i! ) ! • ! ' ! li' iili ! m m f!lH 1! m i m mm i ) (! UBRARY OF CONGRESS DDDnfl31fl4A hf h.J )! MENTAL AND SOCIAL CONDITION OF SAVAGES PRINTED BY SPOTTISWOODE AND CO., NEW-STREET «QUArd5 LONDON }J;'l lb^r:j\f%:lliih'M^ THE ORIGIN OF CIVILISATION AND THE PRIMITIVE CONDITION OF MAN MENTAL and SOCIAL CONDITION of SAVAGES BY SIE JOHN LUBBOCK, ^BAET. MP., F.E.S., D.C.L., LL.D. AUTHOR OF 'PREHISTORIC TIMES' ETC. : HONORARY SECRETARY OF THE LONDON BANKERS PELXOW OP THE SOC. OP ANTIQUARIES ; OF THE GEOLOGICAL, ENTOMOLOGICAL, AND OTHER SOCIETIES : PRESIDENT OP THE LONDON CHAMBER OP COMMERCE AND VICE-CHAIRMAN OF THE LONDON COUNTY COUNCIL Jift^ €bilion, faitfe itwmerons gi,b!bitions NEW YORK D. APPLE TON AND COxMPANY 1898 ^ ^^' ^;c(,<^% 34%!- PEE FACE 1^ my work on ' Prehistoric Times ' I have devoted several chapters to the description of modern savages, because the weapons and implements now Lised by the lower races of men throw much light on the sig- nification and use of those discovered in ancient tumuli, or in the drift gravels ; and because a knowledge of modern savages and their modes of life enables us more accurately to picture, and more vividly to conceive the manners and customs of our ancestors in bygone ages. In the present volume, which is founded on a course of lectures delivered at the Eoyal Institution in the spring of 1868, I projDOse more particularly to describe the social and mental condition of savages, their art, their systems of marriage and of relationship, their re- ligions, language, moral character, and laws. Subse- quently I shall hope to publish those portions of my lectures which have reference to their houses, dress, boats, arms, implements, &c. From the very nnture of the subjects dealt with in the present volume, I shall have to record many actions and ideas very abhorrent to vill PESFACE us ; so many, in fact, that if I pass them without com- ment or condemnation, it is because I am reluctant to fatigue the reader by a wearisome iteration of disap- provaL In the chajDters on Marriage and Keligion more especially, though I have endeavoured to avoid everything that was needlessly offensive, still it was impossible not to mention some facts which are very repugnant to our feelings. Yet were I to express my sentiments in some cases, silence in others might be held to imply indifference, if not approval. Montesquieu ^ commences with an apology that por- tion of his great work which is devoted to Religion. As, he says, ' on pent juger parmi les tenebres celles qui sont les moins epaisses, et parmi les abimes ceux qui sont les moins profonds, ainsi Ton pent chercher entre les religions fausses celles qui sont les plus con- formes au bien de la societe ; celles qui, quoiqu'elles n'aient pas I'effet de mener les hommes aux felicites de I'autre vie, pen vent le plus contribuer a leur bonheur dans celle-ci. Je n'examinerai done les diverses reli- gions du monde que par rapport au bien que Ton en tire dans I'etat civil, soit que je parle de celle qui a sa racine dans le ciel, ou bien de celles qui ont la leur sur la terre.' The difficulty which I have felt has taken a different form, but I deem it necessary to say these few words of explanation, lest I should be supposed to approve that which I do not expressly condemn. ^ Esprit des Lois, liv. xxiy. ch. 1. FEE FACE ix Klemm, in his ' AUgemeine Culturgeschichte der Menscben,' and recently Mr. Wood, in a more popular manner ('Natural History of Man'), have described the various races of man consecutively ; a system which has its advantages, but which does not well bring out the general stages of progress in civilisation. Various other works, amongst which I must specially mention Mliller's ' Geschichte der Americani- schen Urreligionen,' M'Lennan's ' Primitive Marriage,' and Bachofen's ' Das Mutterrecht,' deal w^ith particular portions of the subject. Maine's interesting work on ' Ancient Law,' again, considers man in a more ad- vanced stage than that which is the special subject of my work. * The plan pursued by Tylor in his remarkable work on the ' Early History of Mankind ' more nearly re- sembles that which I have sketched out for myself, but the subject is one which no two minds would view in the same manner, and is so vast that I am sure my friend will not regard me as intruding upon a field which he has done so much to make his own. Nor must I omit to mention Lord Kames' ' His- tory of Man,' and Montesquieu's ' Esprit des Lois,' both of them works of great interest, although written at a time when our knowledge of savage races was even more imperfect than it is now. Yet the materials for such a work as the present are immense, and are daily increasing. Those who take X PEE FACE an interest in the subject become every year more and more numerous ; and while none of my readers can be more sensible of my deficiencies than I am myself, yet, after ten years of study, I have been anxious to pub- lish this portion of my work, in the hope that it may contribute something towards the progress of a science which is in itself of the deepest interest, and which has a pecidiar importance to an empire such as ours, com- prising races in every stage of civilisation yet attained by man. High Elms, Down, Keitt: February, 1870. CONTENTS CHAPTER I. INTRODUCTOEY CHAPTER. PAGE Importance of tlie Subject — Difficulty of the Subject — Inactivity of the Savage Intellect — Condition of the Lowest Races of Men — Curious Customs with reference to Mothers-in-Law — La Couvade — Reasons for La Couvade — Savage Ideas on the Influence of Food — Curious Ideas with reference to Portraits — Use of Prayers as Medicine — Savage Ideas of Disease — Medical Treatment among Savages — Fancies about Twins — Life attributed to Inanimate Ob- jects — Salutations 1 CHAPTER II. ART AND ORNAMENTS. Art as an Ethnological Character — Ancient Art — Art in Africa — Esqui- maux Drawings — The Quippu — Picture-writing — Indian Census Roll — Indian Tombstones — Picture-writing in North America — Indian Biography — Indian Petition — Rock Sculptures — Savage Ornaments — Cheek Studs — Labrets — Ornamentation of the Skin — Tribe Marks — Tattooing — Artificial Alteration of Form — Hairdress- ing — Fiji Head-dresses 38 CHAPTER III. MARRIAGE AND RELATIONSHIP. The Position of Women among Savages — Absence of Affection in Marriage — Absence of Marriage — Relationship among Savages — Different Kinds of Marriage — Polyandry — Separation of Husband and Wife — Absence of Marriage Ceremony — Marriage Ceremonies xii CONTENTS PAGE — Relationships Independent of Marriage — Soutli Sea System of Relationship — Toda System of Relationship — Pre valence of Adoption — The MiUc-tie — Original or Communal Marriage — Origin of Mar- riage — Bachofen's Views — Wrestling for Wives — M'Lennan's Views — The True Explanation — Origin of Marriage by Capture— Preva- lence of Marriage by Capture — Originally a Reality — Subsequently a Form — Hindustan — Central India — Malay Peninsula — Kalmucks — Tonguses— Kamchadales — Mongols — Koreans — Esquimaux — North and South Americans — Fijians — Polynesians — Philippine Islanders — Negritos— Africa — Circassians — Europe — Rome — Poland — Russia — Britain — Explanation of Marriage Ceremonies — Marriage by Confarreatio — Expiation for Marriage — Babylonia — Armenia — Balearic Islands— Temporary Wives — Exogamy— Origin of Ex- ogamy — Prevalence of Exogamy — Australia — Africa — Hindostan — Northern Asia — China — Circassia — North America — South America — The Causes of Polygamy — Polyandry — Polyandry Exceptional — The System of Levirate — Endogamy — The Milk-tie — Relationship through Females — Causes and Wide Distribution of the Custom — Neglect of Paternal Relation — Origin of Relationship in the Male Line — Change from Female to Male Kinship — System of Kinship through Males — The Present System . . . .69 CHAPTER IV. THE ORIGIN OF MARRIAGE. Communal Marriage — Australian Marriage Law — Two Forms of Mar- riage in Australia — Origin of Individual Marriage — Marriage by Capture — Australia — India — Northern Asia — America — Esqui- maux — Indians — South America — Pacific Islands — Africa — Arabia — Europe — Symbol of Capture — Temporary Marriages — Exogamy — Origin of Exogamy — Origin of Marriage — Polyandry — TheLevi- rate — Endogamy — Relationship through Females — Relationship throush Males CHAPTER V. ON THE DEVELOPMENT OF RELATIONSHIPS. On the Development of Relationships — Different Systems of Relation- ships—Classification of Systems — Nature of the Evidence — Custom of addressing Persons by their Relationship — Similarities of System among the Lower Races — Malayans — Fijians — Redskins — Nomen- clature of Relationships — ESect of Female Kinship on S3-stems of Relationship — The Hawaiian System — American Systems — Import- CONTENTS ance of the Mother's Brother in the Family System — The Micmac System — Burmese and Japanese Systems — The Wj^andot System — The Tamil and Fijian Systems — Remarkable Terms in Use — Explanation of the Terms — System of the Oneidas — Otawas — The Kaffir System — Mohegans — Crees — Chippewas — Summary of Red- skin Systems — Hindoo Sj^stems — Karens — Esquimaux — Remarkable Similarities — Indications of Progress — Incompleteness of Systems — Existing Systems Incompatible with the Theory of Degradation — Evidence of Progress — No Evidence of Deoradation — Conclusion . 