaiming any profound knowledge of the Sanskrit, yet I make bold to say that chatvar or chatvaras is a compound word, and hence comparatively modern, when treating of archaic forms of speech. I take it to be composed of chat, or its dialectical variations in quat, pet, pat, tet, fid, original expressions of the conception four, and var, expressing a revolution, a cycle, a total, a tally. With that chat, and pat, the Polynesian pa, fa, and ha, stand in intimate relation ; but inasmuch as the latter never shared in the final consonant which distinguishes THE POL YNESIAN RA CE. 1 5 r the articulation of the word by all the other branches, it is reasonable to infer that the Polynesians had separated from the mother stock before consonantal endings, imply- ing a subsequent compound, had been introduced to give greater variety of meaning and definiteness of expression to words from the same primitive root. 1 Up to this number four it is thus evident, I think, that the Indo-European and the Polynesian numerals were derived from the same mother-tongue, and that the Poly- nesian manner of expressing them was the more ancient ; but it is also evident that, before the Arian stock had passed beyond the quaternary system of counting, one branch — the Polynesian — had also passed beyond the in- fluences, associations, and conditions under which the remaining branches built up, and gave expression to, the higher numerals from four upwards. Under what conditions of civilisation, or from what ethnic relations or neighbourhoods, the Polynesian family adopted the higher numerals from four to ten, is a ques- tion I have not had sufficient means within my reach — if the means exist at all — satisfactorily and positively to solve. There are several reasons which incline me to believe that they were adopted after the family had left the Arian stock, and while in course of amalgamation with the Dravidian in Upper India, or on the Iranian sea- 1 That the final t in chat, tet, quat, lolo, Bum and Savu forms of the &c., was not an original radical in pre-Malay dialects — that this final I the Arian expression for four may, I or d was not an essential, integral think, be inferred from its absence part of the word. I believe it is now in the Anglo-Saxon and Armorican generally understood that the Indo- dialects of the Indo-European family ; European varieties did not derive one the former having fe-o-ver, the latter from the other, or all from the San- pe-var. That two cognate branches skrit ; but that they were, so far back of the same lineage, coming down as knowledge can trace them, con- abreast from the hoariest antiquity, temporary provincialisms, so to say, should have differed, and continued of a common tongue, of whose an- to differ, in the pronunciation of the tique, if not original forms of speech, number four, shows to me — especially some were preserved with greater in view of the entire Polynesian fidelity in one dialect, others in usage, and of the Malgasse, Pulo- others. Nias, Lampong, Bugui, Tidore, Gi- 1 5 2 THE POL YNESIAN RA CE. board, along the Persian Gulf. With all due deference for so industrious and capable workers in philological mines as Bopp and Logan, it is far from obvious that these higher numerals are either reduplications or combinations of the lower numerals existing in the Polynesian family proper, or among its Arian cousins. The universality of their adoption, from Easter Island to Madagascar, and the incredibly small difference in pronunciation, all things considered, amount almost to proof positive that they were introduced into the language, such as they now are, by a people of higher civilisation than the Arians pos- sessed when the Polynesians left them, and while the latter were yet a compact body with small dialectical tendencies, and those of no well defined development. And that people, whether in India, Beluchistan, or further on, was doubtless an early variety of the Dravidian stock, already civilised, or in course of civilisation, by the all- pervading Cushites of the ancient Arabian or Chaldean times. 5. The numeral five, lima, rima, nima, is a purely indigenous Polynesian word, though not without Arian connections and analogies. As soon as the family passed beyond the quaternary system of counting, the hand — lima — became the natural signification and expression for the totality of the five fingers or of a tally. That this word appears exclusively in the Polynesian family as a numeral for five, and not in any of the Indo-European languages, is to me another proof that the former separated from the latter before the quinary system was adopted. The higher Polynesian numerals, six, seven, eight, nine, point strongly to a Dravidian formation grafted on to the Polynesian language, but by what branch of the Dravidian stock the distinctive initial words of those numerals were employed, I am unable to point out, if they are at all derived from any Dravidian source. It would seem as if the Polynesians, after having perfected the quinary system from their own vocabulary, had adopted the terms of a THE POLYNESIAN RACE. 153 foreign vocabulary, already made to hand and more or less corrupted, for the higher numbers. 6. The Polynesian terms for " six," ono, hono, ene, henc, unu, and una, appear to connect themselves with the South Dra vidian forms of the term for "one," such as on-ru, on-du, un-di, though they possibly may refer to one of the early Arian synonyms for " one," which the Latins in the West preserved in iinus, and the Dyak-Idaans of the East retained in uni as expressions for " one." By the time the Polynesians adopted the quinary system, they were already beyond the influence and associations of their Arian con- geners; and having employed their own, the Arian, language in counting one set of fives, the contents of one hand, they employed a foreign language — that of the Dravidians, with whom at that time they probably were in process of amalgamation — to count the next set of numbers, or at least to contribute greatly to their formation. The Gilolo (Galela) and Tidore moi and remoi, "one," strongly point to the North Dravidian (kol) formation of moi, "one." Whether the Polynesian mua, " the first," and muli, " the last," refer themselves, as Bopp is inclined to prefer, to the Sanskrit inula, " root, beginning," or to the Dravidian numeral formation in mo, mu, I will not presume to decide. That the Greek has preserved the root of this word in mounos, monos, " one," " only," " alone," would seem to strengthen the Arian view of the question. But the Sanskrit mula has a well-defined affinity in the Polynesian (Hawaiian) mole, " root, foundation, bottom," as well as in the Latin moles. Professor Bopp, in the work often referred to, thinks the Polynesian forms of hene, fene, ene, for " six," and the Indo-Nesian forms in anam, nencm, &c, are cor- rupted derivations from the Sanskrit (gen.) sannam, a modification of sah-nam from an original sas, " six." This conclusion is etymologically possible, and the analogy of the Latin sent (distrib. num.) " six," would be even stronger and more direct to the Polynesian hene, fene, ene, could it be proven that the Polynesian is only a corrupted form of 1 54 THE POL YNESIAN RA CE. Malay, and the Malay a corrupted form of Sanskrit. But so far all data, historical, traditional, and linguistic, lead in a contrary direction, and thus render Bopp's theory and reasoning untenable, though at the time he wrote those data were not collected. 7. The Polynesian number " seven," fitu, hiku, &c, the Indo-lSTesian pitu, &c, is really more difficult to class than the number six. Professor Bopp, according to his theory, considers the Polynesian forms as a corruption (" Verstiim- melung und Schwachung ") of the Sanskrit sapta, that the first syllable sa has been lost, that the last vowel a has been softened to u or 0, and that to accommodate the Polynesian idiom a light vowel as i has been inserted between the original p and t in the last syllable pta. This is also etymologically possible, but is it historically so ? Professor Buschmann, according to Bopp, loc. cit., con- siders the terms pitu, fitu, hiku, as of purely Polynesian origin, and home-made, so to say. He refers the first syllable to a weakening of the Polynesian pa, fa, ha, "four," and the second syllable to a contraction of the Polynesian tolu, "three," thus assuming that the whole word originally was fa-tolu, 4 + 3 = 7. This reference is ingenious and plausible ; the more so as it starts from a quaternary basis, and builds up the higher numbers on that. But,.. unfortunately for this hypothesis, there is no instance within the whole range of the Polynesian lan- guage, from Madagascar to Easter Island, of any dialectical forms mfa-tu, fe-tu, fa-tolu, fe-tolu, ovfi-tohc, or the same forms commencing with p or h, or with the initial h dropped, as often is the case among the Polynesian dialects. Had the pitu, fitu, been of Polynesian origin, some signs of its gradual corruption from an original fa-tolu to the present pitu, fitu, hiku, itu, could not fail to be found in some one of so many widely scattered and long isolated dialects. Were the word of Polynesian formation, it must have been adopted as the numeral " seven," and the cor- ruption from fa-tolu to fitu taken place, while the Poly- THE POL YNESIAN RA CE. 155 nesians yet were a unit, a comparatively compact body, and before their dispersion East and West over the Indian Ocean. I have attempted to show, however, in previous pages, that, at that period of their existence, the Poly- nesians lived with and among the Dravidian race of India ; that, when they separated from their Arian connections, their system of counting was quaternary ; that the num- ber " five " was only adopted after such separation, in some intermediate stage, inasmuch as their name for " five," though an Arian word, was not the word adopted to express that number by the remaining unbroken tribes of the Arian family ; and that they commenced their numeration of the higher numerals with a decidedly Dravidian word, and in all probability continued the scale upward to " ten " from the same linguistic formation, though I am unable to state the particular Dravidian dia- lect from which the word in question was borrowed. It has a Dravidian postfix, but whether the radical pi is Dravi- dian or Cushite, I am unable to say ; it is not Polynesian. 8. The Polynesian walu, varu, aru, " eight," is doubt- less a foreign word and incorporated as such in the language. It has no Arian affinities, and cannot well, as Professor Bopp endeavours to do, be decomposed into the same elements as the Malay delapan. It has its own radson d'etre and connects itself either as an ^amplification of an ancient Dravidian binary term in bar, var, or, which I am more inclined to think, it is identical with the third person plural of the Dravidian pronoun, aru, avaru varu, " they." The Dravidian connection with the Polynesians when these higher numerals were introduced in the language is further evidenced by the very exceptions to the general term. Thus the Gilolo (Galela) itu-pangi, the Bum (Waiapo) et-rua, " eight," refer plainly to the Dravidian (Tamil) ettu, yettu, " eight," while the ru and tit postfixes in all are distinctively Dravidian family marks that should not be hastily set aside in determining the origin of these words. 1 5 6 THE POL YNESIAN RA CE. g. The Polynesian number "nine," siwa, hitua, iiva, with Indo-Nesian varieties in sia, sio, siau, siam, siwer, I confess myself unable, from my limited philological resources, to trace satisfactorily to its origin. I cannot entertain the idea of Professor Bopp that siwa is a mutila- tion and corruption of the Malay sambilan. It is the same word in Madagascar and in Easter Island, and must have existed in the Polynesian language ages before it came in contact with the Malay. No Polynesian or pre- Malay tribe in the Archipelago comes in any measure near the Malay sambilan in its appellation of the number "nine." Even the Sunda dialect, which has been so greatly Malayified, has not adopted the word, but adheres to what was probably the older Malay formation, and calls " nine " salapan, as it calls " eight " dalapan. The Indo- nesian synonyms for "nine," Javanese sanga, Bugui hassera, and the Dyak varieties petan, paih, piri, iletean, throw no light, so far as I know, on the language from which they all were derived. There is no apparent reason why the Polynesians, after having adopted " six," " seven," and " eight," from a foreign tongue, should have reverted to their own language for the formation of the number " nine." And yet the change of the first vowel from i to a in the Savu saio, and to u in the Pulo-Nias su-wa, might lead to the inference that the first syllable repre- sented the ancient Arian sa, si, demonstrative pronoun, and number one, while the second syllable wa was an abbreviation of the already adopted wain; the more so, as in Mindanao and the Solo Archipel nine is si-au, and in the Sangvir Islands it is ka-si-au, showing a lapse of the middle I not uncommon in some dialects like the Marquesan. And this inference requires additional weight from the Teor si-wcr, where the liquid is retained, though the final u is omitted. The word would then represent i, 8, or 1+8 = 9. 10. The Polynesian "ten," pulu,fulu, Miru, with vary- ing prefixes of sa, se, san, sanga, singa, tanga, ana, ono, THE POLYNESIAN RACE. 157 lionga, is doubtless a genuine Polynesian word. Its literal and archaic meaning in all the dialects is that of " feathers, hair, wool." When the denary system was adopted, " ten" "became a new tally, and was expressed by a word indicat- ing a multitude, as may be seen from its synonym umi lit., "the beard." 1 But the Polynesian puru,fulu, hum is evidently near kindred to the Sanskrit (Ved.) piXru, " much, many, exceeding," though I am not aware that in any of the Indo-European branches it was ever used to express any definite quantity. The concrete sense of " feathers, hair," &c, must be very ancient, however, inas- much as the denary system must have been adopted before the Polynesians occupied the Indian Archipelago, and while yet subject to, or mingling with, the Dra vidian tribes of India, and before the Vedic Arians had crossed the Indus. As to the prefix sa, se, sanga, singa, ana, Kongo, &c, I consider the two first as the old Arian numeral " one," thus making the sa- or se-pula equal to " one ten." Sanga and its varieties is composed of the same numeral sa, and the Polynesian plural prefix na, thus making the word pro- perly written as sa-nga-pulu ; and in that way the original plural sense of nga-pulu, " the feathers," crops out under its later conventional meaning of " ten." That the quaternary system of counting continued long after the denary system was adopted among the Polyne- sians is evidenced in the Hawaiian dialect. There " ten fours " were called a kana-ha or an iako or a ka'au ~ 40 ; " ten forties " were called a laic = 400 ; ten " lau " were one mano = 4000 ; ten " mano " were one kini = 40,000 ; and ten " kini " were one lehu = 400,000, the highest num- ber known to them. The expressions for a single hundred or a single thousand were unknown to the Hawaiians until 1 In Hawaiian, umi, "ten;" umi- Paumotu, id., Fiji kumi, "beard;"rrc- umi, "beard." Samoan umi, " ten fa- kumi, "bearded." The Ceram hutu thorns." Tahitian, umi-umi, "beard;" and hutu-sa, " ten," refer themselves umi, "ten fathoms." Marquesan, to the Tidore and Galela hutu, kumi-kumi, "beard." Maugarewa, "hair." 153 THE POLYNESIAN RACE. the discovery of the group by Cook, and subsequent inter- course with foreigners. Some of these words, like the puhc and umi and hutu, have preserved their original meanings alongside of the later conventional and numeral designa- tions. Thus lau still means " the leaves of trees " in the Hawaiian, Samoan, Marquesan, New Zealand, and Fiji dialects ; in the Tonga contracted to lo ; and still retains the primary sense in composites, as lau-ulu (Samoan), " hair," lit., " the leaves of the head," as lo-nutu (Tonga), " lips," lit., " the leaves of the mouth," et at. In the Indo- nesian dialects, the same word, with the same meaning, still remains in the Ceram lau, laun ; Malay daun, " leaf," and others. Thus lehu, under varying dialectical forms of rehu, reu, lefu, efu, means " ashes, dust ; " in Fiji levu, " large, numerous, great," already a secondary meaning ; in the Amblaw and Javanese ami, in Malay liahu, Sunda lebu, also mean ashes. In the Hawaiian and other Poly- nesian dialects, the duplicated form lehu-lehu means "many, numerous, and an indefinite large number." In none of the Polynesian, however, except the Hawaiian, has this word assumed the conventional sense of a definite number; but in Indonesia we find Malgasse ar-rivu, iooo; Tagal libo, id., Malay sa-ribu, id., Sunda sa-rivu, id., Javan- ese s6vu, id., while in the island of Mysol lafu only repre- sents " ten." Thus the Hawaiian kini, beside the definite number, means " an indefinitely great number, a retinue of persons, a. following ; " Marquesan tini, " many, innumer- able ; " Tahitian po-tini-tini, " an indefinitely large num- ber;" while in Fiji tini, and in Ceram (Camar.) tinein means only "ten." And thus, lastly, mano, which in Hawaiian represents 4000, and in Tonga 10,000, has a primary meaning of " many, numerous, an indefinite mul- titude." \ THE POL YNESIA N RACE. 159 I think the facts collected, in the foregoing attempt to satisfactorily solve the question of the Polynesian origin, will warrant the conclusion that the various branches of that family, from New Zealand to the Hawaiian group, and from Easter Island to the outlying eastern portion of the Fiji Archipel, are descended from a people that was agnate to, but far older than, the Vedic family of the Arian race ; that it entered India before these Vedic Avians ; that there it underwent a mixture with the Dravidian race, which, as in the case of the Vedic Arians themselves, has permanently affected its complexion ; that there also, in greater or less degree, it became moulded to the Cushite- Arabian civilisation of that time ; that, whether driven out of India by force, or voluntarily leaving for colonising pur- poses, it established itself in the Indian Archipelago at an early period, and spread itself from Sumatra to Timor and Luzon ; that here the Cushite influence became paramount to such a degree as to completely engraft its own legends, myths, culte, and partially institutions, upon the folklore and customs of the Polynesians ; that it was followed into this archipelago by Brahmanised or Buddhist Ario-Dra- vidians from the eastern coasts of Deccan, with a probably strong Burmah-Tibetan admixture, who in their turn, but after protracted struggles, obtained the ascendancy, and drove the Polynesians to the mountain ranges and the in- terior of the larger islands, or compelled them to leave altogether; that no particular time can be assigned for leaving the Indian Archipelago and pushing into the Pacific — it may have occurred centuries before the present era, but was certainly not later than about the first cen- tury of it ; that the diversity of features and complexion in the Polynesian family — the frequently broad forehead, Eoman nose, light olive complexion, wavy and sometimes ruddy hair — attest as much its Arian descent and Cushite- connection, as its darker colour, its spreading nostrils, and its black eyes attest its mixture with the Dravidian race ; 1 60 THE POL YNESIAN RA CE. and, finally, that if the present Hindu is a Vedic descen- dant, the Polynesian is & fortiori a Vedic ancestor. In estimating the time of arrival of the Polynesian family in the Pacific, I have been guided almost wholly by their own genealogies and traditions. No