162 CHAPTER VI. RELIGION. Mental Inactivity of Savages — Religious Characteristics of Savages — Religious Characteristics of the Lower Races of Man — Classification of the Lower Religions — Religions according to Sanchoniatho — Religious Condition of the Lowest Races — Absence of Religion — Rudimentary Religions — Religious Ideas as suggested by Sleep — Religious Ideas as suggested by Dreams — Nightmare — Shadows — Thunder — Spirits regarded as Evil — Spirits regarded as causing Disease — Madness reverenced — Belief in Witchcraft — Disbelief among Savages in the Existence of Natural Death — Low Ideas of Spirits entertained by Savages — Greek and Roman Conceptions — Savage Ideas as to Eclipses — Various Notions as to Eclipses — Belief in Ghosts — Future Life dependent on Mode of Death — Belief in the Plurality of Souls — Divination — Sorcery — Confusion of Name and Thing— Confusion of Part and Whole — Similarity of Witchcraft — Wizards — Belief in Witchcraft shared by European Travellers — Sorcerers not necessarily Impostors — Fasting — Religious Dances — Smoking as a Religious Ceremony — Intoxication as a Religious Rite 205 CHAPTER VII. RELIGION (continued). Animal-Worship — Origin of Animal- Worship — The Kobong — The Totem — Totemism in America — Totems in India and Polynesia— Ser- pent-Worship — Serpent- Worship in Asia — Africa — Guinea — Why- dah — Agoye the Fetich of Whydah — Kaffraria — Madagascar — Polynesia — America — The Worship of other Animals — Polynesia — Sandwich Islands — ^Fiji Islands — Siberia— China— India— Ceylon — The Philippines — Africa— Madagascar — Europe — The Custom of Apologising to Animals for killing them— The Worship of the Celestial Bodies— Savage Tendency to Deification — Deities not sup- xiv CONTENTS posed to be Supernatural — Life attributed to Inanimate Objects — Souls attributed to Inanimate Objects — Tree- Worship in Europe — Egypt — Arabia — Congo — India — Ceylon — Hill Tribes of India — Siberia — Sumatra — Philippines — Fijians— North America — Mexico — Peru — Patagonia — Water-Worship — Europe — Siberia—India — Africa — North America — Central America — The Worship of Stones —Attributes of the God Mercury — Siberia — Hindostan — New Zea- land — The Arabians — Phoenicians in Europe — Africa — Polynesia — Fiji Islands — Micronesia — America — Fire- Worship — Vestals — Asia — America — Africa — Sun and Moon Worship — America — India — Asia — Africa — Sundry Worships ...... 261 CHAPTER VIII. RELIGION (concluded). Religion of Australians — Veddahs — Californians — Bachapins — Kaffirs — Fetichism — Hindostan — Negroes — Fetichism in other Races — North America — China — Siberia — Africa — Totemism — Develop- mental and Adaptational Modifications of Religion — Myths — Sha- manism in Siberia — Greenland — Pacific Islands — Africa — India — Idolatry — Origin of Idolatry — Connection with the Worship of An- cestors — India — Africa — Polynesia — Siberia — Solomon's Explana- tion — Idols not mere Emblems — Worship of Men — Worship of Chiefs — Worship of Trarellers — The Worship of Principles Sacrifices — Confusion of the Victim with the Deity— Worship of the Sacrifice — Eating the Sacrifice — Human Sacrifices — Europe — America — The Jews — Temples — Priests— Mystery Men— The Soul Ideas of Heaven — The Future State — Creation — Prayer — Morality — The Progress of Religion— Science and Religion .... 325 CHAPTER IX. CHARACTER AND MORALS. Difficulty of ascertaining the Character of Savage Races — -Insecurity of Life and Property among Savages— Progress in Morals— Moral Condition of Savages— Confusion of Family A flection and Moral Feeling— Absence of Moral Feeling-^-Religion not necessarily con- nected with Morality— Future Life not necessarily one of Punish- ment or Reward— Rank in Heaven — Law and Right Growth of Moral Feeling — Origin of Moral Feeling 394 CONTENTS XV PAGE CHAPTER X. LANGUAGE. Gesture Language — The Origin of Language — All Language reducible to a Few Root-words — Origin of Root-words — Onomatopoeia — Wear and Tear of Words — Nicknames and Slang Terms — Origin of the Terms Father and Mother — Words for Father and Mother in A^arious Languages — The Choice of Root-words — Poverty of Savage Lan- guages — Deficiency in Terms of Afiection — Absence of Abstract Terms — Deficiency in Numerals — Savage Diffieulties in Arithmetic — Use of the Fingers in Arithmetic, as shown in the Names of Numerals — The Origin of the Decimal System 416 CHAPTER XL LAWS. Importance of the Subject — Savage Laws not founded on the Family — Tyranny of Fashion among Savages — Tyranny of Custom among Savages — Superstitious Customs — Rules relating to Legal Cere- monies and Contracts — Court Language — Gradations of Rank — Salu- tations and Ceremonies — Conduct of Public Business — Property in Land — Communal Property — Laws of Inheritance — Absence of Wills — Roman Wills — ^Rights of Children — The Yasu — Custom of naming Parents after Children — Laws of Inheritance — The Punish- ment of Crime — Regulated Revenge — The Laws of Property — Mani- fest and Non-Manifest Thieves — The WerD-eld — General Conclusion 448 APPENDIX. PART I. Difficulty of obtaining Conclusive Evidence— The Stationary Condi- tion of Savages — No Evidence of Earlier Civilisation — Evidence derivable from Domestic Animals and Pottery— Indications of Pro- gress among Savages— Savages not Incapable of Civilisation- — In- digenous Origin of Mexican Civilisation— Progress as indicated by Language — Traces of Barbarism in Civilised Countries— Arbitrary Customs — Unity of the Human Race— -Mental Differences in the Different Races 487 CONTENTS PART II. The Weapons of Monkeys — True Nature of Barbarism — Sequence of Customs — The Diffusion of Mankind — The Influence of External Conditions — The Esquimaux — Original and Universal Barbarism — Supposed Inevitability of Degradation — Supposed Evidence of De-^ oradation — The Survival of Customs — Progress of Religious Ideas — Fetichism — Totemism — Idolatry — The True Theory of the Four Ages — Evidence from Crossed Races — Similarity existing between Savages and Children — Language of Savages — Tendency to Redu- plications — Ancient Ceremonies and Modern Games — Development of the Individual, and that of the Species ..... 60: NOTES 531 INDEX o41 ILLUSTRATIONS DESCEIPTION OF THE PLATES. PLATE PAGE Frontispiece — View op Stonehenge. From an original draw- ing by M. Griset To face Title I. Sketch op Mammoth, on a piece of ivory, found in the Rock- shelter at La Madeleine, in the Dordogne . . To face 38 IL Fijian Modes of Dressing the Hair. After Williams. ' Fiji and the Fijians,' p. 158 To face %^ III. Indian Sacred Stones. After Forbes Leslie. ' Early Races of Scotland,' voL ii. p. 464 To face 808 IV. A Human Sacrifice in Tahiti. After Cook . . To face 371 V. Group of Sacred Stones in the Dekhan. After Forbes Leslie. * Early Races of Scotland,' vol. ii. p. 460 . To face 375 DESCRIPTION OF THE FIGURRS. FIG. 1. Group of Reindeer. From a photogTaph presented to me by M. le Marquis de Vibraye 39 2-1. Drawings on Esquimaux Bone Drillbows. Presented to the Ashmolean Museum at Oxford by Captain Beechey . 41 5. North American Indian Census Roll. After Schoolcraft. ' History of the Indian Tribes,' vol. ii. p. 222 . . . . 47 6. Indian Gravepost. After Schoolcraft. 'History of the Indian Tribes,' vol. i. p. 356 48 7. Indian Gravepost. After Schoolcraft. 'History of the Indian Tribes,' vol. i. p. 356 48 xviii LIST OF ILLUSTRATIONS I'IG. rAGK 8. Indian Baek Lettee. After Schoolcraft. 'History of the Indian Tribes,' vol. i. p. 338 . 49 9. Indian Bark Letter. After Schoolcraft. ' History of the Indian Tribes,' vol. i. p. 336 51 10. Indian Biography. After Schoolcraft. ' History of the Indian Tribes,' vol. i. p. 336 52 11. Indian Petition. After Schoolcraft. 'History of the Indian Tribes,' vol. i. p. 416 53 12. Caroline Islander. After Freycinet. 'Voyage autour du Monde,' pi. 57 64 13. New Zealand Head. After Freycinet. ' Voyage autour du Monde,' pi. 107 65 14. New Zealand Head. After Freycinet. * Voyage autour du Monde,' pi. 1C7 . . . . • 65 15-17. Shoulder-blades PREPARED FOR Divination. After Klemm. ' All. Cultur. d. Mens,' vol. iii. p. 200 243 18, A Sacred Dance of the Virginians. Lafitau, vol. ii. p. 135. 259 19, Agoye, An Idol op Whiddah, Astley's 'Col. of Voyages,' vol. iii. p. 50 272 20, Sacred Stones, Fiji Islands. Williams, loc. cit. vol. i. p. 220 .314 LIST OF THE PRI.YCIPAL WORKS QUOTED IN THIS VOLUME. Adelung, Mithridates. Allen and Thomson, Exped. to the River Niger. Arago, Narrative of a Voyage round the World. Arbousset and Daumas. Tour at the Cape of Good Hope. Asiatic Researches. Astley, Collection of Voyages, Atkinson, Oriental and Western Siberia, „ Upper and 'Lower Amoor. Azara, Voyages dans I'Amerique Me- ridionale. Bachofen, Das Mutterrecht. Baikie, Exploring Voyage up the Rivers Kwora and Binue. Bain, Mental and Moral Science. Baker, Albert Nyanza. „ Nile Tributaries of Abyssinia. Bancroft, Native Races of Pacific States. Barth, Travels in Central Africa, Battel, The Strange Adventures of (Pinkerton's Voyages and Travels). Beechey, Narrative of a Voyage to the Pacific. Bosnian, Description of Guinea (Pin- kerton's Voyages and Travels), Brett, Indian Tribes of Guiana. Brooke, Lapland. Bruce, Travels in Abyssinia. Burchill, Travels in Southern Africa. Burton, Lake Regions of Africa, „ First Footsteps in Africa, „ Abbeokuta and the Cameron Mountains. Burton, City of the Saints. „ Mission to the King of Da- home. Caillie, Travels to Timbuctoo. Callaway, Religious System of the Amazulu. Campbell, Tales of the West High- lands. „ Wild Tribes of Khoudistan Carver, Travels in North America. Casalis, The Basutos. Catlin, North American Indians, Chapman, Travels in S. Africa, Charlevoix, History of Paraguay. Clarke, Travels, Collins, English Colony in Nev/ South Wales. Cook, Voyage round the World. (In Hawkes worth's- Voyages.) „ Second Voyage towards the South Pole. „ Third Voyage to the Pacific Ocean. Cox, Manual of Mythology. Crantz, History of Greenland. Dalton, Descriptive Ethnology of Bengal. Dalzel, Hist, of Dahomy. Darwin, Animals and Plants under Domestication. ,, Origin of Species. „ Researches in Geology and Natural History. Davis (Dr. J, B,), Thesaurus Cranio- rum, Davis, The Chinese. XX LIST OF WOBKS QUOTED Davy, Account of Ceylon, Deane, Worship of the Serpent traced throughout the World. De Brosses, Du Culte des Dieux fetiches. De Hell, Steppes of the Caspian Sea. Denham, Travels in Africa. Depons, Travels in South America. Dias, Diccionario da Lingua Tupy. Dieffenbach, New Zealand. Dobrizhoffer, History of the Abipones. Drury, Adventures in Madagascar. Dubois, Description of the People of India. Dulaure, Histoire abregee des diffe- rents Cultes. Dann, The Oregon Territory. Dupuis, Journal of a Eesidence in Ashantee. D'Urville, Voyage au Pole sud. Earle, Residence in New Zealand. Egede, Greenland. Ellis, Three Visits to Madagascar. ,, Polynesian Researches. Erman, Travels in Siberia. Erskine, Western Pacific. Eyre, Discoveries in Central Australia. Farrar, Origin of Language. „ Primitive Manners and Cus- toms. Fergusson, Tree and Serpent Worship. Fison and Hewitt, The Kamilaroi and Kurnai. Fitzroy, Voyage of the ' Adventure ' and ' Beagle.' Forbes Leslie, Early Races of Scotland. Forster, Observations made during a Voyage round the World. Forsyth, Highlands of Central India. Franklin, Journey to the Shores of the Polar Sea. Eraser, Travels in Koordistan and Me- sopotamia. „ Tour to the Himalaya Moun- tains. Freycinet, Voyage autour du Monde. Fustel de Coulangen, La Cite Antique. Gaius, Commentaries on Roman Law. Galton, Tropical South Africa. Gama, Descripcion historica y crono- logica de las Peclras de Mexico. Garcilasso de la Vega, Commentaries of the Yneas. Gardner, Faiths of the World. Gibbs (H. H.), Romance of the Chevelere Assigne. Girard de Rielle, La Mythologie Com- paree. Girard-Teulon, La Mere chez certains Peuples de I'Antiquite. Gladstone, Juventus Mundi. Goguet, De I'Origine des Lois, des Arts et des Sciences. Graah, Voyage to Greenland. Gray, Travels in Western Africa. Grey (Sir G.), Polynesian Mythology. „ Journal of Two Expedi- tions of Discovery in North- wesit and Western Australia. Hale, Ethnology of the United States Exploring Expedition. ,, Ethnology and Philology. Hallam, History of England. Hamilton, Account of the Kingdom of Nepaul. Hanway, Travels in Persia. Hayes, Open Polar Sea. Hawkesworth, Voyages of Discovery in the Southern Hemisphere. Hearne, Vojage to the Northern Ocean. Herodotus. Hooper, Tents of the Tuski. Humboldt, Personal Researches. Hunter, Comparative Dictionary of the Non- Aryan Languages of India and High Asia. ,, The Annals of Rural Bengal. Hume, Essays. „ History of England. Inman, Ancient Faiths in Ancient Names. James, Expedition to the Rocky Mountains. Jones, Antiquities of the Southern Indians. LIST OF WORKS QUOTED XXI Journal of the Royal Institution. Jukes, Voyage of the ' Fly.' Karnes, History of Man. Kenrick, Phoenicia. Keppel, Visit to the Indian Archi- pelago. „ Expedition to Borneo. Klemm, Allgemeine Culturgeschichte der Menschheit. ,, Werkzeuge und Waffen. Koelle, Polyglotta Africana. Kolben, History of the Cape of Good Hope. Kolff, Voyage of the * Dourga.' Kotzebue, Voyage Round the World. Labat, A'oyage aux lies de I'Amerique. Labillardiere, Voyage in Search of La Perouse. Lafitau, Moeurs des Sauvages ameri- cains. Laird, Expedition into the Interior of Africa. Lander (R. and J .), Niger Expedition. Lang, Aborigines of Australia. Latham, Descriptive Ethnology. Lecky, History of Rationalism. Lewin, Hill Tracts of Chittagong. „ Wild Races of South-Eastern India. Lichtenstein, Travels in South Africa. Livingstone, Missionary Travels and Researches in South Africa. „ Expedition to the Zambesi. Locke, On the Human Understanding. Lubbock, Prehistoric Times. Lyall, Asiatic Studies. Lyon, Journal during the Voyage of Captain Parry. McGillivray, Voyage of the ' Rattle- snake.' ]\IacLean, Compendium of Kaffir Laws and Customs. M'Lennan, Primitive Marriage. „ Studies in Ancient History. McMahon, The Karens of the Golden Chersonese. Maine, Ancient Law. „ Early Law and Customs. Marco Polo, Travels of. Mariner, Tonga Islands. Marsden, History of Sumatra. Martins, Von dem Rechtszustande unter denUreinwohnern Brasiliens. Merolla, Voyage to Congo (Pinker- ton's Voyages and Travels). Metlahkatlah, published by the Church Missionary Societ5^ Metz, Tribes of the Neilgherries. Middendorf, Sibirische Reise. Mollhansen, Journey to the Pacific. Monboddo, Origin and Progress of Language. Montesquieu, Esprit des Lois. Moor, Notices of the Indian Archi- pelago. Morgan, Proc. Acad. Nat. Sci. Phila- delphia. Moser, The Caucasus and its People. Mouhot, Travels in the Central Parts of Indo-China. Miiller (C. 0.), Scientific Mythology. „ (C. S.), Description de toutes les Nations de I'Empire de Russie. „ (F. G.), Geschichte der Ameri- kanischen Urreligionen. „ (Max), Chips from a German Workshop. „ „ Lectures on Language, First Series. „ „ Lectures on Language, Second Series. Nilsson, On the Stone Age. Olaus Magnus. Ortolan, Justinian. Pallas, Voyages en differentes Pro- vinces de I'Empire de Rnssie. „ Voyages entrepris dans les Gouvernements meridionaux de rEmj)ire de Russie. Park, Travels. Parkyns, Life in Abyssinia. Perouse, La, Voyage autour du Monde. Petherick, Egypt, the Soudan, and Central Africa. Phear, The Aryan Village. XXll LIST OF WORKS QUOTED Pliny, Natural History. Post, Die Anfange des Staats- und Eechtslebens. ,, Der Ursprung des Eechts. „ Die Geschlechtsgenosseu- schaft. „ Bausteine fiir eine allgemeine Eechtswissenschaft. ,, Einleitung in eine Naturwissen- schaft des Eechts. Pregevalsky, From Kulga to Lob Nor. Prescott, History" of Mexico. ,, History of Peru. Prichard, Natural History