other his- tory throws any light on their departure, their passage, or their arrival. "When once they entered the Pacific, they were lost, as it were, and forgotten. Among the many legends and traditions still existing in the Asiatic Archi- pelago of large and extensive migrations, occasioned by civil feuds, foreign invasion, or the desolating effects of earthquakes and eruptions, of pestilence and famine, it is impossible to fix upon any one as the one, or the ones, that pushed their fortunes into the Pacific. In the Java and Bugui legends we are told that such or such a prince left with a thousand followers or more to escape oppres- sion, or evil of some kind or other, or to found a new home in a better land : and while not a few of those who found that home on the cis-Papuan side were duly re- ported in song and saga, not one of those who went beyond the Gilolo passage or Torres Straits, and found that home on the trans-Papuan side, remains upon the memory of those from whom they separated, or of those by whom they were displaced. But the Polynesian legends and genealogies themselves, bearing upon this point, are extremely obscure, confused, and contradictory, and consequently difficult to bring into chronological order. The generally-received genealogies of most of the leading Polynesian groups lead up to Wakea, Atea, or Malcea, and his wife Papa, as the earliest progenitors, the first chiefs of their respective groups. Other genealogies, like that of " Kumuhonua," bring the line of Hawaiian chiefs on Hawaiian soil up to Hawaii-loa, who is said to have first discovered and settled on these THE POL YNESIA N RACE. 1 6 1 islands while on a fishing excursion, sailing east from his native home. Another, a Tahitian legend, goes also back of Wakea to Til, whom it makes the first settler or dis- coverer of their group, and whom some Hawaiian legends claim as a brother of Hawaii-loa. But I have shown that the Hawaii-loa legend is probably the concentration of several originally distinct legends upon one person, and that if he of whom the legend speaks was the first dis- coverer and settler of the Hawaiian group, his place on the genealogy is a fatal and irreconcilable anachronism. More- over, according to the legend, the Hawaiian group at the time of its discovery by Hawaii-loa consisted only of the two islands of Hawaii and Maui, the other islands of the group not having yet arisen from the sea ; yet, before the death of the discoverer, they are not only made to rise up from the ocean, but to become wooded, watered, and fertile, and to have been allotted as the homes and principalities of his other children. The Wakea period is almost equally unsatisfactory and difficult a starting-point in computing the age of the Polynesian race in the Pacific. Between the Hawaiian genealogies alone, which lead back to Wakea from the present time, there is a difference between fifty- seven generations on the shortest, and seventy on the longest, a difference representing a period of about 400 years. There may be lacunas on the shorter line ; I am morally sure that there are interpolations on the longer. The latter would represent the year 230 B.C. as a medium year, the former the year 160 A.C. Yet admitting the high anti- quity of the Wakea and Papa legends, it is obvious from the legends themselves that the islands now held by the Poly- nesian race were already peopled in the time of Wakea, and that too by people of his own race and kindred. When or how that people arrived is now an absolute blank. They may have been waifs, they may have been colonists from the eastern fringe of the Polynesian area in the Asiatic Archipelago, but of their traditions or their descent not a vestige remains. The Wakea period eclipsed Vol. 1. l 1 6 2 THE POL YNESIAN RA CE. or obscured all previous movements or migrations in an easterly direction. The Wakean era, however, was undoubtedly one of great disturbance, displacement, and change in the ancient Poly- nesian homesteads. The very fact that so mauy of the principal tribes have retained his legend, though under different forms, and have attempted to localise him and his wife on their own groups, proves to me that he was anterior to, or at least contemporary with, some great popu- lar movement preceding or attending the first considerable exodus into the Pacific, the memory of which was linked to his name, and thus handed down to posterity. His wars with Lihaula, his brother ; his wars with Kaneia-Kumu- honua, in which he was conquered, driven out of the land and fled over the sea, though he is said to have recovered his kingdom afterwards ; his changes in the religious and social institutions of the people, or which have been ascribed to him ; all point to an area of unrest, tribal if not ethnic displacement and material modifications among the Poly- nesian forefathers, but still occurring in some common country, ere the original stream of migration had divided itself over the different Pacific groups where the legend is still preserved. Now this period of Wakea, counting on the shortest Hawaiian genealogy, corresponds with the commencement of the Malay Empire in the Indian Archipelago. In the year j6 a.d., according to Javanese historians, Tritestra invaded Java, and commenced those wars against the Piakshasas, the Polynesio-Cushite pre-Malay inhabitants, which ended in their subjugation, isolation, or expulsion throughout the Archipelago. Eighty years from that time bring us to the period of Wakea, and the same time possibly brought the Malays from Java and Sumatra, where they first set foot, to Timor, Gilolo, and the Philippines. Taking tins epoch, therefore, as the starting-point for the great exodus and general appearance of the Polynesian THE POLYNESIAN RACE. 163 family in the Pacific, there is an interval of time of 900 to 1000 years in which to people the various islands and groups now held by that family, until we meet with the uncontested Hawaiian traditions which affirm that twenty- eight generations ago that group was already densely peopled by that family. But twenty-eight generations only represent a period of somewhat less than 900 years, and within that period there is no distinct tradition or remembrance of the active state of the volcanoes on the leeward islands of the Hawaiian group, or of the upheavals and subsidences to which those islands have been subjected. Yet recent dis- coveries have established the fact that those islands were inhabited before their volcanoes had ceased their action and the land assumed its present form. The legends of Pele, Hialca, and that family of demigods, it is true, would seem to infer the ancient Hawaiian belief that the leeward islands were inhabited while their volcanoes were still active ; but the legend of Pele itself, in its application to the Hawaiian group, when critically considered, must be subsequent to the great commotion which prevailed among the Polynesian tribes about twenty-six or twenty-eight generations ago, and is rather a mythical attempt in after ages to explain the volcanic phenomena of the group, than an historical datum for their occurrence. The tone of the legend, its several associations, and especially the therein, occurring prayer of Malaehaakoa, the Kalm or guard of Hialca, bespeak its later composition from Southern mate- rials recast in a Hawaiian mould. It is impossible to judge the age of a lava flow by its appearance. Portions of the lava stream of 1 840, flowing from the crater of Kilauea into Puna district, Hawaii, and thence to the sea, a distance of from sixteen to twenty miles, was in 1867 covered with a luxuriant vegetation; while older flows in Puna, of which no memory exists ; while the last How from Mount Hualalai in 1791-92, through Kekaha on the west of Hawaii ; and while the 1 64 THE POL YNESIAN RA CE. flow near Keonioio in Honuaula, island of Maui, called Hanahaie, and which is by tradition referred back to the mythological period of Pele and her compeers — look as fresh and glossy to-day as if thrown out but yesterday. Geologically speaking, the leeward islands are the oldest of the group, but both on Oahu and on Molokai human remains have been found imbedded in lava flows of un- doubted antiquity, and of whose occurrence no vestige ol remembrance remains in the Hawaiian folklore. In 1822 the first wells were dug in the city of Honolulu. They passed through some eight or ten feet of surface loam and underlying volcanic sand, when a coral bed of some eight feet in thickness was encountered and cut through, under which the fresh water was reached. In this coral formation were found embedded a human skull and sundry human bones. 1 In 1858, in dredging the harbour of Honolulu, island of Oahu, near the new Esplanade, after scooping up and removing the mud and sand at the bottom of the harbour in about twenty feet of water, it was found that under- neath this sand and mud was a pan of coral rock which it was necessary to break up and remove in order to obtain the required depth of water. This pan was of an average thickness of two feet, and beneath it was a thick couch of black volcanic sand, such as is found some four or five feet beneath the surface throughout the city, and evidently thrown out by the extinct crater of Punch-bowl-hill in some pre-traditional time. Embedded in this black sand, underneath the coral bed, was found the lower part or pointed end of an ancient spear or Oo, about three feet long ; and near to it a rounded small stone, the size of a hen's egg and nearly its shape, of a red, close-grained, compact, and heavy lava, such as is not found in the Punch-bowl-hill formation or its vicinity. The broken spear speaks for itself, and shows that man passed over 1 Hawaiian Club Papers, p. 3 ; by James Hunnewell. Boston, Octo- Article, " Early Wells of Honolulu," ber 1868. THE POLYNESIAN RACE. 165 that spot by water or by land before the formation of that coral pan which now covers the bottom of the harbour and the adjoining reefs. What purposes the stone had sub- served I am not prepared to say, unless it had been used for slings and dropped by the same hand or the same generation that dropped the spear. It bears no geological relation to the black sand around it, to the coral-rock above it, or to the extinct crater one and a quarter mile inland. 1 In 1859 Mr. E. W. Meyer of Kalae, Molokai, found in the side of a canon on his estate — some seventy feet, below the surface rim of the upper level country, and among a stratum of volcanic mud, Creccia, clay and ashes of several feet in thickness — a human skull, whose every cavity was fully and compactly filled with the volcanic deposit sur- rounding it, as if it had been cast in a mould, evidently showing that the skull had been filled while the deposit was yet in a fluid state. As that stratum spreads over a considerable tract of land in the neighbourhood, at a vary- ing depth beneath the surface of from ten to four hundred feet, and as the valleys and gulches, which now intersect it in numerous places, were manifestly formed by erosion — perhaps in some measure also by subsequent earthquake shocks — the great age of that human vestige may be reasonably inferred, though impossible to demonstrate within a period of one or five hundred years preceding the coherent traditional accounts of that island. (Hawaiian traditions on Hawaiian soil, 1 though valuable as national reminiscences, more or less obscured by the lapse of time, do not go back with any historical precision much more than twenty-eight generations from the present, or, say 840 years. Within that period the harbour and neighbouring coast-line of Honolulu have remained nearly 1 The writer was present when Lodge, No. 1, I. O. O. F., no public these articles were dug up from museum existing in Honolulu at the beneath the coral, and deposited time, them in the Library of Excelsior 1 66 THE POL YNESIAN RA CE. what they now are, nor has any subsidence sufficient to account for the formation of that coral bed beneath the city and its harbour, or of subsequent upheaval, or of any eruption from the Punch-bowl-hill crater, been retained on the memory of those twenty-eight generations. Among the Hawaiian genealogies now extant, I am, for reasons which will hereafter appear, disposed to consider the Haloa or Hoohokukalani- Nanaulu- Maweke line as the most reliable. It numbers fifty-six generations from Wdkea to the present time ; twenty-nine from Wakea to and including Maweke, and twenty-seven from Maweke until now. Fifty-six generations, at the recognised term of thirty years to a generation, make 1680 years from now ( 1 870) up to Wakea, the recognised progenitor and head of most of the Southern and Eastern Polynesian branches, and brings his era at about a.d. 190, which would in a great measure correspond with the invasion and spread of the Hindu-Malay family in the Asiatic Archipelago. But the first thirteen names on the Haloa line, to Nanaulu, are now allowed to have been shared, partially if not wholly, with the Marquesan and Tahitian branches of the Polynesian family, possibly also by the Samoan, though I have not now the means of ascertaining. These, then, must have existed elsewhere, and been introduced by the pre-Maweke occupants of the Hawaiian group, which would leave sixteen generations, or about five hundred years, in which to discover and people this group previous to the era of Maweke and his contemporaries, the Pauma- kua of Oahu, the Kuhiailani of Hawaii, the Puna family of chiefs on Kauai, the Hua family on Maui, the Kam- auaua family on Molokai, and others renowned in the legends and songs of the people. By which of these sixteen generations, from Nanaulu down to Maweke, these islands were settled upon, there is nothing positively to show. The historical presumption, however, would indi- cate Nanaulu, the first of these sixteen, as the epoch of such settlements ; and there still exists a Hawaiian tradi- THE POL YNESIAN RA CE. 1 67 tion concerning his grandson Pclickeula, who was a chief on Oahu. The first thirteen generations just referred to, from Wakea to Nanaulu, would thus represent the period of arrival and sejour on the Fiji group, and subsequent dis- persion over the Pacific. Probably the greater portion of those generations passed away on the Fiji group, for it is otherwise inconceivable how so much of Polynesian lan- guage and Polynesian folklore could have been incorpo- rated on the Fijian. And when the expulsion from there took place, several streams of migration issued simulta- neously, or nearly so, to the Samoan, Tonga, Tahiti, and other eastward and northward groups. The Marquesas group could be reached from Tahiti in a straight direction, through the trade- winds, and the Hawaiian from the Mar- quesas, as well as from the Samoan, by taking advantage of the north-east and south-east trade-winds. Whether the expulsion from the Fiji covered one year or fifty years, it does not necessarily follow, as some ethnologists are inclined to hold, that the Polynesians departed en masse either to Tonga or the Samoan group ; and after an in- definite period of residence there, and when population had become redundant, portions of it again moved eastward to Tahiti ; and after another indefinite period, moved north- ward to the Marquesas, and so on ; lastly, to the Hawaiian group. It is natural, and hence more probable, that the Polynesian settlements scattered over the Fiji group were attacked separately and successively, and that each chief- tain, as necessity compelled, fled with his family and fol- lowers in this or that direction, according as the state of 7 O the winds and the season of the year made it most favour- able to go. Many such parties, doubtless, made for the same group, and, finding the land already occupied by previous refugees, continued their course to the eastward and northward, until they found some convenient locality, where they finally established themselves permanently. The Polynesian legends would seem to support this latter 1 63 THE POLYNESIAN RACE. proposition. While it may be an open question whether the Tahitians came by the way of Samoa or direct from Fiji, Tahitian legends claim that one Tii was the first ancestor of Tahitian chiefs on Tahitian soil. Subsequent generations elevated him to the position of a demi-god and grandson of Taaroa, the southern god par excellence of a later creed. But Hawaiian legends claim this same Tii or Kit — who was the last of the thirteen from Wakea that lived elsewhere than on the Hawaiian group — as the father of Nanaulu, with whom Hawaiian aristocracy on Hawaiian soil co:nmences; while his brother Ulu re- mained at the south, and became the ancestor of that enterprising race of chiefs who six hundred years later overran the Pacific, from the Tonga group to the Hawaiian, and who gave rise to an era of commotion and unrest among the Polynesian tribes, the memory whereof is vividly retained in the Hawaiian folklore. With due reservation, therefore, regarding any lio-ht that may hereafter be shed on pre-Wakean voyages and settlements by Polynesians in the Pacific, we arrive at the following leading propositions as chronological sign- posts — approximately, at least — of Polynesian migrations to and in the Pacific : — ist. At the close of the first and during the second cen- tury of the present era the Polynesians left the Asiatic Archipelago and entered the Pacific, establishing them- selves on the Fiji group, and thence spreading to the Samoan, Tonga, and other groups eastward and northward. 2d. During the fifth century a.d. Polynesians settled on the Hawaiian Islands, and remained there, compara- tively unknown, until — 3d. The eleventh century A.D., when several parties of fresh emigrants from the Marquesas, Society, and Samoan groups arrived at the Hawaiian islands, and, for the space of five or six generations, revived and maintained an active intercourse with the first-named groups ; and — 4th. Prom the close of the above migratory era, which THE P OL I 'N ESI AN RA CE. 1 69 may be roughly fixed at the time of Laa-mai-kahihi and his children, about twenty-one generations ago, Hawaiian history runs isolated from the other Polynesian groups, until their re-discovery by Captain Cook in 1778. I have thus attempted to clear the path by which men of more varied knowledge and greater acquirements than myself may travel with increased facility, aud restore the Polynesian race to its proper place in the world's history. The ancient folklore at this end of the road unmistake- ably points to its former connection with those grand old- world peoples, the Arians and Cushites, of whom until the last century we hardly knew anything more than the names. It is for the savans of Europe and America to clear the other end of the road, and the more light they can throw upon those ancient races, the more numerous will be the points of affinity between them and the Polynesians. Hawaiian history, during the first period above referred to, is naturally merged in that of the entire stock which emigrated to the Pacific. Whether the dialectical and other differences which distinguish the Hawaiians from the Southern and "Western groups, and each group from the other, existed as already formed tribal characteristics at the time of the migration, or were developed afterwards through dispersion and isolation, there is nothing positive to determine. They are probably owing to both condi- tions. The same dialectical differences are still recoo-nis- able among their pre-Malay congeners in the Asiatic Archipelago, spite of the corrupting influences to which they have been subjected from Malay, Arab, and Chinese intermixture. These differences, then, must have been older than the first dispersion into the Pacific, though they may have been hardened and deepened by subsequent events. I am thus led to infer that at a period, which, for 17 o THE POLYNESIAN RACE. reference sake, I will call the Wakean period, a number of expeditions, impelled by the pressure of the Malay con- quests, left their ancient homes on the eastern border of the Asiatic Archipel, and proceeding eastward on both sides of the Papuan Archipel, met on the Fiji group, and thence spread to the eastward, southward, and northward. And such I take it is the tenor of Polynesian folklore when critically studied and properly collated. In the Northern and Eastern groups the names of places which coincide with, and evidently were called after, local names in their ancient habitats, are to a remarkable extent drawn from the northern portion of the Asiatic Archi- pelago, the Moluccas, Celebes, Borneo ; whereas the South- eastern groups of Polynesia show an equal preference for names drawn from the southern portion of the above Archipel, the Banda islands, Timor, &c, while both streams unite in, and refer to, the more western islands and countries beyond as a common source for their local nomenclature. In several of the Hawaiian legends respecting Wakea, he is said to have been chief over a country called O-lolo-i- inehani. This word is composed of the prefix 0, the name lolo, and the epithet mehani. That lolo and the reference to it as the western home of Wakea point to that one of the Moluccas which by Spanish, Dutch, and English navi- gators is variously called Gi-lolo, Ji-lolo, Dji-lolo, and l-lolo, I think there is little doubt. The epithet mehani is now obsolete, and its meaning forgotten in the Hawaiian dialect, and I think, but am not sure, in the other Poly- nesian dialects. 