of Man. Proceedings of the American Academy of Arts and Sciences. Proceedings of the Boston Society of Natural History. Proyart, History of Loango (Pinker- ton's Voyages and Travels). Baffles, History of Java. Eeade, Savage Africa. Eenan, Origine du Langage. Eeport of Committee of Legislative Council of Victoria on the Abori- gines. Eeville, Les Eeligions des Peuples non- civilises. Eichardson, Journal of a Boat Jour- ney. Eink, Greenland. Eobertson, History of America. Eoss, Voyage to Baffin's Bay. Eutimeyer, Beitr. zur Kenntniss der fossilen Pferde. Scherzer, Voyage of the 'No vara.' Schoolcraft, Indian Tribes. Seebohm, The English Village Com- munity. Seeman, A Mission to Viti. Shooter, Kafirs of Natal. Shortland, Traditions and Supersti- tions of the New Zealanders. Smith (A.), Theory of Moral Senti- ments and Dissertation on the Origin of Languages. „ (G.), (Bishop of Victoria), Ten Weeks in Japan. „ (L), History of Virginia. Smith (W.), Voyage to Guinea. Smithsonian Eeports. Snowden and Prall, Grammar of the Mpongwe Language. Nev^' York. Speke, Discovery of the Source of the Nile. Spencer (H.), Principles of Sociology. „ and Duncan, Descriptive , Sociology. Spencer's Principles of Biology. Spiers, Life in Ancient India. Spix and Martius, Travels in Brazil. Sproat, Scenes and Studies of Savage Life. Squiers, Serpent Symbol in America, Stephens, South Australia. Stevenson, Tiavels in South America. Strahlenberg, Description of Eussia, Siberia, and Great Tartary, Systems of Land Tenure. Published by the Cobden Club. Tacitus. Tanner, Narrative of a Captivity among the North American Indians. Taylor, New Zealand and its Inhabit- ants. Tertre, History of the Caribby Islands. Tindall, Grammar and Dictionary of the Namaqua (Hottentot) Lan- guage. Transactions of the American Anti- quarian Society Transactions of the Ethnological Soc. Transactions of the E. S. of Victoria. Tuckey, Expedition to explore Eiver Zaire. Tnpper, Punjab Customary Law. Turner, Nineteen Years in Polynesia. Tylor, Anahuac. „ Early History of Man. Upliam, History and Doctrine of Buddhism in Ceylon. Vancouver, Voyage of Discovery. Vogt, Lectures on Man. Waitz, Anthropologic der Naturvolker. Wake, Chapters on Man. LIST OF W0BK8 QUOTED Wallace, Travels in the Amazons and Eio Negro. „ Malay Archipelago. Watson and Kaye, The People of India. Wedgwood, Introduction to the Dic- tionary of the English Language. „ Origin of Language. Whately (Archbishop of Dublin), Political Economy. Whipple, Report on the Indian Tribes. Whitney, Language, and the Science of Language. Wilkes, United States Exploring Ex- pedition. Williams, Fiji and the Fijians. Wood, Natural History of Man. Wrangel, Siberia and the Polar Sea. Wright, Superstitions of England. Wuttke, Die ersten Stufen der Gesch., der Menschheit. Yate, New Zealand. THE ORIGIN OE CIVILISATION CHAPTER I. INTRODUCTION. THE study of the lower races of men, apart from the dh^ect importance which it possesses in an empire like ours, is of great interest from three points of view. In the first place, the condition and habits of existing savages resemble in many ways, though not in all, those of our own ancestors in a period now long gone by : ^ in the second, they illustrate much of what is passing among ourselves — many customs which have evidently no relation to present circumstances ; and some ideas which are rooted in our minds, as fossils are imbedded in the soil : while, thirdly, we can even, by means of them, penetrate some of that mist which separates the present from the future. In fact, the lower races of men in various parts of the world present us with illustrations of a social condition ^ I am very glad to find that so the general conclusions at which I able and cautious a critic as Mr. have arrived. See his Physics and Bagehot has expressed his assent to Politics, 1872, especially the excellent the line of argument here used, and chapter on ' Nation-making.' 2 IMPOBTANCE OF THE SUBJECT ruder, and more archaic, than any which history records as having ever existed among the more advanced races. Even among civilised peoples, however, we find traces of former barbarism. Not only is language in this re- spect extremely instructive ; but laws and customs are often of very ancient origin, and contain symbols which are the relics of former realities. Thus the use of stone knives in certain Egyptian ceremonies points to a time when that people habitually used stone implements. Again, the form of marriage by purchase (coemptio) among the Eomans indicates a period in their history when they habitually bought wives, as so many savage tribes do now. So also the form of capture in weddings can only be explained by the hypothesis that the cap- ture of wives was once a stern reality. In such cases as these the sequence is obvious. The use of stone knives in certain ceremonies is evidently a case of sur- vival, not of invention ; and in the same way the form of capture in weddings would naturally survive the actual reality, while we cannot suppose that the reality would rise out of the symbol. It must not be assumed, however, that the con- dition of primitive man is correctly represented by even the lowest of existing races. The very fact that the latter have remained stationary, that their manners, habits, and mode of life have continued almost unaltered for generations, has created a strict, and often compli- cated, system of customs, from which the former was necessarily free, but which has in some cases gradually acquired even more than the force of law. In order then, to arrive at a clear idea of this primitive con- dition of the human race, we must eliminate these NATURE OF THE EVIBENOE 3 customs from our conception of that condition ; and this we are best enabled to do by a comparison of savao;e tribes belonofino; to different families of the human race. Although the differences of race, of geographical position, and of their general surroundings, have neces- sarily led to considerable divergencies in the social and mental development of different tribes, still I have en- deavoured to show that, in the main, the development of higher and better ideas as to Marriage, Relationships, Law, Religion, &c., has followed in its earlier stages a very similar course even in the most distinct races of man ; and when we find customs and ideas which to us seem absurd or illogical, reappearing in separate families of mankind at the same stage of development, we may safely conclude that, however absurd they may appear to us, they rest on some ground which once appeared sufficient, and are no unmeaning or insignificant acci- dents. It has been said by some writers that savages are merely the degenerate descendants of more civilised an- cestors, and I am far from denying that there are cases of retrogression. But, in the first place, a tribe which had sunk from civilisation into barbarism would by no means exhibit the same features as one which had risen into barbarism from savagery. And, what is even more important, races which fall back in civilisation diminish in numbers. The whole history of man shows how the stronger and progressive increase in numbers, and drive out the weaker and lower races. I have endeavoured, for instance, to show that the ideas on the subject of relationships which are prevalent among the less B 2 4 EVIDENCE OF PBOGBESS advanced races, would naturally arise in the course of progress, but are inconsistent with the theory of degra- dation. So, again, a people who trusted in luck would have no chance in the struggle for existence against one which believed in law : if we find a belief in fetichism interwoven with the religion of even the highest races, it is because these races were Fetichists before they became Buddhist, Mahometan, or Christian. A tribe in which the feeling of relationship was weak and ill- defined would be at a great disadvantage as compared with one in which the family feeling was strong. Hence, although we are very far as yet from having arrived at such a result, I believe it will be possible for us to realise to ourselves a condition through which cur ancestors must have passed in pre-historic times — one more primitive than any of which we have at present an actual example. At any rate, it cannot be doubted that the careful study of manners and customs, traditions and supersti- tions, will eventually solve many difficult problems of Ethnology. This mode of research, however, requires to be used with great caution, and has in fact led to many erroneous conclusions. For instance, in more than one case, savage races have been regarded as de- scendants of the Ten Tribes, because their customs offered some singular points of resemblance with those recorded in the Pentateuch. In these cases, a wider acquaintance with the manners and customs of savage races would have shown that these coincidences, so far from being, as supposed, peculiar to these tribes, were, in fact, common to several, if not to most, of the prin- cipal races of mankind. Much careful study will, there- IMFOETANGE OF THE SUBJECT 5 fore, be required before this class of evidence can be used with safety, though I doubt not that eventually it will be found most instructive. The study of savage life is, moreover, as I have already observed, of peculiar importance to us, forming, as we do, part of a great empire, with, colonies in every part of the world, and fellow-citizens in many stages of civilisation. Of this our Indian possessions afford us a good illustration. 'We have studied the lowland popu- ' lation,' says Mr. Hunter,-'- ' as no conquerors ever studied ' or understood a subject race. Their history, their habits, 'their requirements, their very weaknesses and preju- ' dices are known, and furnish a basis for those political ' inductions which, under the titles of administrative ' foresight and timely reform, meet popular movements ' half-way. The East India Company grudged neither ' honours nor solid rewards to any meritorious effort to * illustrate the peoples whom it ruled.' . . , ' The practical result now appears. English ad- ' ministrators understand the Aryan, and are almost ' totally ignorant of the non- Aryan, population of ' India. They know with remarkable precision how ' a measure will be received by the higher or purely ' Aryan ranks of the community ; they can foresee ' with less certainty its effect upon the lower or semi- ' Aryan classes, but they neither know nor venture to ' predict the results of any line of action among the 'non- Aryan tribes. Political calculations are impos- ' sible without a knowledge of the j)eople. But the evil ' does not stop here. In the void left by ignorance, ' prejudice has taken up its seat, and the calamity of ^ Non -Aryan Languages of India, p. 2. 6 DIFFICULTY OF THE SUBJECT ' the lion- Aryan races is not merely that they are not ^ understood, but that they are misrepresented.' Well, therefore, has it been observed by Sir Henry Maine, in his excellent work on ' Ancient Law,' that, * even if they gave more trouble than they do, no pains * would be wasted in ascertaining the germs out of * which has assuredly been unfolded every form of ' moral restraint which controls our actions and shapes ' our conduct at the present moment. ... As societies * do not advance concurrently, but at different rates of ' progress, there have been epochs at which men trained ^ to habits of methodical observation have really been * in a position to watch and describe the infancy of ' mankind.' ^ He refers particularly to Tacitus, whom he praises for having ' made the most of such an oppor- ' tunity ;' adding, however, 'but the '' Germany," unlike ' most celebrated classical books, has not induced others ' to follow the excellent example set by its author, and ' the amount of this sort of testimony which we possess ' is exceedingly small.' This is, however, I think, far from being reallv the case. At all ejDochs some ' men trained to habits of ' methodical observation have really been in a ]30sition to watch and describe the infancy of mankind,' and the testimony of our modern travellers is in many cases of the same nature as that for which we are indebted to Tacitus. It must, however, be admitted that our information with reference to the social and moral condition of the lower races of man is certainly very far from being satisfactory, either in extent or in accuracy. Travellers ^ Maine's Ancient Law, p. 120. MENTAL CONDITION OF 8AVAGUS 7 naturally find it far easier to describe the houses, boats, food, dress, weapons, and implements of savages, than to understand their thoughts and feelings. The whole mental condition of a savage is so different from ours, that it is often very difficult to follow what is passing in his mind, or to understand the motives by which he is influenced. Many things appear natural and almost self-evident to him, which produce a very different im- pression on us. ' What ! ' said a negro to Burton, ' am ' I to starve, while my sister has children whom she can ' sell ? ' When the natives of the Lower Murray first saw pack oxen, some of them were frightened and took them for demons ^ with spears on their heads,' while others thought they were the wives of the settlers, because they carried the baggage.-^ Moreover, though savages always have a reason, such as it is, for what they do and what they believe, their reasons often are very absurd. The difficulty of ascertaining what is passing in their minds is of course much enhanced by the difficulty of communicating with them. This has produced many laughable mis- takes. Thus, when Labillardiere enquired of the Friendly Islanders the word for 1,000,000, they seem to have thought the question absurd, and answered him by a word which apparently has no meaning ; when he asked for 10,000,000, they said ' laoalai,' which I will leave unexplained; for 100,000,000, 'laounoua,' that is to say, • nonsense ; ' while for the higher numbers they gave him certain coarse expressions, which he has gravely published in his table of numerals. ^ Taplin, The Narinyeri, p. 53. 8 EBE0B8 ARISING FBOM A mistake made by Dampier led to more serious results. He bad met some Australians, and appre- bending an attack, be says : — ' I discbarged my gun to ^ scare tbem, but avoided sbooting any of tbem ; till, ^finding tbe young man in great danger from tbem, ' and myself in some, and tbat tbougb tbe gun bad a little ' frigbtened tbem at first, yet they had soon learnt to despise ^ it, tossing up tbeir bands, and crying "Poob, poob, ' ^'poob ! " and coming on afresb witb a great noise, I ^ tbougbt it bigb time to cbarge again, and sboot one of ^ tbem, wbicb I did . . . and returned back with my men, ' designing to attempt tbe natives no fartber, being very 'sorry for wbat bad happened already.'^ 'Poob, 'poob,' however, or 'puff, puif,' is tbe name which savages, like children, naturally apply to guns. Another source of error is, tbat savages are often reluctant to contradict wbat is said to tbem. Living- stone calls special attention to this as a character- istic of tbe natives of Africa.^ Mr. Oldfield,^ again, speaking of the Australians, tells us : — ' I have found ' this habit of non- contradiction to stand very much ' in my way when making enquiries of tbem, for, as ' my knowledge of their language was only sufficient 'to enable me to seek information on some points ' by putting suggestive questions, in which they im- ' mediately concurred, I was frequently driven nearly ' to my wits' end to arrive at the truth. A native once ' brought me in some specimens of a species of euca- ^lyptus, and being desirous of ascertaining the habit of ' the plant, I asked, " A tall tree ? " to which bis ready * Pinkerton's Voyages, vol. si. p. 473. ^ Expedition to tlie Zamlbesi, p. 309. 3 Trans. Ethn. Soc. N.S. vol. iii. p. 255. IGNOUANGE OF LANGUAGE 9 ' answer was in the affirmative. Not feeling quite ' satisfied, I again demanded, " A low bush ? " to which ' '' Yes," was also the response.' Again, the mind of the savage, like that of the child, is easily fatigued, and he will then give random answers, to spare himself the trouble of thought. Speaking of the Ahts (N.W.America), Mr. Sproat^ says: — 'The ' native mind, to an educated man, seems generally to ' be asleep ; and if you suddenly ask a novel question, ' you have to repeat it while the mind of the savage is ' awaking, and to speak with emphasis until he has quite ' got your meaning. ... A short conversation wearies ' him, particularly if questions are asked that require ' eflforts of thought or memory on his part. The mind ' of the savage then appears to rock to and fro out of ' mere weakness, and he tells lies and talks nonsense.' ' I frequently enquired of the negroes,' says Park, ' what became of the sun during the night, and whether ' we should see the same sun or a different one, in the ' morning ; but I found that they considered the ques- ' tion as very childish. The subject appeared to them ' as placed beyond the reach of human investigation ; 'they had never indulged a conjecture, nor formed any ' hypothesis, about the matter.' - Such ideas are, in fact, entirely beyond the mental range of the lower savages, w^hose extreme mental in- feriority we have much difficulty in realising. Speaking of the wild men in the interior of Borneo, Mr. Dalton^ says that they are found living 'absolutely ^ Scenes and Studies of Savage ^ Moor's Notices of the Indian Life, p. 120. Archipelago, p. 49. See also Keppel's - Park's Travels, vol. i. p. 265. Expedition to Borneo, vol. ii. p. 10. 