1 But in the Amblaw dialect, one of the 1 In Raiatea, of the Society group, just to touch, pass quickly through there is a mountain called Mehani, the air, to disappear, vanish." Epi- where the ghosts of the dead were thets of places may vary in different said to go. That mountain, like so ages and under different conditions, many other places in Polynesia, may and whether the legendary Mehani have been a namesake of some older allies itself to the Amblaw or Poly- locality, though its application as a nesian signification, there can be no residence for departed spirits might doubt of the identification of the tra- indicate a derivation from the Poly- ditional O-lolo with the modern Gi- nesian (Hawaiian) hani, with prefix lolo. mahani, "to step lightly, to graze, or THE POL YNESIAN RA CE. 171 Bandas, south of Burn, the word still remains, and means " red ; " and in Ceram (Ahtiago) la-hanin, " red," is evi- dently the same word. But all the Polynesian tribes invested the red colour with special dignity as a mark of royalty and pre-eminence, and as such it was a most proper epithet for the largest and probably dominant island of the Molucca group. In the legend of Hawaii-loa and his descendants, Ka Oupe-Alii, the grandmother of Wakea's wife Papa, is said to have been a princess from O-lolo-i-mehani. Other legends of Wakea mention that his father Kahxko lived in O-lalo-waia ; l others again give Wakea a land, which they call Hihilcu, as his residence ; and Hawaiian commenta- tors on these legends suggest that O-lolo-i-mehani was the island of Nukahiwa, Marquesas group, or some place upon it, and that O-lalo-waia was some place on Oahu, Hawaiian group, saying that such was one of the ancient names for Oahu, or a portion of it, as Kana-wai-lua-lani was an ancient name for the island of Kauai. But a critical comparison of the legends referring to Wakea brings out the fact that he probably never set foot on either Marquesan or Hawaiian soil ; and that, the names being given by the earlier settlers, the process is both easy and intelligible by which the priests and bards of after- times transferred and localised on their own groups the hero with whose names those places were connected in the ancient legends. In another Hawaiian legend the islands of that group are said to have been created and named by Wakea and Papa. In this cosmogony Wakea is said to have had illicit intercourse with a woman called Hina, and she brought forth the island which Wakea named Molokai. In revenge for this unfaithfulness, Papa co- habited with a man called Lua, and gave birth to the island of Oahu; and in commemoration of this double adultery, the two islands have ever after preserved the sobriquets arising from their birth, viz., Mololcai-Hina 1 See p. 17. i 7 2 THE POLYNESIAN RACE. and Oalm-a-Lua. Under the crndeness and coarseness of the legend we may discover the lingering reminiscence of a geographical and historical fact, namely, the ancient con- nection between the O-lolo or Gi-lolo chief Wakea and the neighbouring island of Morotai, after which the Hawaiian Molokai was undoubtedly named. The reference to that island by that name puts the identity of O-lolo and Gi-lolo beyond much doubt. And the connection of Papa with Oahu points to the central and probably once powerful state of Ouadjon in Celebes, and recalls the legend which makes Papa a lineal and tabued descendant of Hawaii-loa, and claims that Wakea was inferior to her in royal dignity. In the legend of Pupuhuluana the relative position of O-lolo-i-mehani is farther indicated. Papa, under her other name of Haumea, had caused a fearful drought and famine to devastate not only her own island of Oahu, but also Kauai, Maui, and Hawaii, and she herself had retired to a land frequently mentioned in the legends by the name of Nuu-meha-lani. In this distress some people of Oahu fitted out an expedition to procure food from O-lolo- i-mehani, " the land of Makalii," ka aina o Makalii, which land was to the eastward of Oahu. With the exception of the allusion already referred to, of Wakea having been driven out of his country by a hos- tile chief, and fleeing over the ocean, and afterwards con- quering his enemy and recovering his country, there is no Hawaiian legend that I have become acquainted with which refers to any great migration performed by Wakea himself, or by any of his children. Most of the legends which do not treat Wakea and Papa as gods, or endowed with superhuman powers, assume that they were born and bred on the Hawaiian group ; and those who do admit and refer to their foreign origin, yet give no account of their leaving that foreign home for the Pacific, or how they or their descendants arrived here. Though the Marquesan legends are more explicit than any others upon the migrations of their tribes, and prob- THE POL YNESIAN RA CE. 173 ably the entire Polynesian family, and though one of the migrations preserved in their legends bears the name of Atca, there is nothing to connect him with the Wakea of the Hawaiian genealogies. It is probable, moreover, that the branch of the Polynesian family to which the Mar- quesans originally belonged took its departure from Vevau or Wawao, i.e., Timor, and not from Gilolo, and thus came by Torres Straits and the southern passage into the Pacific. They left the marks of their passage in the still retained names of Nuu-mea, near Port de France, in New Cale- donia; in Ua-Hio, a bay in Lifu, one of the Loyalty isles ; in Uca or Ua-ea, one of those islands ; and in a point of land or cape on Uea, called Fae-a- Ue, " the house of Ua." That name of Ua, whoever or whatever he was, meets us again in the Marquesas group, Ua-pou, and Ua-Jmlca or Ua- hunga, according to the harder or softer pronunciation of the natives. A greater acquaintance with the coast-line and the interior of the Papuan groups, and with the people generally, would no doubt greatly increase these vestiges of Polynesian passage and sejour. On the islands of Vate and Mcle, of the New Hebrides, the numeral system is identical with the Polynesian of the earlier dialects, re- taining both the s and the r sounds ; and other indications demonstrate the Polynesian presence and, probably, par- tial absorption in the Papuan race of this group, or, rather, these portions of the group. The two Marquesan accounts of the wanderings of their people ere they reached their present abodes, while they entirely agree in the earlier and later stages of the journey, materially disagree in the middle portions. Apparently they are the representations or reminiscences of two tribes or branches of the same family, travelling together, or fol- lowing each other, over the earlier portions of the journey, then separating for several stages, and finally uniting again, or striking the same trail, so to say, until they arrived at the Marquesas group. These itineraries are called by the principal personages whom they represent, *74 THE POLYNESIAN RACE. or whom the travellers claimed as their ancestors, the Atea and the Tani migrations. Here are their way- bills:— Atea Account Tani Account. From Take-hee-kee to From Take-hee-hee to 53 Ahee-tai » 5' Ahee-take 33 >> Ao-nuu 33 5) Ao-nuu 33 33 Papa-nui 35 33 Papa-nui 53 >3 Take-hee » 55 Take-hee 33 35 Ho-vau 55 55 Nini-oe 55 35 Ao-ewa 35 35 Ani-take 33 3) Ani-take 55 | 33 Ho-vau 33 Hawaii 35 33 Vevau 33 Tu-uma 33 33 Tu-uma 33 Mea-ai 55 33 Mea-ai 53 Fiti-nui 5) 35 Fiti-nui 35 Mata-hou 55 33 Mata-hou »> Tona-nui 55 3) Tona-nui 53 Mau-ewa 55 J3 Mau-ewa S> Pi ina 55 33 Pi ina n Thence " over the ocean" to 35 Ao-ma ama 33 Ao-maama Their name for the Marquesas Islands. The chant or legendary poem which accompanies the Atea account appears to be imperfect or partly forgotten. It gives short and passing descriptions of the eight first stations, then passes over Fiti-nui in silence, then notices Mata-hou, but takes no note of Tona-nui and Mau-ewa. I have seen no chant explanatory of the Tani migration. If any such exists among the Marquesans, it is to be hoped that some resident gentlemen of leisure and archaeological predilections may collect and publish them before the priests of the heathen time and the old people generally, from whom they may be collected, have become extinct. From the chants to which I have had access, through the politeness of Professor W. D. Alexander, and which have been collected and carefully translated by Mr. T. C. Law- THE POLYNESIAN RACE. 175 son, a resident on Hiwaoa or St. Dominica island, the fol- lowing prosaic and historical resume of the migrations of the Take, as the Marquesans are called in the chants, may- be presented : — Take-hee-hce, or Ahee-tai, as another legend calls it, was the oldest original home of which the " Takes " had any remembrance. It is described as a mountain-land with a settlement or inhabited district at Tai ao, another at Meini- taha-hua, and another near the water (lake or river) of Nuu-teea. Wars and commotions having; arisen among themselves, the people were driven out of this land and migrated to — Ao-nuu, which is described in the chant as He hernia hiwaoa mei Ahee-tai, He henua hiwahiwa Ao-mai. " A beautiful country far from Ahee-tai, A beautiful country is Ao-mai." While dwelling in Ao-nuu a chief ruled over the country whose name was Faaina. After him came Anu-o-Aatuna. After that the chief Atea killed Umai, by which civil wars arose, and Atea and many other " Takes " were driven out and obliged to seek new homes in other lands. They then migrated to Papa-nui, which seems to have been reached by sea, for a legend relates that the chief Tiki-Matohe and his wife Hina left Aonuu with their followers and outfit of pigs, fowls, and fruit in a double canoe, and thus, with a favour- ing wind, arrived at Papa-nui. This land is described as a high table-land, surrounded by the sea. It appears also that the Tani branch of the family arrived at Papa-nui after Atea, for one of the chants mentions his cordial reception as one of the same family as Atea, and how, for his entertainment, pigs were brought from Ao-tumi, and turtle from Ono-tapu, and fowls from below Ii-Haioa and Nuu-teea. The next stopping-place was Take-liee, which is said to Tu hiwaoa eeke eelce i te hee. 176 THE POLYNESIAN RACE. Here the two branches seem to have separated, the Tani legend mentioning; five lands not visited, or at least recorded by the Atea legend, while the latter makes only two stop- ping-places between Take-hee and Tu-uma, where the Tani branch seems to have joined it again, or come in upon its- track. But while thus separated the Atea branch visits Hawa-ii, which the legend calls Tax mamao uta-oa tu te Ii. " The distant sea or region, far inland stand the volcanoes." The hupe, kohanui, mio, and temanu trees are said by one chant to have been growing there in abundance. It is also said to have been subject to tremendous hurricanes, fol- lowed by famines. Two of the chants give rather parti- cular descriptions of the Hawa-ii remembered by the Mar- quesans. One mentions five headlands or capes, Fiti- tona-tapu, Pita, Ao, Ao-cna, and Ao-oma, and one mountain which it calls Mouna-tiha-oe. The other chant, of evi- dently later origin, mentions a mountain called Mouna-oa, which is said to have been raging, burning (Ii) on top, and served as a landmark for Tupaa when he left Hawa-ii with his family and followers. The order in which this Hawa-ii appears on the Mar- quesan carte de voyage, and other considerations, make it impossible to identify it with the North Pacific Hawaiian group, or even with the Saiva-ii of the Samoan group. The constant and emphatic expression of all these legends, that the wanderers came from " below," mei iao mai, from the direction towards which the wind was blowing, and were always going " up," iuna, in the direction from which the wind was blowing, makes it evident that the Hawa-ii to which they refer must have been situated to the west- ward or " below " the Fiti, Viti, or Fiji group, from which, with one intermediate station, whose name I am not now able to identify, they proceeded to the Tonga group, Tona- nui, and thence to the Society group, or Mau-ewa, which name I consider to be the same as Ma-cwa, a district on ■THE POLYNESIAN RACE. 177 the island of Huahine, thence to Pi ina, now not known by that name, and thence the wanderers, still going up on the wind, crossed the ocean — una te tai — to the Marquesas or Te Ao-maama. That the Marquesans in aftertimes visited the Hawaiian group there can be little doubt, and it is quite probable that the whole or a portion of the early Hawaiian settlers came from or passed through the Marquesas group ; but that the Havm-ii of the Marquesan carte de voyage is the North Pacific Hawaii is not credible under any proper analysis of the legend. It was, then, to the westward of the Fiji group, and, according to the legend, removed by two stages. But one of these is said in the chant to be " near to Hawa-ii " — Te I'uuma i Hawa-ii tata ae, while the situation of the other Mea-ai is not indicated. We thus find ourselves again in face of a western Hawa-ii, far west of the Fiji group ; but whether it is the same Hawaii to which the Hawaiian legends refer there are no means to decide. Probably it was not. The Hawa, Saiua, and Djawa name and its composites were not uncommon appellations of island places and districts throughout the Asiatic Archipelago, and some one of these may have been the Hawa-ii in question. Here the Tani account of the migrations may offer an indication at least of the direction in which this Haioa-ii is to be sought for. Tracing that account backward from Ao-maama and beyond the Fiji group, through places identical with the Aiea account, we find that Vevau is the station just previous to Tuuma, and not Hawa-ii, as the other account calls it. I have already shown that the Vevau referred to in the earlier Marquesan legends corresponds, in all probability, to Timor of the Asiatic Archipelago ; and thus understood the Tani account ren- ders the journey both intelligible and credible. Whether Hawa-ii in those ancient times was another name for Vevau or Timor, or whether in the Atea account it is used vol. 1. M 1 7 8 THE POL YNESIAN RA CE. as a representative name for the Asiatico- Polynesian area and the eastern and last portion especially, it is now impossible to say. The current traditional belief among the Southern Marquesans, that they came from Hawaii, which in ordi- nary parlance has become synonymous with " the regions below, the invisible world," — and the similarly current belief among the Northern Marquesans, that they came from Vavao, an island " below," i.e., to westward of, Nuka- hiwa, — point to the earlier legend and its two migrations, that of Atea and that of Tani. And dialectical differences between the northern and southern portions of the group confirm the fact of a double origin; whether from two originally distinct tribes, or at two widely separate epochs, I am unable to determine. Mr. Hale, in the " Ethno- graphical portion of the United States Exploring Expedi- tion," p. 127, inclines to the conclusion that the Marquesans were colonists from Sawaii of the Samoan group. I think it quite probable and very natural that a considerable portion of the Marquesans did come from the Samoa, either direct or via the Society group ; but the legendary Hawa-ii and Vevau of the Marquesans lay unquestionably farther west than either the Samoan or the Tonga group. There is no time, or attempt at specification of time, connected with these Marquesan legends ; and the con- formity of names in the legends with those on the only Marquesan genealogy which I have seen will not even warrant a conjecture. A better acquaintance with, and a critical comparison of, the Marquesan genealogies still extant might furnish some approximative data for deter- mining the period of these migrations. 