10 ABJECT CONDITION OF TUB ^ in a state of nature, who neither cultivate the ground ^ nor live in huts ; who neither eat rice nor salt, and who * do not associate with each other, but rove about some ' woods, like wild beasts ; the sexes meet in the jungle, * or the man carries away a woman from some campong. ^ When the children are old enough to shift for them- * selves, they usually separate, neither one afterwards ' thinking of the other. At night they sleep under some 4arge tree, the branches of which hang low ; on ^ these they fasten the children in a kind of swing ; ' around the tree they make a fire to keep off the wild ' beasts and snakes. They cover themselves with a piece ' of bark, and in this also they wrap their children ; it ' is soft and warm, but will not keep out the rain. The ' poor creatures are looked on and treated by the other ' Dyaks as wild beasts.' Lichtenstein describes a Bushman as presenting ^ the true physiognomy of the small blue ape of Caf- ^ fraria. What gives the more verity to such a com- ^ parison was the vivacity of his eyes, and the flexibility ' of his eye-brows. . . . Even his nostrils and the ^ corners of his mouth, nay his very ears, moved in- ^ voluntarily. . . . There was not, on the contrary, a ' single feature in his countenance that evinced a con- ' sciousness of mental powers.' -^ Under these circumstances it cannot be wondered at that we have most contradictory accounts as to the cha- racter and mental condition of savages. Nevertheless, by comparing together the accounts of different tra- vellers, we can to a great extent avoid these sources of ^ Lichtenstein, vol. ii. p. 224. L0WE8T EACU8 OF MEN 11 error ; and we are very much aided in this by the re- markable similarity between different races. So striking, indeed, is this, that different races in similar stages of development often present more features of resemblance to one another than the same race does to itself in different stages of its history. Some ideas, which seem to us at first inexplicable and fantastic, are yet very widely distributed. Thus among many races a woman is absolutely forbidden to speak to her son-in-law. Franklin ^ tells us that among the American Indians of the far North ' it is considered ^ extremely improper for a mother-in-law to speak or ' even look at him ; and when she has a communication ^ to make to him it is the etiquette that she should turn ' her back upon him, and address him only through the ' medium of a third person.' Further south, among the Omahaws, ^ neither the ^ father-in-law nor mother-in-law will hold any direct ^ communication with their son-in-law.' ^ Harmon says that among the Indians east of the Rocky Mountains the same rule prevails. Baegert ^ mentions that among the Indians of California ' the son-in-law was not ' allowed, for some time, to look into the face of his ' mother-in-law, or his wife's nearest relations, but had ' to step on one side, or to hide himself when these ' women were present.' Lafitau,^ indeed, makes the same statements as re- gards the North American Indians generally. We find ^ Journey to the Shores of the Translated by 0. Rau, in Smith- Polar Sea, vol. i. p. 137. sonian Rep. for 1863-4, p. 368. ^ James's Expedition to the ^ Moeurs des Sauvages Ameri- Rocky Mountains, vol. i. p. 232. cains, vol. i. p. 676. 2 Account of California, 1773. 12 GUBI0U8 CUSTOMS WITH BEFEBENGE TO it among the Crees and Dacotahs, and again in Florida. Rochefort mentions it among the Caribs, and in South America it recurs among the Arawaks. In Asia, among the Mongols and Kalmucks, a woman must not speak to her father-in-law nor sit down in his presence. Among the Ostiaks of Siberia,^ ' une fille mariee evite autant qu'il lui est possible la presence du pere de son mari, tant qu'elle n'a pas d'enfant ; et le mari, pendant ce temps, n'ose pas paroitre devant la mere de sa femme. S'ils se rencontrent par hasard, le mari lui tourne le dos, et la femme se couvre le visage. On ne donne point de nom aux filles ostiakes ; lors- qu'elles sont mariees, les hommes les nomment Imi, femmes. Les femmes, par respect pour leurs maris, ne les appellent pas par leur nom ; elles se servent du mot de Take, hommes.' In China, according to Duhalde, the father-in-law, after the wedding day, * never sees the face of his * daughter-in-law again ; he never visits her,' and if they chance to meet he hides himself.^ A similar custom prevails in Borneo and in the Fiji Islands. In Australia, also. Eyre states that a man must not pronounce the name of his father in-law, his mother-in-law, or his son-in-law. Dubois mentions that in certain districts of Hindo- stan a woman ' is not permitted to speak to her mother- ' in-law. When any task is prescribed to her, she shows ' her acquiescence only by signs ; ' a contrivance, he 1 Pallas, vol. iv. pp. 71, 577. Russie, pt. i. pp. 191-203; pt. ii. He makes the same statement with p. 104. reference to the Samoyedes, loc. cit. ^ Astley's Collection of Voyages, p. 99. See also Miiller, Description vol. iv. p. 91. de toutes les Nations de I'Empire de M0THEB8-IN-LAW 13 sarcastically adds, ' well adapted for securing domestic • tranquillity.' ^ In Central Africa, Gail lie '^ observes that, ' from this moment the lover is not to see the father and mother of his future bride : he takes the greatest care to avoid them, and if by chance they perceive him they cover their faces, as if all ties of friendship were broken. I tried in vain to discover the origin of this whimsical custom ; the only answer I could obtain was, " It's our '' way." The custom extends beyond the relations : if the lover is of a different camp, he avoids all the in- habitants of the lady's camp, except a few intimate friends whom he is permitted to visit. A little tent is generally set up for him, under which he remains all day, and if he is obliged to come out, or to cross the camp, he covers his face. He is not allowed to see his intended during the day, but, when everybody is at rest, he creeps into her tent and remains with her till daybreak.' Among the Kaffirs^ a married woman 'is required to " hlonipa " her father-in-law and all her husband's male relations in the ascending line — that is, to be cut off from all intercourse with them. She is not "allowed to pronounce their names, even men- tally ; and whenever the emphatic syllable of either of their names occurs in any other word, she must avoid it, by either substituting an entirely new word, or at least another syllable, in its place. The son-in-law is placed under certain restrictions towards his mother- in-law. He cannot enjoy her society or remain in the ^ On the People of India, p. 235. ^ KaiRr Laws and Customs, pp. - Caillie's Travels to TimlDuctoo, 95, 96. Yol. i. p. 94. 