1 1 Mr. Hale of the United States and other plants with them. Eighty- Exploring Expedition under Com- eight generations, at thirty years modore Wilkes, in the section of each, make 2640 years back of 1812-14 Ethnology and Philology, p. 128, when Commodore Porter visited the quoting from Commodore Porter, Marquesans. I have little doubt that states "that the chief Ke-ata-mti of the Ke-ata-nui genealogy was as in- Nukahiwa was the eighty-eighth gene- flated as the Hawaiian genealogies of ration from Oataia (Atea) and his the Ulu-Hcma line, by admixture wife Ananoona.v/ho came from Vavao with Tahitian, Samoan, and possibly and brought bread-fruit, sugar-cane, Tongan collateral issues. THE POL YNESIAN RA CE. 179 I am very little acquainted with the Sarnoan traditions and legendary lore, and am unable, therefore, to state what reference, if any, the ancient legends of that group may make to the Polynesian migrations into the Pacific, the time of their occurring, or whence they started. The name of the Samoan group, however, affords, in my opinion, some indication of the extraction of the people who named and inhabit it. The group is called by the natives Samoa, in the Tonga and other dialects Hamoa. The early Spanish visitors to the Molucca islands give the ancient names of Gilolo as " Maurica " and " Bato-chine," and mention the middle part of Gilolo as being called Gamoca-nora. The affinity or identity of Gamoca, as the Spaniards pronounced it, and Hamoa or Samoa is intel- ligible, and no doubt will be unquestioned by Polynesian scholars ; l but the epithet nora I am unable to explain, unless it connects with the Polynesian (Hawaiian) noa, meaning " constantly burning, unquenchable as a volcano," and thus referring to the former active state of the volcanoes on Gilolo. In the absence of positive evidence to the contrary, it is therefore extremely probable that the Samoan s came from the Gilolo group and to the north of the Papuan Archipel ; and with them, or by the same route, came the Hawaiians, 1 In the account of the United States would differ from the Malay extrac- Exploring Expedition, Com. Wilkes, tion and the later adoption of the sect. Ethnology and Philology, Mr. nanie. It has heen shown, I hope, H. Hale, p. 120, proposes that Samoa in previous pages, that the Polynesian is a Malay word and adopted by the owes nothing to the Malay per se, Navigator Islanders to designate their least of all the names of its places group; its meaning being "all," scil. and islands. As to the later adoption of that group, and equivalent to the of the name by the Samoans them- American expression of "Union" selves, I think it questionable, inas- when speaking of the United States much as the name was known not as a whole, united country. He only to the Tongans, who, however, intimates that it must have been of may be considered as comparative later adoption, and that the earlier neighbours, but also to the Tahitians emigrants from the Samoas only knew and the still farther distant Hawai- that group by the name of its largest ians,the latterof whom still designate island, Sawaii. W T ith due deference a land on the island of Maui by the to so able and careful a writer, I name of Hamoa. i So THE POLYNESIAN RACE. possibly also the Society Islanders ; while the Marquesans and the Tongans came by the southern route and Torres Straits, the former from Timor, the latter from Buru. From what has been already said, it is equally probable that some portion of the Fiji group was the primary rendezvous of these two, three, or more streams of migra- tion, and that, whether expelled or leaving voluntarily, a new division took place there according to tribal, dialectical. or other affinities and predilections, some seeking new homes in the north-east, others in the east and south-east. And it has been shown by one genealogy at least that this ethnic movement embraced a period of from seven to thirteen generations previous to the forty-third recognised, and generally considered as authentic, ancestor of the present Hawaiian chief families. Of these thirteen names born on most of the Hawaiian genealogies very little is known that throws any historical light on that period. David Malo, a Hawaiian gentleman, educated by the earlier missionaries, states in his " Ha- waiian Antiquities," that many well-informed people of the olden time maintained that the six first generations after Wakea still lived in O-lolo-i-mehani. Be that as it may, it is evident that the Tahiti mentioned in these earlier legends, — to and from which Papa, Wakea's wife, made so many voyages, where she took other husbands and had other children, from whom the Polynesian Tahitians claim their descent, and where she finally died, — could not have been the Tahiti of the South Pacific, but must be sought for in some of the islands of the Asiatic Archipelago. It is presumable that, when in after ages the intercourse between the Polynesian tribes was renewed, the scenes of those early legends was shifted and modified to suit the requirements of the new area which they then occupied ; and thus O-lolo-i-mehani became located on Oahu of the Hawaiian group, while the Tahiti of the legend was transposed to Tahiti of the Georgian or Society group. THE POLYNESIAN RACE. 181 Before proceeding further, as we are entering on pecu- liarly Hawaiian domain, it may be proper in this place to insert the various genealogies current among the Ha- waiians, so that the reader may understand the force of our subsequent criticism and attempt at correction. We will commence with the different genealogies which, start- ing from the first man, lead down to Wakea and Papa ; then those which, starting from WaJcea and Papa, lead down to the present time. THE GENEALOGY OF KUMUHONUA. The letter k. means kame or husband ; w. means wahine or wife. 1 Kumuhonua, k. Lalo-Honua, w. 2 ( Laka, or Kolo-i-ke Ao, k. Papa ia Laka, w. < Kulu-Ipo, or Kolo-i-ke Po, or Ahu, k. \ Kapili, or Kaiki-ku-a-Kane, k. 3 Ka Moolewa, k. Olepuu-Honua, w. 4 Maluapo, k. Laweao, w. 5 Kinilan-a-Mano, k. Upolu, w. 6 Halo, k. Kini Ewalu, w. 7 Ka Mano Lani, k. Ka Lani-a-Noho, w. 8 Ka Maka o ka Lani, k. Ka Moo Lani, w. 9 Ka Lei Lani, k. Opua Hiki, w. IO Ka La Lii, k. Ke Ao Melemele, w. 1 1 Haule, k. Loaaio, w. 12 Inii Nanea, k. Imi Walea, w. 13 Nuu, or Kahinalii, k. 2 Kapili, k. Nohi-nohi-nohele, w, 3 Ka Wa Kahiko, k. Luhiluhi Heleae, w. 4 Ka Wa Kupua, k. Kahiko-o-Lupa, w. /Kahiko Lei Kau, w. I Kahiko Lei Ulu, w. 5 < Kahiko Lei Honua, k. I Hakoakoa Lau Leia, k. vKupo, w. 6 Ke Ake Nui, k. C Mauli Neweneweloa, k. 7 1 Ke Olai Maolina a kane. Nahae-i-Kua, to. ( Ka Ipo Lau Leia-i Heleua, w. ( Kalani-Hoohonua, w. Muo Lani, w. 182 THE POLYNESIAN RACE. 8 Ka Lei Lani, k. Apaiki, w. 9 Hauli i Honua, k. Laa-a, w. IO Ka La Lili, k. Ke Ao Melemele, w, ii Lalo-o-Kona, k. Ka Mole Aniani, w. 12 Hoo Nanea, k. Hoo-Walea, w. 13 Nuu or Kahinalii, k. J 3 Nuu, k. Lili-noe, %o. Hi r Nalu Akua, h. Nalu Hoohua, A. (. Nalu Manamana, A;. Ka Ali Akea, w. 15 Naeheehe Lani, k. Kawowo-i-Lani Hikimoe 16 Ka Hakui Moku Lei, k. Ke kai Holana, w. 17 18 i9 20 21 22 w. Ke kai Lei, k. Ka Haku Lani, k. Hele i Kahiki Ku. k. Ka Noelo Hikina, k. Hele i ka Moo Loa, k. Ke Au Apaapaa, k. Lua Nuu, or Kanehoalani, Nalu Lei, w. Moeana-i-lalo, w. Hooneenee i ka Hikina, w. Hala Po Loa, w. Kawehe'a'ao, w. Ke Au Laelae, w. 14 Nalu Manamana, k. 1 5 Ka Io Lani, k. 16 Hakui Moku, k. 17 Nunu Lani, or Imi Lani, k. 1 8 Honua ka Moku, k. 19 Neenee Papu Lani, k. 20 Hele i kua Hikina, k. 21 Hele Moo Loa, k. 22 Ke Ao Apaapaa, k. 23 Lua Nuu, or Kanehoalani, k. Manamana-ia-Kuluea, w, Kawowo-i-Lani, w. Lu-i-ka Po, w. Pili Po, vo. Anahulu ka Po, w. Wehe ka Po, w. Hala ka Po, vo. Ka Wanaao, w. Ke Ao Laelae, w. 23 Lua Nuu, k, C Aliu, w. \ MeeHiwa, Hakulani, Po Malie, vj. 24 \ Ku Nawao, k. Kalani Mene Hune, k. Aholoholo, k. Ka Imi Puka Ku,Kinilau-a-Mano, k. Ka Hoolulii Kupaa, w. Ka Mole Hikina Kuahine, w. THE POL YNESIA X RA CE. 1 8 3 / Ka Hekili Paapaaina, k. Ke Apaapa Nuu, k. Ke Apaapa Lani, k. Nakeke i Lani, k. Kahiki Apaapa Nuu, /;. Kahiki Apaapa Lani, k. Nakolokolo Lani, k. Nakeke Honna, k. Ku i ka Ewa Lani, k. Ka Uwai o ka Moku, k. Hoopale Honua, k. 26 ' Newenewe Maolina i Kahiki-kn, k. Nowelo Hikina, w. 27 Kaokao Kalani, k. Heba ka Moku, w. 28 Aniani Ku, k. Ke kai Pahola, w. 29 Aniani Kalani, h. Ka Mee Nui Hikina, w. 30 ( Hawaii Loa, or Ke kowa i Hawaii, k. Hualalai, w. Kii, k. Kana Loa, k. Laa Kapu, k. { Maui-ai-Alii, k. 31 < Oaliu, rv. Ku Nui ai a ke Akua, /;. ( Kauai, k. 32 Ku Nui Akea, k. Kahiki "Walea, w. 33 Ke Lii Alia, k. Kahiki Alii, w. 34 Ke Milia, k. Polohainalii, w. 35 Ke Lii Ku, or Eleeleualani,7;.Ka Oupe Alii, w. 36 Ku Kalani Ehu, k. Ka Haka ua koko, w. 37 Papa Nui Hanau Moku, w. Wukea, k. I have another genealogy from Kumuhonua to Papa and Wakea, purporting to be the genealogy followed by the ancient Hawaiian priests of the Paao line. That genealogy inserts ten generations between Newenewe Maolina and Hawaii Loa, Nos. 26 and 30, and thirty-four generations between Hawaii Loa and Papa Nui, Nos. 30 and 37 on the foregoing list. It is said to have been confirmed and approved by the late chiefess Luahine, wife of KaoleioJcu, the first born son of Kamelmmelw L., and grandmother of the present high chiefess, Mrs. Pauahi Bishop ; but it is evident at almost the first glance that, even if those 1 84 THE POL YNESIAN RA CE. ancient priests had correctly preserved the tradition of the number of links in this genealogical chain, yet the naming of them has been an entirely arbitrary operation in far subsequent times, — presenting more the appearance of a geographical list of lands and islands known to the com- pilers, personified and genealogically arranged, than a proper pedigree of genuine names. It is very probable that this last arrangement of the Kumuhonua genealogy was another of those curious interpolations made after that great Southern influx in the Hawaiian group, to which I have alluded in previous pages, and by the Kahunas or priests of that period and of those invaders. The Genealogy of Kumtj-tjli. This genealogy was much praised by the ancient Ha- waiian chiefs. It is quoted in the famous chant of Kualii, the warrior king of Oahu, and was recited in honour of Kcopuolani, the wife of Kamehameha I., and who was a tabued scion of the Maui line of kin2;s. It runs thus : — r Kane. \ Kanaloa. Ukina-opiopio, w. j Kauakalii. ' Maliu. / Hulihonua, k. j Laka, k. Keakahulilani, v>. ") Kamooalewa, k. ' Maluakapo, k. 2 Laka, k. Kapapaiakele, w. 3 Kamooalewa, k. Olepuukahonua, w. 4 Maluapo, k. Lawekeao, w. 5 Kinilauamano, k. Ulupalu or Upolu, w. 6 Halo, k. Koniewalu, w. 7 Kamanonookalani, k. Kalani a noho, w. 8 Kamakaoliolani, k. Kaehuaokalani, w. 9 Keohokalani, k. Kaamookalani, w. io Kaleiokalaui, k. Kaopuahihi, w. 1 1 Kalalii, k. Keaomele, w. 12 Haule, k. Loaa, w. 1 3 Nanea, k. Walea, w. 14 Nana Nuu. k. Lalokana, w. r//£ po LYNES1AN RACE. 185 15 Lalokona. Lalohoaniani, w. i6 Hanuapoiluna, k. Hanuapoilalo, w. 17 Pokinikini, k. Polehulehu, w. 18 Pomanomano, k. Pohakoikoi, w. 19 Kupukupunuu, k. Kupukupulani, w, 20 Kamoleokahonua. Keaaokahonua, w. 21 Kapaiaokalani, k. Kanikekaa, w. 22 Ohemoku, k. Pinainai, w. 23 Mahulu, k. Hiona, w. 24 ]\Iilipomea, k. Hanahanaiau, w. 25 Hookumukapu, k. Hoaono, u\ 26 Luakahakona, k. Niau, w. 27 Kahiko, h. Kupulanakehau, w. 28 Wakea, k. Papa Nui, w. The correspondence between the fourteen first genera- tions of this genealogy — with the exception No. g, Keoho, which in some versions is omitted — with the first thirteen of the Kumnhonva genealogy is, to say the least, remark- able. But the introduction of the four divinities, Kane, Kanaloa, Kmtakahi, and Maliu, as the natural parents of the first man Hulihonua, and as part and portion of a human genealogy, is, according to my understanding of Polynesian folklore, a clear indication that this genealogy was compiled from pre-existing materials after the influx of the southern element, about 800 years ago. The Kurau- honua legend draws a broad distinction between man and his Maker, which this legend ignores. Its later origin is, moreover, evidenced by the introduction of Kanaloa, the great southern god of later times, although, as a compro- mise or a concession to the primary and comparatively purer creed of the Hawaiians, he is placed in a secondary position to Kane. After Nuu, Nana-Nuu, or Kahinalii, this correspondence ceases, the Kumu-idi genealogy lead- ing down to Wakea, whereas the other leads down to his wife Papa. i 86 THE POLYNESIAN RACE. THE GENEALOGY OF OPUKAHONUA. I ( Opukahonua, k. < Lolomu, k. ( Mihi, h. !Kananamukumanao, k. Ohikimakaloa, w. Hekilikaaka, k. | Nakolowailani, k. I Ahulukaaala. w. Lana, 70. Ohikimakaloa, id. 3 Mihi, k. Ahulukaaala, to. 4 Kapuaululana, k. Holani, w. 5 Kekamaluahaku, k. Laamea, to. 6 Lanipipili, k. (" Laakeakapu, w. \ Hinaimanau, to. A Lanioaka. Laakealaakona, k. Kamaleilani, w. 8 Haulanuiakea, k. Manau, to. 9 Kahaloalena, k. Laumaewa, iv. IO Kahalolenaula, k. ( Kanehoalani, w. < Hinakului, w. \ Kaihikapualamea, w. 11 ( ' Kaiwilaniolua. Kanehoalani, w. 11 Kapumaweolani. Haweaoku, w. ill w Kukonalaa. Kaenakulani, w. I2 I ' Kalaniwahine, k. Manuiakane. , Kalanipaumake. Malela, w. ! 3 I ' Kamakahiwa, k. Loi, 10. I3 < Makakaile, k. Paweo, w. 13 ' 1 Makakailenuiaola, k. \ 14 Kikenui a Ewa, k. > Ewa, w. 14 Kalanirnanuia, k. ) l S\ ' Kahiko, k. Kupulanakehau, w. 15 \ Kupulanakehau, w. 15 ) Kukalaniehu, k. Kahakauakoko, to. 15 1 -. Kahakauakoko, w. 16 Wakea, k. 16 Papa, w. ■ THE POLYNESIAN RACE. 187 THE GENEALOGY OF KAPAPAIAKEA. Quoted in the Chant of Kualii, the King of Oahu. 1 Kapapaiakea, k. 2 Hinakapeau, k. 3 Ukinaopiopi, k. 4 Kalei, k. 5 Kaiakea, k. 5 Kamoanakea, k. Iwikauikauanui, k. 5 Hulukeeaea, Jc. 6 Hauiikaiapokahi, k. 7 Maihea, k. 7 Uliuli, t. 8 Kahiko, h. 9 Wakea, k. Kauhihi, w. Ukinohunohu, iff. Moakuanana, iff. Keelekoha, w. Kaehokumanawa, w. Kauakaliikuaana, w. Kahakuakea, no. Wahineikapeakapu, w. Kahakapolani, w. Niau, w. Kupulanakehau, to. Papa-nui, w. Hinakapean. Ukinaopiopi. Kalei. Kaiakea. ( Kamoanaakea. ( Hulnkeeaea. ( Hauii. ( Hauee. Kanehoalani. Hauinui naholoholo. Hauiikaiapokahi. Uliuli. Maihea. Kaukeano. Mehameha. Kahiko. Wakea. THE GENEALOGY OF WELAAHILANI. 1 Welaahilani, 1c. Owe, w. 2 Kahiko-Luamea, k. Kupulanakehau, 10. 3 Wakea, k. Papa, w. Kahiko. Wakea. k. Lihauula, k. Makuu, Jc. According to the genealogy called Kumuvlvpo, a woman called Lailai was the first person on earth, de- scended from Po or chaos. From her and her husband Kealihcahilani, the rest of mankind were derived. Their son was Kahiko, the father of Wakea. According to the tradition called Pvanue, the creators of heaven and earth, and the progenitors of mankind, were Kunmkanikekaa and her husband Paialani. i88 THE POLYNESIAN RACE. Among the various Hawaiian genealogies I consider the Nanaulu line as the most reliable and least affected by the interpolations and confusion introduced by the southern element so often referred to. It was extensively, almost exclusively, patronised by the Kauai and Oahu chiefs, and seldom referred to by the Maui, — hardly ever by the Hawaii chiefs. I will, therefore, commence with that and bring it down to the person of the present reigning sovereign Kalakaua. I would premise by saying that there exist two versions of the earlier portion of this genealogy, from Wakea to Kii, one descending from Wakea's son Haloa, the other from his daughter Hoohokukalani. The former was the most generally current of later times, but the latter appears to me to be the most archaic as well as the most trustworthy, for reasons which will appear when I come to treat of the Ulu line, and as the number of generations is the same on both, though the arrangement is somewhat different, I prefer to follow the latter in this earlier portion down to Nana-ulu. THE NANA-ULU GENEALOGY. I Wakea, k. Papa, w. f Haloa, k. \ Hoohokukalani, w. 2 Hoohokukalani, w. Manouluae, k. Waia, k. 3 Waia, k. Huhune, w. Wailoa. 4 Wailoa, k. Hikawaopualanea, w. Kakaihili, 5 Kakaihili, k. Haulani, w. Kia. 6 Kia, k. Kamole, w. Ole. 7 Ole, k. Haii, w. Pupue. 8 Pupue, k. Kamahele, w. Manaku. 9 Manaku, k. Hikohaale, w. Nukahakoa. io Nukahakoa, k. Koulamaikalani, w. Luanuu. 1 1 Luanuu, k. Kawaamaukele, w. Kahiko. 12 Kahiko, k. Kaea, 10. Kii. 13 Kii, k. Hinakoula, w ( Nana-ulu. (Ulu. 14 Nanaulu, k. Ulukou, w. Nanamea. 15 Nanaraea, k. Puia, w. Pehekeula. THE POLYNESIAN RACE. 189 16 Pebekeula, k. 17 Pehekemana, k. 18 Nanamua, k. 19 Nanaikeauhaku, k. 20 Keaoa, k. 21 Hekumu, k. 11 Umalei, /•. 23 Kalai, k. 24 Malelevvaa, k. 25 Hopoe, h. 26 Makalawena, k. 27 Lelehooma, &. 28 Kekupahaikala, k. 29 Maweke, k. 30 Mulielealii, A;. 31 Moikeha, k. 32 Hookamalii, X". 33 Kahai, k. 34 Kuolono, k. 35 Maelo, w. 36 Laulikewa, k. 37 Kahuoi, ft. 38 Puaakahuoi, k. 39 Kukahiaililani, h. 40 Mailikukahi, A;. 41 Kalona-nui, k. 42 Kalamakua, &. 43 Laielolielohe, w. 44 Piikea, iv. 45 Kumulae, #. 46 Makua, k. 47 I.,* 48 Ahu-a-I., k. 49 Kapaihi-a-Ahu, A;. 50 Heulu, k. 51 Keawe-a-Hevhi, k. 52 Keohohiwa, w. 53 Aikanaka, k. 54 Keohokalole, w?. Uluae, w. Nanahapa, w. Nanahope, w. Elelni, w. Waohala, w. Kumukoa, 10. Umaumanana, w. Laikapa, w. Pililohai, w. Hauananaia, w. Koihouhoua, w. Hapuu, w. Maihikea, w. Naiolaukea, w. Wehelani, w. Henauulua, w. Keahiula, 10. Kelieau, w. Kaneakaleleoi, w. Lauli-a-Laa, k. Akepamaikalani, w. Pelea, w. Nononui, w. Kokalola, iv. Kanepukoa, iv. Kaipuholua, w. Keleanuinohoanaapiapi, w, Piilani, k. Umi-a-Liloa, k. Kunuu-nui-puawalau, w. Kapohelemai, w. Kawalu, w. Kaoui, w. Umiulaakaahuumanu, v. Ikuaana, to. Ululani, w. Kepookalani, k. Kamae, w. Kapaakea, k. 55 Pehekemana. Nanamua. Nanaikeauhaku. Keaoa. Hekumu. Umalei. Kalai. Malelewaa. Hopoe. Makalawena. Lelehooma. Kekupahaikala. Maweke. Mulielealii. Moikeha. Hookamalii. Kahai. Kuolono. Maelo, w. Laulihewa. Kahuoi. Puaakahuoi. Kukahiaililani. Mailikukahi. Kalona-nui. Kalona-iki. Kalamakua. Laielolielohe. Piikea. Kumulae. Makua. I. Ahu-a-I. Kapaihi-a-Ahu. Heulu. Keawe-a-:Heulu. Keohohiwa. Aikanaka. Keohokalole. Kalakaua. 190 THE POLYNESIAN RACE. The Ulu genealogy was the one most in vogue among Hawaii and Maui chiefs. It divides in two principal branches, the Puna, and LTe?na, and although, through sub- sequent intermarriages, every aristocratic family in the land can trace itself up to one or the other, and both ; yet, for reasons now not well understood or perhaps forgotten, the Hawaii and Maui chiefs, with peculiar pride and per- tinacity, preferred to ascend to Wakea on the Hema line, while the Kauai and Oahu chiefs clung to the Puna line with a pride and affection hardly less than that with which they regarded the Nana-ulu line just quoted. I will now insert the Ulu-Hema line as currently adopted in the time of Kamehameha I., and first published by David Malo in 1838, in his Moolelo Hawaii, " Hawaiian History." THE ULU GENEALOGY. 1 Wakea, k. ( Papa, w. \ Hookokukalani^ w. c Hoohokukalani, w. \ Haloa. 2 Haloa, k. Hinamanouluae, w. Waia. 3 Waia, k. Huhune, w. Hinanalo. 4 Hinanalo, k. Haunuu, w. Nanakehili. 5 Nanakehili, k. Kaulani, w. Wailoa. 6 Wailoa, k. Hikawaopuaianea, u 1. Kio. 7 Kio, k. Kamole, w. Ole. 8 Ole, k. Hai, w. Pupue. 9 Pupue, Tc. Kamahele, w. Manaku. 10 Manaku, k. Hikohaale, w. Kahiko. 1 1 Kahiko, k. Kaea, w. Luanuu. 12 Luanuu, k. Kawaamaukele, w. Kii. 1 3 Kii, k. Hiuakoula, w. Ulu. Nana-ulu. 14 Ulu, k. Kapunuu, w. ( Nana. < Kapulani. ( Nanaie. 1 5 Nanaie, k. Kahaumokuleia, w. Nanailani. 16 Nanailani, k. Hinakinau, w. Waikulani. 17 Waikulani, k. Kekauilani, w. Kuheleimoana. 18 Kuheleimoana, k. Mapunaiaala, w. Konohiki. 19 Konohiki, k. Hikaululena, w. Wawena. THE POLYNESIAN RACE. 191 20 Wawena, ft. Hinamahuia, w. Akalana. 21 Akalana, ft. Hinakawea, w. / Maui-mua. \ Maui-hope. j Mauikiikii. ' Mauiakalana. 22 Mauiakalana, k. Hinakealohaila, w. Nanaruaoa. 23 Nanaruaoa, k. Hinaikapaikua, w. Nanakulei. 24 Nanakulei, k. Kakaukuhonua, w. Nanakaoko. 25 Nanakaoko, k. Kahihiokalani, w. Heleipawa. 26 Heleipawa, A. KookookumaikalanijU'. Hulumanailani. 27 Hulumanailani, A;. Hinamaikalani. w. Aikanaka. 28 Aikanaka, k. ( Hinahanaia Kama- ( lama, w. ( Puna. \ Hema. 29 Hema, ft. Ulumahalioa, w. Kakai. 30 Kahai, k. Hinauluohia, w. Wahioloa. 31 Wahioloa, ft. Koolaukahili, w. Laka. 32 Laka, k. Hikawaelena, w. Luanuu. 33 Luanuu,/:. Kapokulaiula, w. Kamea. 34 Kamea, k. Popomaili, w. Pohukaina. 35 Pohukaina, k. Huahuakapalei, w. Hua. 36 Hua, ft. Hikimolulolea, w. Pau. 37 P ;iu , k Kapohaakia, w. Huanuikalalailai. 38 Huanuikalaiailai,ft. J Kapoea, w. Paumakua. \ Molehai, w. Kuhelani. 39 Paumakua, ft. Manokalililani, w. Haho. 40 Hako,ft. Kauilaianapa, w. Palena. 41 Palena, ft. Hikawainui, w. ( Haualaa-nui. ( Hanalaa-iki. 42 Hanalaanui, ft. Maliuia, w. Lanakawai. 43 Lanakawai, ft. Kololiialiiokawai, w. Laau. 44 Laau, ft. Kukamolimolialoha,w. Pili. 45 Pili, ft. Hinaauaku, u: Koa. 46 Koa, ft. Hinaauniai, w. Ole. 47 Ole, ft. Hinamailelii, w. Kukohou. 48 Kukohou, ft. Hinakeuki, w. Kaniuhi. 49 Kaniuhi, ft. Hiliamakani, u: Kanipahu. 50 Kanipahu, ft. ( Hualani, w. \ Alaikauakoko, w. Kalahuinoku. Kalapana. 51 Kalapana, ft. Makeamalamaiha- Kahainioeleaikaaiku- nae, w. pou. 52 Kaliaimoeleaikaai - Kapoakauluhailaa, w, . Kalaunuiohua. kupou, ft. 53 Kalaunuiokua, ft. Kalieke, w. Kuaiwa 192 THE POLYNESIAN RACE 54 Kuaiwa, k. Kamuleilani, w. 55 Kohoukapu, k. Laakapu, to, 56 KauholanuimahUj/jNeula, to. 57 Kiha, k. 58 Liloa, k. 59 Umi, k. Waoilea, w. 60 Kealiiokaloa, k. 61 Kukailani, k. ( Kohoukapu. < Hukulani. ( Manauea. Kauholanuimahu. Kiha. Liloa. ( Pinea, w. Hakau. ( Akahiakuleana, w. Umi. (Kulamea, to. KapunanahuanuiaumL Makaalua, to. Nohowaaumi. Kapukini, w. t Kealiiokaloa. } Kapulani. ( Keawenuiaumi. Piikea, to. \ Aihakoko. ( Kumalae. Makuahineapalaka, w. Kukailani. Kaohukiokalaui, w. 62 Makakaualii, k. Kapukamola, to. 60 Keawenuiaumi, k. Koihalawai, w. 61 Kanaloakuaana, k. Kaikilani, to. ( Kaikilani. ( Makakaualii. Iwikauikaua. Kanaloakuaaua. SKealiiokalani. Keakealanikane. Kalanioumi. Keakamahana. Keakealani. Keawe. Kalanikauleleikaiwi. Kalanikauleleiaiwi, w. ( Keeaumoku. ( Kekela. 66 Keeaumoku, k. Kamakaimoku, w. Kalanikupuapaikala- ninui. 66 Kekela, w. Haae, k. Kekuiapoiwa. 67 Kalanikupuapai- Kekuiapoiwa, to. 68 Kamehameha. kalaninui, k. 62 Keakealanikane, k. Kealiiokalani, w. 63 Iwikauikaua, k. Keakamahana, to. 64 Kanaloakapulehu,&Keakealaui, w. Kaneikauaiwilani,/: 65 Keawe, k. Such is the genealogy which the bards and priests at the Court of Kamehameha I. recited in his honour. Con- tinuing this line to the present time, from Keawe s half- sister, Kalanikauleleiaiwi, No. 61, we have 65 Kalanikaulele- iaiwi, w. 66 Keawepoepoe, k. 67 Kameeiamoku, k. 68 Kepookalani, k. 69 Aikanaka, k. 70 Keohokalole, w. 7 1 Kaldkaua, k. THE POLYNESIAN R Lonoikakaupu, k. Kanoena, w. Kamakaeheikuli, w. Keohohiwa, w. Kamae, w. Kapaakea, k. Kapiolani, w. ACE. Keawepoepoe. Kameeiamoku. Kepookalani. Aikanaka. Keohokalole. Kalakaua. *93 The other line of the Ulu-Hema genealogy, dividing at LTanalaa-iki, and attributed to the Maui chiefs, runs as follows : — ■ 42 Hanalaa-iki, k. 43 Mauiloa, k. 44 Alau, k. 45 Kaneinokuhealii,& 46 Lonoraai, k. 47 Wakalana, k. 48 Alo, k. 49 Kaheka, k. 50 Mapuleo, k. 5 1 Paukei, k. 52 Luakoa, k. 53 Kuliimana, k. 54 Kamaluohua, k. 55 Loe, k. 56 Kahaokuohua, k. 57 Kaulaliea, k. 58 Kakae, k. 59 Kaliekili, k. 60 Kawaokaohele, k. 61 Piilani, lc. 62 Kihapiilani, k. 63 Kamalalawalu, k. 64 Kauhi a Kama, I; 65 Kaulanikaumakao- ( KaneakanLi, wakea, k. I Makakuwahine, w. 66 Lonohonuakini, k. Kalanikauanakini- lani, w. 67 Kaulaliea, lc. Papaikaniau, w. 68 Kekaulike, k. Kekuiapoiwanui, w. Vol. l Kapukapu, w. Kaukua, w. Moikeaea, w. Keikauhale, w. Kolu, w. Kauai, w, Puhia, w. Maiaoula, w. Kamaiokalani, w. Painalea, w. Hinaapoapo. Kaumana, w. Kapu, w. Waoliaakuna, ic. Hikakaiula, w. Kapohanaaupuni, w. Kapohauola, w. Haukanuimaka- raaka, w. Kepalaoa, w. Laielohelohe, w. Kumaka, w. I'iilaniwahine, w. Kapukini, w. w. Mauiloa. Alau. Kanemokuhealii. Lononiai. Wakalana. Alo. Kaheka. Mapuleo. Paukei. Luakoa. Kuhimana. Kamaluohua. Loe. Kahaokuohua. Kaulahea. Kakae. Kahekili. Kawaokaohele. Piilani. Kihapiilani. Kamalalawalu. Kauhi a Kama. Kaulanikaumakao- wakea. Lonohonuakini. Umialiloa. Kaulahea. Kekaulike. Kahekili. N 194 THE POLYNESIAN RACE. 69 Kahekili, Jc. Kauwaliine, %o. Kalanikupule. 70 Kalanikupule, Jc. the last independent king of Maui, conquered by Kame- hameha I. Continuing this line, however, from Umialiloa, the brother of Lonohonuakini, No. 66, we descend to the present generation, as follows : — 66 Umialiloa, Jc. 67 Kuimiheua I., Jc 68 Niau, w. 69 Ululaui, w. 70 Keoholiiwa, %o. 71 Aikanaka, Jc. 72 Keohokalole, w. 73 Kalalcaua, Jc. Kuihewamakawalu,w. Kuimiheua. Kalanikueivvalono, w. Niau. Mokulani, Jc. Ululani. Keawe a Heulu, Jc. Keoholiiwa. Kepookalani, Jo. Aikanaka. Kamae, w. Keohokalole. Kapaahea, h. Kalakaua. Kapiolani, 10. The other branch of the Ulu genealogy, descending from Hernias brother Tuna, is equally voluminous and equally subject to different versions, between which great discre- pancy occurs. The one quoted by several ancient Haw- aiians, scil., Kamakau among others, and recited when rival heralds sang the praises of their chiefs, runs as follows : — 29 Puna-imua, 7c. Hainalau, w. Ua. 30 Ua, Jc. Kahilinai, w. UamaikalanL 31 Uamaikalani, Jc. Haimakalani, vj. Uanini. 32 Uanini, Jc. Welihaakona, w. Auanini. 33 Auanini, Jc. Maunakuahaoka- lani, ic. Newalani. 34 Newalani, Jc. Kahihiikaale, w. Lonohuanewa. 35 Lonohuanewa, Jc. Loiloa, w. Lonowahilani. 36 Lonowahilani, Jc. Kahikihaaueue, w. Pau. 27 Pau, Jc. Kapalakuakalani, w. Paumakua. 38 Paumakua, Jc. Keananui, w. Moeanaimua. 39 Moeanaimua, Jc. Alahoe, %v. Kumakaha. 40 Kumakaha, Jc. Moanaaulii, w. Nana. 41 Nana, Jc. Haakaleikini, w. Luahiwa. 42 Luahiwa, Jc. Kilohana, 10. Ahukai. 43 Ahukai, Jc. Keakamilo, w. Laa. 44 Laa, Jc. Kaikulani, to. Laamaikahiki. 45 Laamaikahiki, Jc. Hoakamaikapuai- helu, w. Lauli-a-Laa. THE POLYNESIAN RACE. *95 46 Lauli-a-Laa, k. 47 Laulihewa, k. Waolena, w. Manoopupaipai, w. Maelo, w. Ahukini-a-Laa. Kukona-Laa. Laulihewa. This is the same Lauliheica as No. 36 on the Nana-ulu genealogy. Other versions of this portion of the Puna line are considerably shorter, and hence, in my opinion, more correct. I now proceed to give the Puna line from Ahukini- a-Laa, one of the sons of Laamaikahiki, down to the pre- sent time, it comprises the descent of Kaumualii, the last independent king of the island of Kauai : — 46 Ahukini-a-Laa, k. Hai-a-Kamaio, w. Kamahano. 47 Kamahano, k. Kaaueanuiokalani, w, , Luanuu. 48 Luanuu, k. Kalanimoeikawai- kai, w. Kukona. 4Q Kukona, k. Laupuapuamaa, iv Manokalanipo. 50 Manokalanipo, /:. Naekapulani, to. Kaumakamano. 5 1 Kaumakamano, k. Kapoinukai, w. Kahakuakane. 52 Kaliakuakane, k. Manukaikoo, u: Kmvalupaukamoku 53 Kuwalupaukanio- Hameawahaula, w. Kab akumakapa weo. ku, k. 54 Kahakumakapa- weo, k. 55 Kalanikukuma, k. Kahakukukaena, v: Kalanikukuma. Kapoleikauila, u\ Ilihiwalani. 56 Ilihiwalani, k. Kaiuili, «7. Kauihi a Hiwa. 57 Kauihi a Hiwa, k. Kueluakawai, u\ Kaneiahaka. 58 Kaneiahaka, w. Kealoki, k. Kapulauki 59 Kapulauki, w. Kainaaila, k. Kuluina. 60 Kuluina. Kauakalrilau. Lonoikahaupu. 61 Lonoikaliaupu, k. Kamuokaurueheiwa,? '•. Kaumeheiwa. 62 Kaumeheiwa. Kaapuwai. Kamakahelei. 63 Kaiuakahelei, w. Kaeokulani, k. Kaumualii. 64 Kaumualii, /:. Kapuaamohu, w. Kinoike. 65 Kinoike, w. Kuliio, k. Kapiolani. 66 Kapiolani, w. Kalakaua. The Oahu chiefs, claiming descent under the Nana-ulu genealogy, mostly derive from Kalona-iki, the son of Mailikukahi, No. 40, and the Eoyal Kualii family line runs thus : — 196 THE POLYNESIAN RACE. 41 Kalona-iki, k. Kikenui-a-Ewa, w. 42 Piliwale, k. 43 Kukaniloko, w. 44 Kalaimanuia, w ! Piliwale, Jc. Kamaleamaka, 1c. Lo-Lale, k, Kukaniloko. Kalaimanuia. Kaihikapu. Paakanilea, w. Luaia, k. Lupekapukeahoina kalii, k. 45 Kaihikapu-a-Man- Kaunuiakanehoala- Kakuhihewa. uia, k. ni, w. 46 Kakukihewa, k. ( Kahaiaonuiakauai- ( Kanekapu. \ lana, w. \ Kaihikapu. v- Kaakaualani, w. Kauakahinui. 47 Kaihikapu-a-Ka- Kalua-a-Hookila, w. Kahoowaha, kuhihewa, k. 48 Kahoowahaoka- Kawelolauhuki, w. Kauakahi. lani, k. 49 Kauakahi-a-Ka- Makulua, w. Kualii. hoowaha, k. 50 Kualii, k. Kalanikahiniakei- Peleioholani alii, w. Kukuiaimakalani, w. 51 Peleioholani, k. Lonokahikini, w. Kumahana. 52 Kumahana, k. Kaneoneo. I am not aware that any lineal descendants of PelioJw- lani still survive, but there are numerous scions of the Kualii house, through his daughter Ktikuiaimdkalani and granddaughter Kalanipo, still alive in the fourth and fifth generation, thus bringing this line down to Nos. 55 and 56, corresponding exactly with the Kalona-nui branch of the Nanauhi genealogy. To reconcile these different genealogies is impossible ; to reconstruct them by the exercise of a proper criticism and with the light thrown upon them by the legends and chants still preserved — regarding the contemporaneity, intermarriages, wars, &c, of various chiefs on different lines — may be practicable, at least approximatively, and the result of my endeavours in that direction will appear in the following synchronical list of the Nana-ulu and Ulu lines. A few prefatory remarks, however, on these lines, as recorded on the Hawaiian genealogies, may be THE POL YNESIA N RA CE. 197 necessary for the better understanding of the list and the necessity, on behalf of historical truth, of reducing the Ulu line to more moderate proportions, and leaving it in an apparently incomplete condition, compared with the Nana-ulu. There are no legends of much historical value referring to the long line of chiefs from Nanaulu to and including Mawehe, embracing a period of fifteen generations, or about 450 years. Even the family — Nana — name had ceased to appear as a component part of the chiefs' names. But it is a significant fact, and of considerable importance, that out of all the genealogies of different Hawaiian chief families now known and recited, not one falls in upon the main line of either Nanaulu or Ulu above the time of Mawehe or Paumakua, with the exception of the Puna and LTema divisions of the Ulu line. From these two (Maiveke and Paumakua) the bare stems without colla- teral offshoots run up to Kii, and from him to Wakca. In their time, then, probably commencing some genera- tions earlier, certainly continuing several generations later, took place that general movement and displacement of Polynesian tribes which sent the Hawaiians southward, and the Southerners northward, in cpiest of new homes, adventures, or renown, of which the Hawaiian legends are so full and circumstantial. The Maiveke family, through Ids numerous sons and grandchildren, was probably the most powerful of the original chief families descended from Nanaulu, and were thus able to hold their ground against the intrusion and influence of the southern element, and retain their genealogy intact and unmixed ; while most, if not all the other chief families on the same line, of which scattered notices occur here and there in the legends, were gradually absorbed or superseded by the southern chiefs who claimed descent from Ulu, through Puna and Hema. Whatever legends may have existed, connected with names previous to Maiveke, they were apparently swal- lowed up and forgotten in the new era then inaugurated. 198 THE POLYNESIAN RACE. The historical value of the Ulu line, as recorded on Hawaiian genealogies, in the yvs-Maweke, ~pi'e-Paumakua period, is very small and very doubtful. In critically exa- mining the -post- Paumakua period, numerous opportuni- ties present themselves from time to time to compare the various genealogies which lead up to Paumakua among themselves, and with others that lead up to Maweke, as well as with the legends connected with the prominent chiefs of either line, thus testing their correctness, and enabling the inquirer to detect and adjust their inaccur- acies. The pre- Paumakua period furnishes neither so many nor so clear tests for historical criticism. The legends have grown into myths, and the myths have de- generated into fables. Still probabilities are not wanting, though very little can be definitely proven. It is certain that during the migratory period of which Maweke and Paumakua are the central figures, the Haw- aiian group was visited by expeditions from the Samoan, Society, and Marquesas groups, and that Hawaiian ex- peditions visited them in return. It thus appears both natural and probable that several different versions of the southern or Ulu legends and genealogies were introduced by the immigrant chiefs, their priests and followers, which, as the southern element became dominant and consoli- dated, were localised and incorporated on the general folk- lore of the Hawaiian group, and the different genealogies of the leaders of these expeditions were pieced together into one connected whole. Thus the Puna and Hema divisions of the Ulu line become so disproportionately longer than the Nanaulu straight line or its various branches. By counting upward from the present genera- tion, having due regard to the evidence furnished by the accompanying legends as to intermarriages and other social and political relations between the chiefs of the Nanaulu and Ulu lines, it becomes an undoubted historical fact that Maweke and Paumakua were contemporaries, the former being the twenty-seventh, and the latter the twenty-sixth, THE POLYNESIAN RACE. 199 generation from and inclusive of the present. If we now count from Nanaulu and Ulu, admitted by all genealogies and legends of both lines to have been brothers and sons of Kii, down to Maweke and Paumahua, we find only fifteen generations on the former line, and twenty-five and twenty-four respectively on the two divisions of the latter line, besides the discrepancies between the two divisions themselves, some making the Puna branch even longer than the Hema branch. It is almost certain that a number of names on the Ulu line were those of chiefs in some of the southern groups, who never set foot on Hawaiian soil, but whose names and whose legends were imported by southern emigrants, and to whom dynastic ambition and national vanity after- wards assigned a locus standi on Hawaiian legends, and a birthplace and burialplace on the Hawaiian group. Glimpses of southern legends and genealogies in New Zealand, Tonga, Samoa, Society, and Marquesas groups confirm this proposition still more. The Maui legends, the Maui family of four brothers, and their parent A-Icalana, Karana, or Taranga, and the grandmother Hina-mahuia, are found upon all those groups in slightly different ver- sions. The legend of Maui-kiihii or Maui-tiki-tiki, the youngest of the family, being out fishing, and catching the various Hawaiian islands on his hook, attempting to drag them ashore at Hilo and join them to Hawaii, is found nearly literally the same on New Zealand. On Tonga the same legend obtained, but they ascribe the act to Tangaloa instead of Maid. Near Puuepa, district of N. Kohala Hawaii, a stone is still shown which is said to bear the impress of Maui's fish-hook called Manaiakalana. Near the south end of Hawke's Bay, in the district of Here- taunga, New Zealand, Maui's fish-hook is still said to be preserved ; and at Tonga, a place called Hounga is pointed out as the spot where the hook caught in the rocks, and the hook itself was said to have still been in the pos- session of the Tut- Tonga family some thirty years before 20O THE POL YNESIAN RA CE. Mariner's time and residence on that group. Maui's ex- ploits in discovering fire are the common property, under various versions, of all the Polynesian groups. The deified ancestress and grandmother of Maui, in the New Zealand legends called Mahuika, is evidently the same as his grand- mother Hina-Mahuia on the Hawaiian Ulu genealogy ; and the Samoan Mafuie betrays a confused reminiscence of the same legends. These legends were undoubtedly older than the Polynesian exodus into the Pacific. On Borneo a legend still exists that that island formerly was composed of a number of smaller islands, which by some miracu- lous process were joined together. It is just to conclude, therefore, that the Maui family and legends were not only not indigenous to Hawaiian soil or contemporary with any chiefs on the Nanaulu line, but it is very questionable whether their origin does not date back to the pre-Pacific period of the Polynesian race. The next interpolation or, rather, insertion in the wrong place of the Ulu line, to which I will call attention, is the Nana family ending, or, according to the royal Hawaiian genealogy published by D. Malo, and referred to above on page 191, continued through Heleipawa, though several other genealogies end that family with a chief called Kapawa, which is no doubt the correct version. Accord- ing to one genealogy there were four Nana preceding Kapawa ; according to another there were three Nana ; and according to the old genealogy above referred to, page 184, there were but two Nana preceding Kapawa. I think there is little doubt that this family and their descendant Kapawa were actual chiefs on the Hawaiian group. The building up and consecration of Kukanilolzo, on the island of Oahu, that peculiarly hallowed place in all subsequent ages of Hawaiian history as the birthplace of the highest " Kapu " chiefs, is universally and continu- ously ascribed to Kapawa's father Nanahaoho. As to the. time of Kapawa, the legend of Paao — a Southerner of great rank and a high-priest, whose family was established THE POLYNESIAN RACE. 201 during this MaiceJce-PaumaJcua period as par excellence the priestly caste, and whose descendants survive to this day — expressly confirms Kapawa's contemporaneity with this migratory period. The legend states that when Pili-Jcaiaea arrived from Tahiti or Kahiki, " the Nana chiefs of Hawaii were extinct on account of the crimes of Kapawa, the chief of Hawaii at that time : " Ua pan na Alii Nana o Hawaii-nei i Jcahewa Kapaiva, he Alii Haivaii ia manaiua. What this great crime or fault may have been is not stated. Paao, the high priest, who had then already arrived and established himself, sent to Kahiki, that foreign, southern land, for Pili, who, on his invitation and through his instrumentality, became a king on Hawaii. Thus Kapaiva and Pili were contemporaries, and Kapawa's grandfather, Nana-maoa, was contemporary with or of the period of PaumaJcua ; and the family was probably of that same Southern, Ulu, descent as Puna and PaumaJcua, as, though living for some generations on the Hawaiian group previous to Pili, they were never included on the original Hawaiian Nanaulu line. In confirmation of this Southern extraction of the Kapawa family, several legends give strong, though inferential testimony. Thus Hina-i-Jcapa- ikua, the wife of Nana-maoa and grandmother of Kapawa, is also called the grandmother of Niheu-JcoloJie, who was the recognised grandson of Kuheailani, the brother of Pau- maJcua. Thus the same lady is called the grandmother of Kaulu, sometimes called Kaulu-a-Kalana, the renowned navigator and explorer of those days, whose astrologer and soothsayer — Kilo-kilo — named Luliau- Kapawa, is admitted to have been a man from KaJiiJci, and is by some said to have introduced the "Kapu" system in the Hawaiian group ; and they are both stated to have been contem- porary with KaJiiwa-Jcaapu, the wife of Hina-Jcai-mauli- awa, the grandson of MaweJce. I feel justified, therefore, in placing Kapawa within the period of Maweke's and Paumakua's grandchildren, and as contemporary with Pili. The Kapawa family, whether consisting of two or 202 THE POL YNESIA N RA CE. more Nana previous to Kapawa, is evidently greatly mis- placed on the Ulu-Hema genealogy, and belongs to the latter instead of the earlier portion of the line. Still another large excision must be made from the Ulu line, as represented on the Hawaiian genealogies, previous to the Paumakiia period. The four first names of the Hema division of the Ulu line, though referred to in song and saga as heroes of Hawaiian birth, were really but another importation and adaptation by that oft-mentioned southern element of their own legends and genealogies to their altered circumstances in the Hawaiian group. The Hawaiian genealogies make Puna and Hema sons of Aikanaka and his mythical wife Hina-hanaiakama- lama, with the cognomen Lonomoku, and descendants on the Ulu line from Heleipawa, or, as others say, Kapawa. The longest genealogies introduce nine generations, the shortest four, between Puna and Hema respectively and Paumakua, whom both divisions claim as a common ancestor for succeeding generations. The same uncer- tainty obtains on both divisions from Paumakua to the time of Pili, some having two, others four, and still others five generations during that period. It is significant, moreover, and to be observed, that no prominent name on the Ulu line, previous to Paumakua, occurs upon the legends connected with the Nanaulu line, except the Puna family of Kauai, who claimed to be, and probably were, of the Ulu descent, and with whom Maweke's grand- son Moikeha allied himself after his return from Kahiki. No crossings of intermarriages, no intercourse of peace or war, are recorded as having occurred between the two lines. They appear to have been mutually ignorant of each other's existence ; and yet the Ulu legends represent the Ulu chiefs of this pre-Paumakua period as having been born, lived, and died side by side of the Nanaulu chiefs, whose bare names have been preserved through the Maweke family, but whose legends were obliterated and forgotten in the superior eclat and later introduction of THE POL YNESIA N RA CE. 203 that southern, Ulu, element with its own peculiar genea- logies, legends, and innovations of various kinds. In comparing the New Zealand legends, as published by Sir George Grey, I find that the New Zealanders claim descent from the island of Sawaii in the Samoan group, which they pronounce Hawaiki, and that among other prominent names occurring in their ancestral