14 MOTHERS-IN-LAW ^ same hut with her ; nor can he pronounce her name. ^ Among the Bushmen in the far South, Chapman re- counts exactly the same thing, yet none of these obser- vers had any idea how general the custom is. In Australia, among the aborigines of Victoria, ' it ^ is compulsory on the mothers-in-law to avoid the sight ^ of their sons-in-law, by making the mothers-in-law ^ take a very circuitous route on all occasions to avoid * being seen, and they hide the face and figure with the ' rug which the female carries about her.' ^ So strict is the rule, that if married men are jealous of any one, they sometimes promise to give him a daughter in marriage. This places the wife, according to custom, in the position of a mother-in-law, and renders any communication between her and her future son-in-law a capital crime. More or less similar customs occur among the Dyaks, and other races, and cannot possibly be without a cause. Mr. Tylor, who has some very interesting remarks on these customs in his ' Early History of Man,' observes that 'it is hard even to 2:uess what state of thing's ' could have brought them into existence,' nor, so far as I am aware, has any one else attempted to explain them. In the Chapter on Marriage I shall, however point out the manner in which I conceive that they have arisen. Another curious custom is that known in Bearn under the name of La Couvade. Probably every Eno-- lishman who had not studied other races would assume as a matter of course, that on the birth of a child the ^ Keport of Select Committee on Aborigines, Victoria, 1859, p. 73. LA GOUVADU 15 motlier would everywhere be put to bed and nursed. But this is not the case. In many races the father, and not the mother, is doctored when a baby is born. Yet though this custom seems so ludicrous to us, it is very widely distributed. Commencing with South America, DobritzhofFer tells us that ' no sooner do you ^ hear that a woman has borne a child, than you see the ^ husband lying in bed huddled up with mats and skins, * lest some ruder breath of air should touch him, fasting, ^ kept in private, and for a number of days abstaining ^ religiously from certain viands : you would swear it ^ was he who had had the child. ... I had read about ' this in old times, and laughed at it, never thinking I ^ could believe such madness, and I used to suspect ^ that this barbarian custom was related more in jest ' than in earnest ; but at last I saw it with my own ^ eyes among the Abipones.' In Brazil, among the Coroados, Martins tells us that ^ as soon as the woman is evidently pregnant, or has ' been delivered, the man withdraws. A strict regimen ' is observed before the birth ; the man and the woman ' refrain for a time from the flesh of certain animals, and ' live chiefly on fish and fruits.' ^ Further north, in Guiana, Mr. Brett ^ observes that ' some of the men of the Acawoio and Caribi nations, ' when they have reason to expect an increase of their ' families, consider themselves bound to abstain from ^ certain kinds of meat, lest the expected child should, * in some very mysterious way, be injured by their par- ' taking of it. The Acouri (or Agouti) is thus tabooed, ^ Spix's and Martius's Travels in ^ Brett's Indian Tribes of Guiana, Brazil, vol. ii. p. 247. p. 355. 16 REASON FOB LA COJJYABE lest, like that little animal, the child should be meagre : the Haimara, also, lest it should be blind — the outer coating of the eye of that fish suggestmg film or cataract ; the Lahba, lest the infant's mouth should protrude like the labba's or lest it be spotted like the labba, which spots would ultimately become ulcers.' And again : — ' On the birth of a child, the ancient Indian etiquette requires the father to take to his ham- mock, where he remains some days as if he were sick, and receives the cons^ratulations and condolence of his friends. An instance of this custom came under my own observation, where the man, in robust health and excellent condition, without a single bodily ailment, was lying in his hammock in the most provoking manner, and carefully and respectfully attended by the women, while the mother of the new-born infant was cooking — none apparently regarding her ! ' ^ Similar statements have been made by various other travellers, including De Tertre, Giliz, Biet, Fermin, and in fact almost all who have written on the natives of South America. In North America, Bancroft mentions the existence of a similar custom among the natives of California and New Mexico. Remy states that among the Shoshones, when a woman is in labour, the husband also is bound ' to remain in seclusion, away from every one, even from ' his wife.' ^ In Greenland, after a woman is confined, the 'husband must forbear working for some weeks, « neither must they drive any trade during that time ; '^ in Kamskatka, for some time before the birth of a baby, 1 Brett, loG. cit. p. 101. ^ Journey to the Great Salt Lake City, p. 126. 2 Egede's Greenland, p. 196. BEASON FOB LA COUVADE 17 the husband must do no hard work. In South India, Mr. Tylor^ quotes Mr. F. W. Jennings as stating that among natives of the higher castes about Madras, Seringapatam, and on the Malabar Coast, ' a man, at the ' birth of his first son or daughter by the chief wife, or ^ for any son afterwards, will retire to bed for a lunar ^ month, living principally on a rice diet, abstaining *from exciting food and from smokmg.' In Fiji, also, when a child is born, the father, as well as the mother, is careful to abstain from eating anything which might disagree with the infant. Similar notions occur among the Chinese of West Yunnan, among the Dyaks of Borneo, in Melanesia, in Madagascar, on the west coast of Africa, among the Kaffirs, in the north of Spain, in Corsica, and in the south of France, where it is called ' faire la couvade.' While, however, I regard this curious custom as of much ethnological interest, I cannot agree with Mr. Tylor in regarding it as evidence that the races by whom it is practised belong to one variety of the human species.^ On the contrary, I believe that it originated indepen- dently in several distinct parts of the world. It is of course evident that a custom so ancient, and so widely spread, must have its origin in some idea which satisfies the savage mind. Several explanations have been suggested. Professor Max Muller,^ in his ' Chips from a Grerman Workshop,' says : — ' It is clear ' that the poor husband was at first tyrannised over by 'his female relations, and afterwards frio^htened into ^ Tylor's Early History of Man, ^ Chips from a German Work- 2nd ed., p. 301. shop, vol. ii. p. 281. ' Loc. cit. p. 296. 18 SAVAGE IDEAS OK TEE INFLUENCE OF FOOD ^superstition. He then began to make a martyr of ' himself till he made himself really ill, or took to his ' bed in self-defence. Strange and absurd as the cou- ' vade appears at first sight, there is something in it ' with which, we believe, most mothers-in-law can ' sympathise.' Lafitau^ regards it as arising from a dim recollection of original sin ; rejecting the Carib and Abipon explanation, which I have little doubt is the correct one, that they do it because they believe that if the father engaged in any rough work, or was careless in his diet, ' cela feroit mal a I'enfant, et que cet enfant ' participeroit a tons les defauts naturels des animaux * dont le pere auroit mange.' This idea — namely, that a person imbibes the characteristics of an animal which he eats — is very widely distributed. In India, Forsyth mentions that Mahouts often give their elephant ' a piece of a tiger's ^ liver to make him courageous, and the eyes of the ' brown horned owl to make him see well at night.' ^ The Malays at Singapore also give a large price for the flesh of the tiger, not because they like it, but because they believe that the man who eats tiger ^ acquires the ^ sagacity as well as the courage of that animal,'^ an idea which occurs among several of the Indian hill tribes.^ ' The Dyaks of Borneo have a prejudice against the ^ flesh of deer, which the men may not eat, but which is ^allowed to women and children. The reason given ' for this is, that if the warriors eat the flesh of deer 1 Moeurs des Saiivages Am6ri- ^ Keppel's Visit to the Indian cains, vol. i. p. 259. Archipelago, p. 13. 2 Forsyth's Highlands of Central * Dalton's Des. Ethn. of Bengal, India, p. 452. p. 38. SAVAGE IDEAS ON THE INFLUENCE OF FOOD 19 * they become as faint-hearted as that animal.' ^ ' In ^ ancient times those who wished for children used to ^ eat frogs, because that animal lays so many eggs.' ^ The Caribs will not eat the flesh of pigs or of tor- toises, lest their eyes should become as small as those of these animals.^ The Dacotahs eat the liver of the dog, in order to possess the sagacity and courage of that animal.^ The Arabs also imjDute the passionate and revengeful character of their countrymen to the use of camel's flesh.