tales, pre- vious to their departure from Hawaiki, are four that ap- pear also on the Hawaiian Ulu line between Aikanaka and Paumakua. In the New Zealand legends they appear. as chiefs or Ariki of Hawaiki, following one another in the same succession as on the Hawaiian genealogy. Their names are — the Hawaiian pronunciation in brackets — Hema [Hema], Tawhaki [Kahai], Wahieroa [Wahieloa], liaka [Laka]. Each of these chiefs have been natura- lised, so to say, and localised on the Hawaiian group by Hawaiian legends ; yet as there is no reasonable proba- bility that the New Zealanders took their departure from the Hawaiian instead of the Samoan group, and as their evidence is positive as to the residence of those chiefs on the Hawaiki, which they knew and from which they departed for New Zealand, I am forced to conclude that their introduction on the Hawaiian genealogies was the work of that migratory period, to which I have so often alluded, and was a local adaptation in after ages of pre- viously existing legends, when the memory of the mother- country had become indistinct, and when little more was known of them except the one main fact that they stood on the genealogical list of the Hawaiian chiefs of the Ulu line; a fact which was never allowed to be forgotten under the old system, however much local associations mi^ht be forgotten or altered. It is hardly historically possible that there could have been two series of chiefs in the Samoan and Hawaiian groups, with identical names and in the same succession ; with one transposition alone, the same identity holds good in the names of three of their wives, viz. : — 204- THE POLYNESIAN RACE. New Zealand. Hawaiian. Heraa, k. Uru-tcmga, w. Hema, k. Ulu-mahelioa, w. Tawhaki, k. Hine-piri-piri, w. Kahai, k. Hina-uluohia, w. "Wahieroa, k. Kura, w. Wahieloa, k. Koolaukahili, w. Raka, k. Tongarautawhiri, w. Laka, k. Hikawaelena, w. Thus, on the testimony of the New Zealand legends, these chiefs were not original on the North Pacific Hawaii, of which the New Zealanders knew apparently nothing, but on the South Pacific Samoan Sawaii, from which they claimed descent, from which they emigrated, and whose legends they brought with them to their new homes in Ao-tea-roa or New Zealand. Thus, while the Nanaulu genealogy for the period be- tween Kii and Maiveke has been assailed by no doubts and by no diversity of opinion among subsequent generations of Hawaiians, the Ulu genealogy has been subject to numerous varying constructions, no two agreeing together throughout, and as a Hawaiian genealogy on Hawaiian soil is disproven in several places by its own discrepancy with the Nanaulu line, as well as by the direct testimony of the legends and genealogies of the South Pacific groups. With these introductory remarks I will now give the Hawaiian genealogy from Nanaulu and Ulu down, as I consider it ought to be rendered, when the sources have been critically examined and properly collated ; showing at the same time the collateral branches both of the Puna and Hema lines of the JJla division, as well as the main branches of the Nanaulu-Mawcke division. (For Genealo- gical Table, see Appendix No. IX.) Of the legends which treat of Wahea and his wife Papa, not much bearing the impress of ancient and original tradition has been preserved. What has been preserved, however, establishes the fact, as previously noticed, that Wahea was a chief on one of the Molucca islands (Gilolo), previous to, perhaps contemporary with, the great exodus THE POLYNESIAN RACE. 205 of the Polynesian family from the Asiatic Archipelago. His reign seems to have been chequered by wars and re- verses. Certain great changes in the social system of the people, the strengthening of the Kapus and the introduc- tion of new ones, are vaguely ascribed to him. His life seems to have been troubled by rebellion at home and by foreign pressure from without. The domestic relations between him and his wife Papa appear to have been very unfortunate, and form by far the greatest portion of the subject-matter of the legends referring to those personages. Wakea, however, seems not to have been without defenders of his good name, for there were legends existing in David Malo's time, say fifty years ago, which asserted that Hoo- hokukalani, the reputed, and on the most prevalent genea- logies recognised, daughter of Wakea and Papa, was not their child at all, but was the daughter of Wakea' s high priest Komoawa and Ins wife Pojjokolonuha ; and I have one genealogy which, while it recognises Hooliokukalani as the daughter of Papa and Wakea, gives her Manauluae as husband and Waia as their son. The domestic scandal of Wakea 's incest, on which later versions of the Wakea legends lay so much emphasis, appears therefore not to have been fully believed in more ancient times, and I feel justified in considering it as an unfounded gravamen of a character remembered only by succeeding generations for its oppres- siveness and tyranny. I find no personal description in the legends of Wakea, but Papa is represented as a comely woman, " very fair and almost white." She is said to have become crazy or distracted on account of her domestic troubles with her husband, who publicly divorced her, according to ancient custom, by " spitting in her face." She is represented as having lived to a very old age, and as having died in Waicri, a place in Tahiti. In after ages she was deified under the name of Haumea. Of the remaining names from Wakea to Nanaulu and ITlu, there exist no legends of any historical value. On the whereabouts of their residence and the exploits of their 206 THE POLYNESIAN RACE. lives, tradition is apparently silent ; though some Hawaiian commentators of the ancient legends, according to D. Malo, asserted that the first six lived in Kahihi (Tahiti), that is in some foreign land outside of the Hawaiian group. Of Kii, No. 1 3 on the list, and the last of the first series, nothing is known except that he was the father 01 the two brothers Nanaulu and Ulu, from whom the northern and southern Polynesians respectively claimed their descent, and in whose time the probable separation of the two branches took place ; the Nanaulu branch pro- ceeding northward and settling on the Hawaiian group with a possible sejour or rest on the Marquesas group, though nothing in the legends remain to indicate such a fact ; and the Ulu branch remaining on the islands of the South Pacific, keeping up a not unfrequent intercourse between them, forgetting or ignoring their northern brethren for a period that may be roughly stated to have extended over ten to twelve generations. Towards the close of this period, from Nanaulu to MdwcJce, as a central figure, the Hawaiian seclusion or isolation was interrupted by the arrival of sundry parties from the South, or, as the legends call it, Kahiki, claim- ing descent from Ulu through either the Puna or Hema division. Such were the Puna family established on Kauai, with whom Maivekes grandson Moikeha allied him- self. Such were the Nanamaoa or Nana-a-Maui family established on Oahu, and obtaining ascendancy for some time on Hawaii. Such the Huanuikalalailai family, whom the Hema division in subsequent times claimed as the father of Paumakua, and probably others whose legends have perished. What particular groups of southern islands those emigrants came from no vestiges on existing legends remain to indicate. It is probable that the Puna family came from or through the Marquesas group. The name is familiar and common on the Marquesan genealogy in my possession about thirty-five to forty-two generations THE POL YNESIA N RA CE. 207 ago, and may have been the stock from which the Haw- aiian Punas descended. It is almost certain, taking the concurrent testimony of the legends as the arbitrium of conflicting genealogies, that several of those earlier names mentioned on the Ulu line, both before and after the Puna-LTema divisions, were con- temporary. Thus the genealogies represent Laau-alii as the father of Pili-kaiaea, but the legends are unanimous that Pill came to Hawaii from Kahiki in the time of Laaualii, and that Pili succeeded Kapawa in the govern- ment of Hawaii. Hence Laaualii and Kapawa must have been contemporary. Thus JSua, or, as he is otherwise called, Hua-a-kapuaimanaku, who on the genealogies is placed as the grandfather of Huanui-Kalalailai, was in reality, according to the legends and the known contem- poraneity of his associates, at least five generations later. Discarding, therefore, the earlier portion of the Ulu line as of any historical value in the reconstruction of Hawai- ian genealogies for the period previous to Mawekc and Paumakua, the uncontested Nanaulu line remains for our guidance, showing a period of fifteen generations previous to Maweke, during which the Hawaiian group was in- habited by the Polynesian race, practising under its own line of chiefs the customs and religion which they brought with them. It is now nearly impossible to separate those customs and that religion from what subsequently ob- tained, after the great migratory wave of the eleventh century had passed over Polynesia and thoroughly inun- dated the Hawaiian group with a new order of things. Glimpses, however, of the former condition of the Hawai- ian portion of the Polynesian family appear here and there in the legends immediately relating to this migratory period ; and from a careful inquiry into their contents and bearing, I am led to believe that the Katie worship in greater simplicity, with the customs it enjoined or which grew out of it, and notices of which have been given in 208 THE POLYNESIAN RACE. previous pages, was the prevalent creed of those ancient Hawaiians ; that the Kapus were few and the ceremonials easy ; that human sacrifices were not practised, and cannibalism unknown ; and that government was more of a patriarchal than of a regal nature. By counting upwards from the present time, the Hawai- ian genealogies and legends have enabled me to establish, approximatively, the period of Wakea at about the middle or latter part of the second century a.d. But in examin- ing the genealogies bearing upon the pre-Wakea period, it becomes evident that the thirty-seven generations embraced upon even the longest of them — that from Kumuhonua, or the first man, to Wakca, — is entirely inadequate to represent the continued existence of the human race during that interval, and that there must be large and important gaps in that genealogy. All the other Hawaiian genealogies, covering the pre-Wakea period, are equally if not more defective. There is evidently a large gap among the generations immediately succeeding the twelve sons of Kinilau. Up to that time I look upon the Kumuhonua genealogy as merely a reflex of the Cushite knowledge and Cushite reminiscences im- parted to the Polynesians while yet sojourning in India, or during their early residence in the Asiatic Archipelago. A number of families in ancient history seem to have adopted this distant and mysterious twelve-ship as their national point of departure, and to have carried it with them wherever they spread. All previous to that was to them a common heirloom ; all subsequent became national divergence, complexity, confusion, and oblivion. The manifest relation, in many places, between this genealogy (the Kumuhonua) and Arabian, Chaldeo-Hebraic, and other Eastern genealogies and legends, clearly proves the common origin of them all. The ten generations between the sons of Kinilau and the time of Wakea must hence cover a period of some thousands of years. They repre- sent probably only the most prominent figures on the THE POLYNESIAN RACE. 209 traditional canvass of that half-forgotten period, including the discoveries and exploits of Hawaii-loa. It is a period extending from their connection with, and absorption of, the Cushite element, to their expulsion from the Asiatic Archipelago by the Malays. Having thus attempted to show who the Polynesians are, whence they came, and how connected with the old- world peoples of historic renown, I will in the next portion of this work endeavour to sketch Hawaiian history proper, from the period of Maweke and Paumakua to the times of Kamehameha I. — so far as such history may be gathered, from a critical research into the legends, traditions, songs, and genealogies of the Hawaiian people. It will be observed that in this sketch of the Polynesian race I have not referred to the tribes occupying Western Polynesia, the Micronesian and Caroline groups. I am not acquainted with their languages, and very imperfectly with their traditions. That they are a branch of the same great race there can be little doubt ; but they were pro- bably of much later separation from the Asiatic Archi- pelago, and had been either there, or have been subsequently, subjected to intermixture with foreign elements to so great an extent as to destroy the Polynesian character of the language, and that general homogeneity of customs and traditions which is so conspicuous a link of connection between the Southern, Eastern, and Northern groups of Polynesia. VOL. T. APPENDIX. No. I., page 63. Some writers, supporting themselves by a legend said to have come from Hawaii, that the world was produced from an egg. 1 I find a relation of that legend to the Brahminical doctrine of the World-egg. I have been unable to discover or collect such a legend on the Hawaiian group ; nor do I know of any resembling it on the other island groups, unless in some distorted form it may refer to the Fiji legend which says that mankind sprang from two eggs that were hatched by the god Ndengei. There is a Hawaiian legend, however, which ascribes the creation of the world to Wakea and Papa in this way : " Papa, the wife of Wakea, begat a calabash — ipu — including bowl and cover. Wakea threw the cover upward, and it became the heaven. From the inside meat 1 In a "Journal of a Tour around a woman, a hog and a dog, and a pair Hawaii," "by a deputation from the of fowls, came in a canoe from the Mission of those islands," Boston, Society Islands, landed on the east- 1825 — of which deputation Rev. Mr. ern shores, and were the progenitors Ellis, of Polynesian and Madagascar of the present inhabitants." It is fame, was one — it is said, page 197, much to be regretted that this tradi- that in a conversation with Mr. John tion, of which Mr. Young gave only a Young, who had resided on the meagre resume, should have utterly islands since 1790, "Mr. Young said perished from the land during the the natives had several traditions, fifty years since the above "Tour one of which was, that an immense around Hawaii." It certainly must bird laid an egg on the water, which have been an earlier and purer form soon burst and produced the island of of the subsequent tradition of Papa Hawaii, and shortly after a man and and her calabash. 2 1 2 THE POL YNESIAN RA CE. and seeds Wakea made the sun, moon, stars, and sky ; fiom the juice he made the rain, and from the bowl he made the land and the sea." I now give the Brahminical account of creation, as gathered from an " Analysis of the Code of Menu," pub- lished in the "Asiatic Journal," ' November 1827. It says : — " The universe existed only in darkness, imperceptible, undefinable, as if wholly immersed in sleep. The self- existing power, himself undiscerned, with five elements and other principles, appeared in glory, dispelling the gloom. ' He, whom the mind alone can perceive, whose essence eludes the external organs, who has no visible parts, who exists from eternity, even he, the soul of all beings, whom no being can comprehend, shone forth in person.' Having willed to produce various beings from his own divine substance, first with a thought he created the waters ; ' the waters are called ndrd, because they were the production of Nara, or the spirit of God ; and since they were his first ayana, or place of motion, he thence is named N&r&yana, or moving on the waters.' .... The Hindu legislator proceeds to tell, that the self- existing power placed a productive seed in the waters, which became ' an egg, bright as gold, blazing like the luminary with a thousand beams,' from whence he was born himself, ' the divine male, famed in all worlds under the appellation of Brahma.' " Compare with the above the following extract from the Manelc-Maya, the classical work on Javanese mythology before the introduction of Mohammedanism. I quote from " Oceanie," par M. de Pdenzi, vol. i. p. 75. " Avant que les cieux et la terre fussent cree's, Sang- yang-tvisesa (le Tout-Puissant) existait. Cette divinity, etait placee au centre de l'univers ; elle desira interieure- ment que le Eegulateur supreme lui accordat un souhait. Aussitot tous les elements se heurterent, et ii entendit, au milieu d'eux, une repetition de sons semblable au batte- THE POL YNESIA N RA CE. 2 1 3 merit rapide d'une cloclie. II leva les yeux, et il vit un globe suspendu au-dessus de sa tete, il le prit et le separa en trois parties : une partie devint les cieux et la terre, une autre partie devint le soleil et la lune, et la troisieme fut l'liomme, ou Manek-Maya. La volonte de Sang-yang- wisesa ayant ete accomplie, il voulut bien parler a, Manek- Maya, et lui dit : Tu seras applet Sang-yang-gouron ; je place une entiere confiance en toi; je te donne la terre et tout ce qui en depend, afin que tu en uses et que tu en disposes selon ton plaisir. Apres ces paroles, le Tout- Puissant disparut." The Hawaiian legend, as I have collected it, is possibly a corruption of the Javanese myth ; but whether either of them refers to Menu's account of creation for their origin, may, I think, admit of a doubt, unless the Brahminical account itself is a copy, or a compromise, of some pre- viously existing Cushite-Dravidian cast of thought, har- dened into myth or legend. The chaotic condition of the world, the Narayana or " moving on the waters," referred to by Menu, are certainly Cushite modes of thought, and bespeak their kindred to the Polynesian Po, and the Tahitian Tino- Taata} and the Hawaiian Lono-nolio-i-ka- wai? as well as with the Egyptian Koub and his mys- terious boat. 1 Vide, p. 64. 2 Vide, p. 94. ( 214) "No. II., page 63. Te Vanana na Tanaoa. In the beginning, space and companions. I te tumu ona-ona a na hoa. Space was the high heaven. Ona-ona oia te iku-ani, Tanaoa filled and dwelt in the whole heavens. Tanaoa 1 hahapi a nonoho i na ani otoa And Mutuhei was entwined above. A Mutuhei 2 ua hei ma una, There was no voice, there was no sound ; Koe na eeo, koe na tani, No living things were moving. Aoe e ae na mea polioe, There was no day, there was no light. Koe na A, maama hoe A dark, black night. He tano-tano ke-ke po, O Tanaoa he ruled the night. Tanaoa vivini-ia te po, Mutuhei was a Spirit pervading and vast. 10 Mutuhei uhane vae-vae a oa. From within Tanaoa came forth Atea. Mei ioto Tanaoa tihe ae Atea, 5 Life vigorous, power great. Pohoe oko, mana nut, O Atea he ruled the day Atea vivini-ia te'A, And drove away Tanaoa. A tatai pu ia Tanaoa. 1 Darkness. ' Silence. 3 Light, the Sun THE POL YNESIA N RACE. 215 Between Day and Night, Atea and Tanaoa, / vavena o te A me po, a A tea me Tanaoa Sprang up wars, fierce and long. Tupu ae na toua a-ha oa-oa, Atea and Tanaoa, great wrath and contention. Atea a Tanaoa, a-ha nui a nanaku Tanaoa confined, Atea soared onward, Tanaoa tamau ae, Atea hee anatu Tanaoa dark as ever. Tanaoa kelce pe ananu Atea very good and very active. 20 Atea meita meitai a ta-ana-ana From within Atea came forth Ono. Mei ioto Atea taha-taha ae te Ono 1 O Ono he ruled the sound Ono vivini 'ia te tani And broke up Mutuhei. A tafati-fati 'ia Mutuhei Here a great division was made / tenei he pahei nui haka' ia In the company of Atanua. / na hoa o Atanua 2 Atanua was beautiful and good Atanua pootu a meitai Adorned with riches very great. Tapi i taia tae-tae ma-iko-iko Atanua was fair, very rich and soft. Atanua teea, taetae nui a peehu, Atea and Atanua embraced each other. Atea me Atanua popoho'ia kohua Atanua produces abundantly of living things. 30 Atanua tupu oko i na mea pohoe Atea took Atanua for wife. Atea too'ia Atanua mea vahine Atea and Ono pass onward, pass upward. Atea me Ono hee anatu, hee ma una Atea the body, Ono the Spirit. Atea tino, Uhane Ono 1 Articulated sound, the voice. 2 The Dawn. 1 16 THE POL YNESIAN RA CE. Atea with Ono in one place. Atea me Ono etahi ona Atea the substance, Ono the Atea tono, moid Ono Atea produces the very hot fire. Atea tupu i te alii vea-vea Ono is powerful and great Ono mana oko nunui' ia Atea is adorned with riches changeable and dazzling:. Atea tapi i te taetae take take a ponio-nio Ono is adorned with princely wealth and power. Ono tapi i te taetae Hakaiki me te mana They two the same glory. 40 Aua eua etahi koaa. Atea the body, Ono the Spirit. Atea te tino, Uhane Ono Atea the substance, Ono the Atea te tono, moid Ono, And dwelt as kings in the most beautiful places A nonoho hakaiki na ona meitai oko Supported on thrones, large, many-coloured, wondrous. Hakatu mauna na paipai nui take take a-anaau They dwelt above, they dwelt beyond. A 110J10 una, a nonohu atu They ruled the space of heaven, Mea haatoitoi te va-vae-ani And the large entire sky, Me na ikuani nui otoa And all the powers thereof. A m.e otoa na mana i ke ia The first Lords dwelling on high. Te tau Fatit o'mua nonoho tikitiki, wondrous thrones, good and bright 50 E na paipai aanaau meitai ta-ana-ana wondrous thrones, whereon to seat the great Lord Atea. E na paipai aanaau mea paipai ia te Fatu-nui Atea. O thrones placed in the middle of the upper heavens. E na paipai hakatu i vavena te ani una THE POL YNESIAN RA CE. 217 O thrones whereon to seat the Lord of love; E na paipai mea paipai 'ia te Fatti te hina-nau The great Lord Atea established in love, Te Fatu nui Atea haatuia i te hinanau To love the fair Atanua. Mea hinanau 'ia te pootu Atanua, Atanua shades the neck of Atea. Man kaki Atanua no Atea A woman of great wealth is Atanua, He vahine taetae nut Atanua Which she brought from out of night, Toi mai 'ia met ioto te po Gathered for Atea. Hai-hai 'ia mai no Atea Nothing was given back to night, 60 Aoe he mea tuu atu no te po Atea gave nothing back to Tanaoa, Aoe he mea no Atea tuu atu no Tanaoa,. Who thus was chased to distant regions, Pehea tatai 'ia vahi oa Where the light of day was not known ; Koe e itea te ao-o-te a No wealth, no warmth ; Taetae koe, mahanahana koe, Confined, lying beneath the feat of Atanua, Tamau moeana iao te tapu vae no Atanua Very cold, dreary, dark, without companions ; Anu oko aa-naho Kevokevo koe na hoa Nothing of all his wealth remained. Koe to'ia taetae a na mea otoa Cold, shivering, engulfed ; behold indeed ! Anu kamaiko uuku ia aa ehoa. O dark Tanaoa engulfed in the long nights E keke Tanaoa uuku ia i na po a oa Secure sits Atea on his wondrous throne, 70 Mau Atea una to'ia paipai aanaau, And dwells as Chief in his domains. Nonoho Hakaiki i to'ia pai aina 2 1 8 THE POL YNESIAN RA CE. Born is his first son, his princely son. Tupu to'ia tama mua, to'ia tama Hakaiki. O the great Prince, O the sacred superior. te Hakaiki nui o te una tapu O the princely son, first born of divine power ! te tama Hakaiki fanau mua o te mana na etua O the Lord of everything, here, there, and always. te Hakaiki o na mea otoa eia aia a e ia mai a oa O the Lord of the heavens and the entire sky. te Hakaiki o te vaevaani a na ikuani otoa O the princely son, first born of the exalted power. te tama Hakaiki fanau mua o te mana tiki-tiki. O the son, equal with the father and with Ono. te tama tia me te motua a me Ono Dwelling in the same place. Etahi ona a te nonoho Joined are they three in the same power, 8o Poho 'ia toko tou etahi koaa. The Father, Ono, and the Son. Te motua, Ono a te Tama • One tree (trunk, cause) was formed from those three. Te tumu tahi koaa mei na toko tou The tree producing in the heavens above Te tumu tupu ia i te vaevaani una All the good and wondrous families in love. / te tau huaa meitai aanaau i te hina-nau The tree of life, firm rooted in heaven above Te tumu o te pohoe, man te aka i te ani una : The tree producing in all the heavens Te tumu tupu i te ani otoa The bright and sprightly sons. / na tau tama poninnio a ta anaana From Atea they were born as his sons. No Atea hakatupu nui ia atou i te tama Atea, the exalted Lord of everything ! Atea te Fatu tikitiki o na mea otoa O Atea, their life, body, and spirit. 90 Atea to atou pohoe, tino, moui a uhane. THE POLYNESIAN RACE. 219 The foregoing chant is extremely valuable as a relic of Polynesian folklore. It is now impossible to determine the age of its composition ; but to judge from the rugged- ness of its diction, it must be of very high antiquity. It is an allegory, no doubt, but the consciousness of its being an allegory had not yet faded from the mind of the com- poser, nor, perhaps, from the people before whom it was chanted. It points to a period of the human mind when the thoughts of sa^es still lingered and laboured in the border land between material facts and metaphysical ab- stractions ; when Tanaoa was still half the real darkness of night, chaos, and half the deified impersonation of an evil principle, antagonistic to the powers of light; when Aten, was still half the actual sun, springing forth from, succeed- ing to, and dispelling the gloom and darkness of night, and half the deified impersonation of creative power ; when Ono was still half the mere actual sound, the busy hum of a living, active, moving world, just awakening from the torpor and silence of night, and half the deified imper- sonation of speech and intelligent communication, an evolution of, and a companion to, Atea ; when Atanua was still the mere Dawn, the result of the apparent contest between Darkness and Light, " encircling the neck of the sun," as well as the goddess wife of Atea. This chant must be at least as old as the period when the Vedic poets sang the praises of Indra and the charms of Ushas. It sounds like a lost hymn of the Vedas, or, perhaps, of the pre- Vedic period. Its whole tenor, style, and imagery are thoroughly Arian. Even here the conception of a tripli- cate Godhead occurs : perhaps the prototype of the Chaldean Anu, Bel, Hea, as well as of the Indian Tri- murti, and is but another version of the Hawaiian Kane, Ku, Lono. ( 220 ) No. III., page 63. Probably one of the grandest religious poems, once cur- rent among the Polynesians, and relating to the creation of the world, is that which Mr. Moerenheut has preserved in his " Voyage aux isles du Grand Ocean." Though but a fragment of what was probably a series of religious poems, yet its lofty tone and archaic simplicity of expres- sion make it extremely valuable as a testimony to the ancient belief of the Polynesians. As published by Mr. Moerenheut there are several errors of orthography which I have endeavoured to correct ; but there are also some other unintelligible parts, whether owing to a bad manu- script copy or to careless printing, which I have enclosed in brackets, being unable to give an English translation of the same. Those who have an opportunity and are com- petent to compare Mr. Moerenheut's translation with the original and with mine, will perceive that though his is more florid and free, and mine more literal, yet the spirit of the poem is fully preserved. The poem accords so thoroughly with the Marquesan and Hawaiian poems on the same subject, that there can be no doubt of its very great antiquity, although the introduction of Taaroa as the Great Creator would seem to indicate a later period for its composition than that of the Hawaiian and Mar- quesan chants on creation and cognate subjects. I am unable, at present, to indicate the period of Polynesian life, when the attributes and powers of Kane, or Tane, or Atea (for they are but synonyms of the same conception) were transferred to Taaroa or Tangaroa, who, to judge THE POL \ r NESIA N RA CE. 221 from the Hawaiian and Marquesan folklore, was ori- ginally conceived of as the very opposite in attributes and functions. It is admitted even in Tahitian folklore, that at some remote period the Tone worship was superseded by, and subordinated to, that of Ono on nearly all the islands of the Society group except Huahine ; and at that time probably the legend arose which made Tarie and Ono to be brothers, and sons of Taaroa. With these considerations, and others that have been set forth elsewhere in this work, I am satisfied that this Tahitian chant of creation is older than the period when Taaroa was elevated by the southern groups into the primacy of Godhead, and that its intrinsic evidence con- nects it with the remarkable series of ancient chants, once common to the Polynesian race as an heirloom from the past, of purer creed and loftier conceptions, and of which the Marquesans and Hawaiians have preserved such inter- esting relics. For better reference I have interlineated the trans- lation. He abides — Taaroa by name — Paraki, Taaroa te ioa, In the immensity of space. Roto ia te aere. There was no earth, there was no heaven, Aita fenua, aita rai, There was no sea, there was no mankind. Aita tai, aita taata. Taaroa calls on high ; Tiaoro Taaroa i nia; He changed himself fully. . . . Fuariro noa ihora oia {i te ohe narea ei). Taaroa is the root; Te tiimu Taaroa ; The rocks (or foundation) ; Te papa ; Taaroa is the sands : Taaroa te one ; 222 THE POLYNESIAN RACE. Taaroa stretches out the branches (is wide-spreading). Toro Taaroa ia naio. Taaroa is the light ; Taaroa tei te ao ; Taaroa is within ; Taaroa tei roto ; Taaroa is . . . Taaroa (te nahora ;) 1 Taaroa is below; Taaroa tei raro ; Taaroa is enduring ; Taaroa te taii ; Taaroa is wise ; Taaroa te paari ; He created the land of Hawaii ; Fanau fenua Hawaii; Hawaii great and sacred, Hawaii nui raa, As a crust (or shell) for Taaroa. Ei paa no Taaroa. The earth is dancing (moving). Te ori-ori ra fenua. O foundations, rocks, E te hemic, e te papa, O sands ! here, here. E te one ! 0, o. Brought hither, press together the earth ; O-toina mai, pohia tei fenua/ Press, press again ! Pohia, popohia l They do not. . . . Aita ia (e far ire) Stretch out the seven heavens ; let ignorance cease. Toro o kitu te rai ; e pan maua. Create the heavens, let darkness cease. Fanau ai te rai, pau mouri, 1 Nahora, if not a misprint, pro- canoe ; " to the Samoan Fola and bably refers to the Tahitian Hora- Hawaiian Hola, "to spread out, to hora, " a platform, the deck of a unfold, to open." THE POLYNESIAN RACE. 223 Let anxiety cease within ; . . . Mataroa e pau roto ; [pau ahai te pautia). Let repose (immobility) cease ; E pau noho ; Let the period of messengers cease 3 E pau va arere ; It is the time of the speaker. E te va orero-reo. Fill up (complete) the foundations, E faai te tumu, Fill up the rocks, E faai te papa, Fill up the sands. E faai one. The heavens are enclosing (surrounding), Fa-opia rai, And hung up are the heavens A toto te rai In the depths ; la hohonu ; Finished be the world of Hawaii. E pau fenua no Hawaii. Mr. Horatio Hale, " United States Exploration Expedi- tion," under Commodore Wilkes, section, "Ethnography and Philology," Philadelphia, 1846, p. 125, refers to this same Tahitian chant as published by Mr Moerenhout, and sees in it another evidence of the Tahitian descent from the Samoan island of Sawaii ; the more so, as in another poem, connected by Mr. Moerenhout with the former, it is said that the god Boo created Uporo, another island of the Samoan group. Mr. Hale is doubtless correct in trac- ing the Tahitians to the Samoan group, though possibly some of them came direct from the Fiji group at the time of their expulsion ; but the evident relation of this Tahi- tian chant to those of the Marquesas group, which posi- 224 THE POLYNESIAN RACE. tively locate the " Hawaii " of which they speak far to the westward of the Fiji group, prevents me from concurring with Mr. Hale in assigning no higher or older origin to this and those chants. I regret that I have not any Samoan legends or chants comparable in date to those of the other groups. If any such exist, I shall be much deceived if they also do not refer to a Hawaii far beyond, and to the westward of their own Sawaii. What Mr. Hale calls " the third portion " of this chant, as arranged and published by Mr. Moerenhout, and treat- ing of the genesis of the Tahitian gods, is evidently a separate poem, and of very much later date ; in short, a local theogony, not even fully recognised on the Society group, and unknown in the neighbouring groups. ( 22 5 ) "No. TV., page 90. Te Tai Toko (The Deluge). Part I. Tlie Lord Ocean is a going 1 Te Fatu Moana ua hoe 'ia To pass over the whole dry land. E taha ta te Moo oa A respite is granted He koina e vae ana For seven days. Na mou atea eitu Who would have thought to bury the great earth Oai tuto e tomi 'ia tepapanui Tinaku. In a roaring flood 1 E. Ma he tai-toko e hetu. E. Ho, ho, in the enclosure ! Ho, ho, i tepapua Ho ! the twisted ropes ! Ho, ho, te tau haulrii Here is confusion among Eta e tohu 'ia i vavena The generations (different kinds) of animals 10 Te tai te puaa O we are the kind, we are the kind, maua he tai, maua he tai O we are reserved from the flood maua a Ice iho e tai Reserved on the flood, E ke iho i tai The flood, the roaring. E. He tai-toko e hetu. E. vol. i. r THE POLYNESIAN RACE. And it will fall over the valleys, A e vi una i na kavai Pass over the plains, Taha una te tohua It will bury the mountains, Tomi 'ia te tau mouna And envelop the hill-sides, A e tupo te van O the flood, the roaring. E. te tai-toko e hetu. E. Ho ! in the enclosure. 20 Ho, ho, i te papua Ho ! the twisted ropes, Ho, ho, te tau hauhii For to tie up in couples Mea pitiki i tahuna. The (various) kinds of animals Te tai te puaa. The white kinds, Te tai te mono The striped kinds, Te tai o te hahei The spotted kinds, Te tai o te patipati The black kinds, Te tai o te papanu The horned kinds, Te tai te Mvikivi The big lizard kinds, Te tai te huho-oa. The small lizard kinds, 30 Te tai te huho-poto the flood, the roarinsr. E. he tai toko e hetu. E. High above the ocean, Tie tie te moana Build a house upon it, Halca haka he hae ma eia APPENDIX. 227 A storied house, the house. He Jute papa, te hae A house with chambers, the house. He hae puJio, te hae A house with windows, the house. He hae puta maama, te hae, A very large house, the house. He hae oa-oa, te hae A house to keep alive He hae mea haapohoe The (various) kinds of animals. Te tai te puaa. the flood, the roaring. E. 40 he tai toko e hetu. E. Ho, ho, there in the enclosure. Ho, ho, ina i te papua Ho, ho, the long-twisted ropes Ho, ho, te hauhii oa To tie up and make fast in couples Mea nati a haamau i tahuna The (various) kinds of animals. Te tai te puaa One man before, Fetu-amo-amo. He enata imua Fetu-amo-amo. One man behind, Ia-fetu-tini. He enata i mui la-feiu-tini The animals between, making great noise. Te puaa te vavena e tani huina O the flood, the roaring. E. te tai toko c hetu. E. Eh ; bear away (carry away) ; Here. E amo E. Eia Carry away the animals ; Here. 50 E amo te puaa. Eia Carry them away to the sea. Here E amo atu atou i tai. Eia 0, the long deep wood (a name for the house or vessel). Kakaveie-oa. Eia [Here. 228 THE POLYNESIAN RACE. the God of destruction (causing evil). Here te Etua o te hakanau. Eia Hina-touti-ani. Here Hina-touti-ani. Eia Hina-te-ao-ihi. Here Hina-te-ao-ihi. Eia O Hina-te-upu-motu. Here O Hina-te-upu-motu. Eia O Hina-te-ao-meha. Here Hina-te-ao-meha. Eia O Fetu-moana. Here. te Fetu-moana. Eia Fetu-tau-ani. Here te Fetu-tau-ani. Eia Fetu-amo-amo. Here 6o Fetu-amo-amo. Eia O Ia-fetu-tini. Here. Ia-fetu-tini. Eia O the flood, the roaring. E. he tai toko e hetu. E. A man before, with the offerings, He enata i mua i te utunu O Fetu-moana. te Fetu-moana A man behind, clinging to the offerings. He enata i mui te pikia i te utunu Fetu-tau-ani. te Fetu-tati-ani A turtle between, making great noise He hono te vavena-e-tani-huina O the flood, the roaring. E. te tai toko e hetu E. Cut off, cut off your ear ; this is a bad house Tipia, tipia to oe puaina, te hae pe Ha For to cook food for the God * * * * 70 Mea tuna kai no te Etua ke huha ko huha The four-faced priests * * * * Te tau taua mata fa ke huha ko huha APPENDIX. 229 House fast asleep. God the destroyer. Hae momoe, Etua te hakanau Crush, crackle, a stinking crowd. A omi hu, tax piau Bring together, pell-mell, E hau 'ia kohua All the heaven-fed animals. Ani otoa tafau puaa Sleeps the sacred supporter in this noise. Moe te tapu tutui i teia rnu Noise, God, noise, with God arise ! Mu Etua mu, ma Etua va God wills it. Etua Jcaki hia. Here is manifest the trouble (storm), Eia ua atea te toua A trouble that is great and manifest, 80 He toua te mea nui i atea And it is roaring, and it is working, A ua hetu e hana nei A rain like a solid cloud. He ua mea ata talii Bring together, pell-mell, E hau ia kohua. All the heaven-fed animals. Ani otoa tafau puaa Sleeps the sacred supporter. Moe te tapu tutui Shaken up and mixed up is the earth. Ua upu a uu-uu te fenua I consent and let loose N'au e ae tuku atu * * * * a confused noise Te malu he mu * * * * make a buzzing noise, Matu a mu a mu * * * * arise, arise go Matua a va a va * * * I will it thus. Matu t'au kaki tenei. THE POLYNESIAN RACE Part II. O, the * * * new E te kou hou O, the mountain ridges E te vau va-a Tf * * * * E te mota Some * * men He mou uu Enata Are arriving here, Tu-tu ana nei People in the storm (war, trouble) ; Tai i te toua A veil on the head He pae i te olio A paddle in the hand He hoe i te iima E, arrivals, come and push back ioo E tutu ina amai e hoe The ocean to the centre. Te moana ie vene. E, the house, E. E puho E. Here I am aground. Eia toko ae au The Fetu-moana E. Te Fetu-moana E. Hearken up there Hakaono oe una nei The Lord-Ocean consents Te Fatu-moana e ao-'ia That the dry land appears. Tefenua moo e haaitea. The Lord-Ocean, E. Te Fatu-moana, E. Ah, quick the * * * new A-ve te ko}i hou APPENDIX. 231 The * * * new, here it is 1 10 He hou hou e ia mai * * * * A te mota. In channels receding. / kava mild Ha. The Lord-Ocean. E. Te Fatu-moana. E. Ah ! quick the * new A-ve te kou hou * * long, and when I * * Kou oa a no au e mota * I will offer seven sacred offerings E utunu ait eitu tapu taetae And seven sucklings that shall cry A eitu mamau a te ve To the Lord-Ocean. No te Fatu-moana The Lord has assented that the earth Te Fatu ua ao te fenua Shall now be dry. 120 E moo ana mai. E, the traveller, E te teetina The traveller of Tanaoa, Te teetina Tanaoa Over the sea of Havaii, Una te tai o Havaii Stretch thy bones thither, Te ivi a ke atu Stretch thy bones hither, Te ivi a ke mai Over the sea of Havaii, Una te tai Havaii. Tanaoa, rest on the curling wave, E noho Tanaoa no te hae-hae Remain at the stern of the vessel, E maohe i te mui te vaa 232 THE POLYNESIAN RACE. Strike, strike your legs, Tanaoa E paki-pakia to vae Tanaoa Tanaoa, I will it thus. 130 Tanaoa au Tcaki hia. Tanaoa, why do you return 1 Tanaoa heaha to oe hua Returned is the North wind with the A hua te tiu me te hafa Not found is a place where to alight. Aoe koaa e tau ae mei nei at a Tanaoa, I will it thus. Tanaoa au kaki hia Alight, Tanaoa, on the sands. E tau Tanaoa i te one-one Call Tanaoa here E vevau Tanaoa nei tahu mai Do not go away. A umoi a hee atu Strike, strike thy breast, Tanaoa. E paki-pakia te vaa Tanaoa Tanaoa, yes I will it thus. Tanaoa ee au kaki hia E the traveller, 140 E te teetina The traveller of Moepo, Te teetina Moepo Over the sea of Havaii Una te tai Havaii Thy bones stretch thither, To ivi a ke atu Thy bones stretch hither, To ivi a ke mai Over the sea of Havaii. Una te tai Havaii Ah, alight, alight here. E a a tau-tau mai. E the Lord Ocean. E. E te Fatu-moana. E, APPENDIX. 233 The four bowls, and the four bowls Te efa ipu-ipu, a te efa ipu-ipu Are safely landed here. Ua tau meitai nei. Great mountain ridges, ridges of Havaii 150 Va-va nui 'ia te va-va Havaii Great mountain ridges, ridges of Matahou, Va-va nui Ha te va-va o Matahou. Whereon to thread and stamp. ilea kihahi a Jcahi. Ah, here is the Moepo A eia te Moepo Bringing aloft what has been gathered. E hai ina mai una kohi-kohi. Part III. Ask, ask, the sorcerer (the high-priest), Uizui te tupua Generations new, generations past, Tai hou, tai hee, Who is the flower above there ] Oai oia te pua una nei ? It is Atii-hau-hua. Atii-hau-hua. The Tiki-vae-tahi. Te Tiki-vae-tahi. What is the work of that God, 160 Healia te hana o tena etua That is here revealed Tefai mai ae With that face that is so bright, Me tena ao te io mai ae And with that noise arising 1 A me tena mu ua va E generations, E. E, tai, E. 234 THE POLYNESIAN RACE. Generations go (spread) again, Tai a-hee-hou E quick over the plain, E a-ve una te tohua Return and stand with Tanaoa Te hua a ua tu me Tanaoa I shall arrive, hearken, A tu-tu au, e ono Hark, hark, arise, get up, Ono, ono, tu ae va-a Ho, Ho, arise, God wills it thus. 170 Ho, Ho, va, Etna JcaJci hia. Ask, ask, the sorcerer, Ui, ui te tupua Who is the flower inland here 1 Oai oia te pua iuta nei ? It is Ka-ka-me-vau. Ka-ka-me-vau. The God with the white teeth. Te Etua niho teea Hark, it is he, I arise, hearken, Ono oia tutu au e ono E generations, E. E, tai, E. Generations go again Tai a hee hou Quickly over the plain. E a-ve una te tohua Hark, hark, arise, get up, Ono, ono, tu ae va-a Ho, Ho, arise, God wills it thus. 1 So Ho, Ho, va Etua hahi hia. Ask, ask, the sorcerer, Ui ui te tupua Who is the flower seaward here ? Oai oia te pua tai nei ? It is the Fatu-moana, Oia te Fatumoana APPENDIX. 235 He is going to sacrifice Na hoe 'ia e tooo The sorcerer here below. Te tupua iao nei the black eel (water-snake) te puhi ke he The eel with ugly head. Te puhi olio ino. Who is the flower tied here 1 Oai oia te pua nalci nei? It is Tu-mata-te-vai. Oia Tu-mata-te-vai. Who is the flower before here 1 190 Oai oia te pua mua nei? It is Au-te-una-tapu. Oai te o Au-te-una-tapu, Who is the flower behind here ? Oai te pua imui nei ? It is Mau-te-anua-nua. Mau-te-anua-nua. Who is the strange flower here ? Oai te pua hiva nei ? 1 am here, Tumu-tupu-fenua. 195 au tenei te Tumu-tupu-fenua. Believing that Mr. Lawson — from whose MS. collection the foregoing chant has been copied — in his endeavour to be literal in his translation, has sometimes become un- intelligible, I have attempted a translation that would in some measure obviate that defect, but this, like almost all the ancient Polynesian chants, is replete with tropes and allusions of which the original meaning is in many instances now forgotten or only acquired with great diffi- culty. The words marked with asterisks in the translation are such words as I was either unable to find in the only Marquesan vocabulary within my reach (that by the Abbe Mosblech. Paris 1843), or only found with a meaning that would have made no sense of the context. (236) No. V., see page, 1 16, n. 2. Dieffenbach, in "Travels in New Zealand," p. 28, &c, describes the baptism of infants ; that the priest with a green branch, dipped in a calabash of water, sprinkled the child and recited a prayer over it. ' The prayer differed for boys and for girls. The following are the prayers : — For Boys. Tohia te tama nei Kia riri, kia nguha Kani o tu me te nganahau Ka riri ke tai no Tu Ka nguha ki tai no Tu Koropana ki tai no Tu E pa te karanga ki tai no Tu Me te nganahau ki tai no Tu Taku tama nei kia tohia Koropana ki tai no Tu Pa mai te karanga ki tai no Tu Ko te kawa o karaka wati O riri ai koe, e nguha ai koe E ngana ai koe E toa ai koe E kano ai koe Ko Tu iho uhia Ko Kongo iho uhia. J o - For Girls. Tohia te tama nei Kia riri, kia nguha te tama neL APPENDIX. 237 Kani o Tu me te nganakau Ka riri ki tai no Tu Ka wakataka te watu Kania ma taratara Te hau o Uenuku Pulia ka mama tauira o Tu Ka mama tauira Kongo Ho. Ka kai Tu Ka kai Kongo Ka kai te Wakariki He haba He hau ora He kau rangatira Kei runga, kei te rangi Ka puha te rangi E iri iria koe ki te iri iri Hahau kai mau tangaengae Haere ki te vvahie mau tangaenfrae Watu kakahu mou taii£raeno;ae. I am inclined to think that Dieffenbach has not correctly apprehended some of the words in the above prayers, or that they have been misprinted. ( 2j8; No. VI., page 128. The following are a number of signs and omens current among the Hawaiians in heathen times, and not yet en- tirely disregarded : — Opeakua : "Hands crossed behind." If a Hawaiian was going out on business or on pleasure, and met another person with his hands crossed behind his back, it was an unlucky omen ; but if it occurred twice on the same journey, it became a sign of success. Maka-paa: "A blind person." If you met a blind per- son on the road, it was a bad sign. If you met two blind ones, the sign was good. Kahea-kua-ia : " Calling after one." If, starting on a journey, you were called after or called back by some- body, it was a bad sign. Therefore, to prevent being thus troubled, the traveller always told those whom he left where he was going, his errand, &c. Kuapuu-hohailua : " Meeting a humpback." If on your journey you met a humpbacked person, it was a bad sign. If two such were met with, the sign was good. Hoo-kua-kii: "Arms akimbo." If you met a person with his or her arms akimbo, hands resting on the hips, it was a bad sign. Hoihou-i-hopie : " Eeturning." If, starting away from a place, and having actually proceeded some distance from the house, however short, you turned back after something forgotten or left, it was a bad sign. Ku-ia ka wawai : " Stumbling." If you stumble or stub your. feet in walking, it is a bad sign. APPENDIX. 239 Makole : " Sore-eyed." If you meet a person with sore or inflamed eyes, the sign is bad. Kukue : " Lame." If you meet a lame or deformed per- son, it is a bad sign. Maia : " Bananas." If you are going on business, and meet a man carrying a banana bunch, you will not pros- per, and would do well to defer it; but if that cannot be done, then, to avoid the evil omen, you should either touch the bananas with your hand, or grasp them, and then proceed on your journey without looking back. Alae : " A water-fowl." If the bird called alae was heard crying in the neighbourhood of a village, it was a sign of the death of somebody there. Kuukuu : " Spider." If the long-legged spider drops dow T n from above in front of you, or on your bosom, it is a good sign, foreboding either presents or strangers ; if he drops on either side, or behind you, the sign brings you no good. Hulahula ka Mdka : " Twitching of the eyes." If the eye twitches or throbs suddenly, it is the sign of the arrival of strangers, or of approaching wailing for some one that is dead. Kani-ana ka ula ka pepeiao : " Einging sounds in the ear." If you have a ringing sound in your ears, it is a sign that you are spoken evil of by some one. If in the right ear, by a man ; if in the left ear, by a woman. Sometimes it indicates approaching sickness. Okakala or Malana ka poo : " Shuddering, shivering of the head." If you feel a sudden shivering or itching of the scalp or skin of the head, as if a louse were crawling, it is a sign that you are spoken evil of. Koni na wawae: "Throbbing of the feet." If you feel a beating, creeping, throbbing sensation in the foot, it is a sign that either you will go on a sudden journey, or that strangers are arriving. ( 240 ) No. VII. Vide page 132. There are numerous other customs, traits, and peculiarities observed by the Polynesians, which find remarkable ana- logies and coincidences among the nations to the west of them, from whom they sprang, or with whom they coha- bited during unknown periods of their former national life. Each one, singly, is but a drop in the stream of evidence which tends to connect the Polynesian with the Cushite and Arian races ; but, taken together, they supplement in a large measure the coincidences previously referred to, and strengthen the evidence of that connection beyond the possibility of contravention. As with my limited means of reference I am unable to say whether these coincidences ultimately refer to Arian or Cushite sources, seeing that the former borrowed so much from the latter, I merely present them en Hoc, that men more able than myself may classify them hereafter. We find, then — 1. The Hawaiian soothsayer or kilo-kilo turned always to the north when observing the heavens for signs or omens, or when regarding the flight of birds for similar purposes. The ancient Hindus turned also to the north for divining purposes, and so did the Iranians before the schism, after which they placed the devas in the north; so did the Greeks, and so did the ancient Scandinavians before their conversion to Christianity. II. Hogs were of the most precious offerings to Polynesian APPENDIX. 241 gods, and hogs' -meat the most delicious food of the people. While the Egyptian, the Hebrew, and the Brahmanised Hindu abominated swine, and the contact therewith, the Arian, Goth, and Scandinavian sacrificed swine as well as sheep, cattle, and horses to their gods, and the boar was a daily feast to the heroes of that northern Valhalla. The Greeks held swine in high estimation, and Homer gives to a swineherd the title " divine." in. The Egyptians were permitted to marry their sisters by the same father and mother, and in patriarchal times among the Hebrews a man might marry a sister, the daughter of his father only, though it was afterwards for- bidden in Leviticus (chap, xviii.). According to Persian law such marriages were not permitted ; but the Greeks and Eomans seem to have admitted the practice in earlier times, if the proceedings of the Olympian gods are an index of primitive manners — Saturn and Rhea, Jupiter and Juno. Among Hawaiian chiefs such marriages were not un- common, even in earliest times, and the offspring of such unions were invested with higher rank, and called Aliipio, taking precedence over brothers and sisters of different unions. IV. In the Hawaiian, Marquesas, and Tahitian groups the first prisoner taken in war was invariably offered as a sacrifice to the particular god of the captors. " The custom of sacrificing their first prisoner (in war) is ascribed by Procopius to the Thulitce or Scandinavians " (" Bell. Goth." ii. 15). " The Germans made their first cap- tive contend with a champion of their own race, and took the result as an omen of success or failure " (" Tacit. Germ." 10), vide Rawlinson's "Herodotus," vii., 180, n. 4. vol. 1. Q 242 THE POL YNESIA N EA CE. In Ceylon and Southern India, whenever a favour is solicited, peace made, or an interview desired, presents are always sent before. 1 In Hawaii and elsewhere in Polynesia presents always accompanied the visitor, or were sent before. VI. In India an unhealthy country is said to " eat up the inhabitants," and a victorious or oppressive rajah is said to " eat up the country." 2 In Hawaii the expression Ai-moha, "eating up the land," is an epithet of chiefs. VII. In India the expression " to live in the shadow," i.e., under the security and defence of another, is very common. 3 In Hawaiian the expression e noho ma Tea main, " to dwell in the shadow," i.e., under the protection of such or such a chief, is frequently heard. VIII. In App. chap. v. book ii. of Eawlinson's edition of " Herodotus," I read that the hieroglyphic sign for a nega- tive " is a pair of extended arms with the palms of the hands downwards, preceding the verb." Before such an action of the hands could have become a recognised hiero- glyphic sign of a negative, it must have been a common and generally adopted manner of expressing a negative in actual everyday life, a gesticulation as significant and as well understood to the Egyptians, and perhaps the entire Cushite race, as a shrug of the shoulders or a shake of the head is to many nations of modern time. 1 Oriental Illustrations, by Joseph Eoberts, p. 22. London, 1835. s Ibid., p. 10 1. 8 Ibid. APPENDIX. 243 I know not if such a manner of expressing a negative still obtains among the Cushite descendants in N.E. Africa, in Asia, or the Archipelago, but the self-same identical manner of inverting the hands, "palms down- wards," in sig%of a negative answer, prevails throughout Polynesia. Ask a person if he has such or such a thing, and, two to one, instead of saying " No," he will turn his hand or hands " palms downwards," in sign of a negative answer. (244 ) No. VIII., 'page, 144. Mr. Crawfurd, in his " Grammar and Dictionary of the Malay Language," vol. i. p. 134-35, considers that those who hold that the Polynesian " language and race are essentially the same as the Malay," are undoubtedly " under a great mistake," and advance " a gratuitous assumption." And, though he resolutely repudiates the idea that there is anything, physically or linguistically, to connect them as springing from the same race, or that the former descended from the latter — yet, in order to account for the few Malay and Javanese words which, according to him, have found entrance into the Polynesian language, he resorts to the hypothesis that at some remote period, while the Polynesians were still living in a body, before their dispersion over the East Pacific, they had been visited by a fleet of Malay rovers, who introduced to the then uncultivated Polynesians the knowledge of the taro, yam, cocoa-nut-palm, sugar-cane, and the numeral system, p. 144 ct seq. ; and he fixes upon the Tonga or Friendly Islands as the country where this encounter took place. Mr. Crawfurd argues that, as those articles, taro, yam, &c, bear Malay names, ergo they are of Malay origin, and most probably brought by Malays to the Polynesians. Let us consider these names. Mr. Crawfurd identifies the Polynesian taro, halo, arum esculentum, with the Javanese talds, and the Polynesian to, ho, sugar-cane, with the Javanese and Malay tabu. I would not, on slight grounds, question the conclusions arrived at by a gentleman who APPENDIX. 245 has done so much, and done it so worthily and well, to illumine the dark and unknown parts of the Asiatic island-world ; but when Mr. Crawfurd wrote, the data bearing upon Polynesian life, language, customs, and tradi- tions were scanty, detached, frequently one-sided, and hence not always reliable. Consequently, when Mr, Crawfurd deriving talo from talds, and to from tabu, I have no doubt that every competent Polynesian scholar of the present day, foreign or native, would dissent from such derivation as contrary to the very genius and idiom of the Polynesian. Had talds and tabu at any time been introduced as foreign words in the Polynesian language, the form the former would have assumed could not pos- sibly have been any other than talasa, talaha, talafa, or even talaka, according to the peculiar dialect wherein adopted ; and the form of the latter would have been tapu, tafu, tahu, or tawu. With one exception, I know of no single instance where a foreign word, introduced in the Polyne- sian, and ending with a consonant, is not invariably fol- lowed by a vowel to enable the Polynesian to pronounce it ; and I know of no instance where a foreign dissyllable, ending with a vowel, has been contracted into a mono- syllable in the Polynesian. The exception alluded to is in the Hawaiian dialect, where a few foreign words end- ing in er or ar have the r elided and the entire syllable sounded as a, as dia for " deer," bea for " bear," wineka for " vinegar," leta for " letter," and some others ; though the rule is not general, for we find per contra foreign words like hamcde for " hammer," lepcla for " leper," and others. And so violent a contraction as to from tabu is entirely unheard of in the Polynesian. Faro and tdlas, to and tabic may possibly be related; but if so, it is for the very opposite reason, viz., that the latter names are derived from the former, and not as Mr. Crawfurd claims it. Taro is not a staple food of the Malays or Javanese, who, when they arrived from India, brought with them their rice-eat- ing proclivities, and spread the use of the article as well as 246 THE POLYNESIAN RACE. the name, throughout the Archipelago. But taro has from time immemorial been the staple food of the Polynesians ; and so with the sugar-cane. The Hawaiians ascribe the introduction of taro to their renowned ancestor Wakea ; but, according to the most reliable and rational of their traditions, it will be seen that Wctkca was a Gilolo chief, in times previous to the Polynesian migrations, who never put foot on any of the Pacific groups now inhabited by those who claim descent from him. That the Polynesian ufi, uhi, and u'i, "yam," and niu, " cocoa-nut," are identically the same words as the Malay and Javanese ubi, uwi, and the Javanese nu, there can be no doubt. But assuming that yams and cocoa-nuts were not indigenous on the groups of the Pacific — which, however, has yet to be proven — is it not as likely that the first Polynesian emigrants from the Archipelago (Asiatic) brought those articles with them, as that they were subsequently brought to them by Malay rovers ? Besides, I am inclined to think that Tcdlapa is the genuine proper Malay name for cocoa-nut, and nu only adopted by them since their arrival in the Archi- pelago, and adopted from the previous inhabitants, the Polynesians and their congeners. Among the thirty-three names for cocoa-nut recorded in the Appendix to Mr. E. A. Wallace's "Malay Archipelago," p. 611, there are thirty-one entirely different from the Malay name, which is there given as 'kalapa or kldpa, and twenty-three which are evidently related to the Polynesian niu, though more or less corrupted, of which only two, the Salibabo and Ceram (Gah), appear to have retained the Polynesian word in its purity. In regard to the numeral system, I have shown (p. 144 d seq.) the sources which in all probability contributed to form the Polynesian numerals, and for which they are not beholden to either Malays or Javanese. The notion entertained by Mr. Crawfurd, that the Tonga Islands were the cradle of the Polynesian race, APPENDIX. 247 from whence they spread over the Pacific Ocean, after having received the benefits of the intercourse with Malay and Javanese sea-rovers, does not at this day require con- sideration. I prefer to follow Mr. Horatio Hale in his excellent work on the " Ethnology and Philology of Poly- nesia — United States' Exploring Expedition," published 1 846, wherein it is convincingly shown that the primary rendezvous of the Polynesian emigrants from the Asiatic Archipelago was at the Fiji group, and that when driven out from there they scattered east, south, and north over the Pacific Ocean. Malay and Javanese rovers may have followed them to the Fijis ; but if so, there is no trace of such occurrence in the traditions, customs, or language of the Polynesians. "Whatever there may be in common between the Malays and Javanese, on one hand, and the Polynesians, on the other, must be sought for in circum- stances unconnected with ethnic consanguinity, and exist- ing previous to the migrations of the latter into the Pacific. Append 15 Nanarnea. Puia, w. I 16 Peliekeula. Ulnae, w. I 17 Pehekemana. Nanahapa, w. I 18 Nan am u a. Nanabope, w. I 19 Nanaikeauhaku. Elehu, w. I 20 Keaoa. "Waohala, w. I 21 Hekumu. Kumukoa, w. I 22 Umalei. Umaumanana, ■>• I 23 Kalai. Laikap- 24 Mai Pil- 14. NANAULU, /r. ULUKOU, w. >5 T teCbLD-URl c 02 7 9S 7 •Wf 2 DEC 1 6 Form — _ 7 - AT LOg ANGILW tf 3 1158' 00271 51 OS UC SOUTHERN REGIONAL LIBRARY FACILITY A A 000 159 293