^ In Siberia the bear is eaten under the idea that its flesh ' gives a zest for the chase, and ' renders them proof against fear.' ^ The Kaffirs also prepare a powder 'made of the dried flesh of various ^ wild beasts, intending by the administering of this ' compound to impart to the men the qualities of the ^ several animals.' ^ Tylor^ mentions that an 'English merchant in ^ Shanghai, at the time of the Taeping attack, met his ' Chinese servant carrying home a heart, and asked him ' what he had got there. He said it was the heart of a ^ rebel, and that he was going to take it home and eat 'it to make him brave.' The New Zealanders, after baptising an infant, used to make it swallow pebbles, so that its heart might be hard and incapable of pity.^ Even cannibalism is sometimes due to this idea, and the New Zealanders eat their most formidable enemies partly ^ Keppel's Expedition to Borneo, ^ Astley's Collection of Voyages, vol. i. p. 231. vol. ii. p. 143. ^ Inman's Ancient Faiths in An- " Atkinson's Upper and Lovrer cient Names, p. 383. Amoor, p. 462. ^ Miiller's Geschichte der Ameri- '^ Callaway's Religious System of canischen Urreli^ionen, p. 221. tlie Amazulu, pt. iv. p. 438. * Schoolcraft's Indian Tribes, vol, ^ Early History of Man, p. 131 . ii. p. 80. ^ Yate's New Zealand, p. 82. c 2 20 SAVAGE IDEAS WITS REFERENCE TO PORTRAITS for this reason. Until quite recent times many medical remedies were selected on this principle. It is from the same kind of idea that ' eyebright,' because the flower somewhat resembles an eye, was supposed to be good for ocular complaints. To us the idea seems absurd. Not so to children. I have myself heard a little girl say to her brother, ' If ' you eat so much goose you will be quite silly ; ' and there are perhaps few children to whom the induction would not seem perfectly legitimate. From the same notion, the Esquimaux, ' to render ' barren women fertile or teeming, take old pieces of the * soles of our shoes to hang about them ; for, as they ' take our nation to be more fertile, and of a stronger ' disposition of body than theirs, they fancy the virtue ^ of our body communicates itself to our clothing.' ^ In fact, savages do not act without reason, any more than we do, though their reasons may often be bad ones and seem to us singularly absurd. Thus they have a great dread of having their portraits taken. The better the likeness, the worse they think for the sitter ; so much life could not be put into the copy, except at the expense of the original. Once, when a good deal an- noyed by some Indians, Kane got rid of them instantly by threatening to draw theui if they remained. Catlin tells an amusing, but melancholy, anecdote in reference to this feeling. On one occasion he was drawing a chief named Mahtocheega, in profile. This, when ob- served, excited much commotion among the Indians : ' Why was half his face left out ? ' they asked ; ' Mah- ' tocheega was never ashamed to look a white man in the ^ Egede's Greenland, p. 198. 0UBI0U8 IDEAS WITH BEFUBUNGJE TO PORTRAITS 21 ' face.' Malitocheega himself does not seem to have taken any offence, but Shonka, ' the Bog,* took advan- tage of the idea to taunt him. ' The Englishman knows,' he said, ' that you are but half a man ; he has painted ' but one half of your face, and he knows that the rest ' is good for nothing.' This view of the case led to a fight, in which poor Mahtocheega was shot ; and, as ill luck w^ould have it, the bullet by which he was killed tore away just that part of the face which had been omitted in the drawing. This was very unfortunate for Mr. Catlin, who had great difficulty in making his escape, and lived some months after in fear of his life ; nor was the matter settled until both Shonka and his brother had been killed in revenge for the death of Mahtocheega. Franklin also mentions that the North American Indians ' prize pictures very highly, and esteem any 'they can get, however badly executed, as efficient charms.' ^ The natives of Bornou had a similar horror of being ' written ; ' they said ' that they did not ' like it ; that the Sheik did not like it ; that it was a ' sin ; and I am quite sure, from the impression, that we ' had much better never have produced the book at all.' ^ The Fetich women in Dahome, says Burton, ' were easily ' dispersed by their likenesses being sketched.' ^ In his Travels in Lapland, Sir A. de C. Brooke says : — ' I ' could clearly perceive ^ that many of them imagined ' the magical art to be connected with what I was doing, ^ and on this account showed signs of uneasiness, till ^ Voyage to the Polar Seas^ ii. 6. ^ Mission to the King of Da- ~ Denham's Travels in Africa, home, i. 278. vol. i. p. 27-5. ' '^ Brooke's Lapland, p. 354. 22 SAVAGE IDEAS IN BEGAED TO WRITING ' reassured by some of the merchants. An instance of ' this happened one morning, when a Laplander knocked ' at the door of my chamber, and entered it, as they ' usually did, without further ceremony. Having come ' from Alten to Hammerfest on some business, curiosity ' had induced him, previously to his return, to pay the ' Englishman a visit. After a dram he seemed quite at, ' his ease ; and producing my pencil, I proceeded, as he ' stood, to sketch his portrait. His countenance now * immediately changed, and taking up his cap, he was on ^ the point of making an abrupt exit, without my being ^ able to conjecture the cause. Ashe spoke only his ^ own tongue, I was obliged to have recourse to as- ' sistance ; when I found that his alarm was occasioned ' by my employment, which he at once comprehended, ' but suspected that, by obtaining a likeness of him, I ' should acquire over him a certain power and influence ' that might be prejudicial. He therefore refused to ' allow it, and expressed a wish, before any other steps ^ were taken, to return to Alten, and ask the permission ' of his master.' Mr. Ellis mentions the existence of a similar feeling in Madagascar.^ We can hardly wonder that writing should seem to savages even more magical than drawing. Carver, for instance, allowed the North American Indians to open a book as often as and wherever they pleased, and then told them the number of leaves. ' The only way they ' could account,' he says, ' for my knowledge, was by ' concluding that the book was a spirit, and whispered ^ me answers to whatever I demanded of it.' '^ ^ Three Visits to Madagascar, p. 358. 2 Travels, p. 265. SAVAGE IDEAS IN BEGABD TO WRITING 23 Father Btiegert mentions ^ that ' a certain missionary ^ sent a native to one of his colleagues, with some loaves ^ of bread and a letter statmg their number. The mes- ' senger ate a part of the bread, and the theft was con- ' sequently discovered. Another time when he had to ' deliver four loaves, he ate two of them, but hid the ' accompanying letter under a stone while he was thus ' engaged, believing that his conduct would not be * revealed this time, as the letter had not seen him in ' the act of eating the loaves.' Further north, the Minatarrees, seeing Catlin intent over a copy of the ' New York Commercial Advertiser,' were much puzzled, but at length came to the conclu- sion that it was a medicine- cloth for sore eyes. One of them eventually bought it for a high price.''^ This use of writing as a medicine prevails largely in Africa, where the priests or wizards w^ite a prayer on a piece of board, wash it off, and make the patient drink it. Caillie ^ met with a man who had a great reputation for sanctity, and who made his living by writing prayers on a board, washing them off, and then selling the water, which was sprinkled over various objects and supposed to improve or protect them. Mungo Park on one occasion profited by this idea. ' A Bambarran having,' he says, ' heard that I was a ' Christian, immediately thought of procuring a saphie ; ^ and for this purpose brought out his walha or writing- ' board, assuring me that he would dress me a supper ' of rice if I would write him a saphie to protect him ' from wicked men. The proposal was of too great ^ Smithsonian Report, 1864, p. 379. ^^ 2 American Indians, vol. ii. p. 92. ^ Travels, vol. i. p. 262. 24 USJE OF FBAYEBS AS MEBIGINE ' consequence to me to be refused ; I therefore wrote the ' board full from top to bottom on both sides ; and my ^landlord, to be cerlain of having the whole force of ' the charm, washed the writing from the board into a ^calabash with a little water, and, having said a few ' prayers over it, drank this powerful draught ; after ' which, lest a smgie word should escape, he licked the 'board until it was quite dry.'^ The same practice occurs in India, where, however. Sir A. Lyall tells me that the native practitioner may sometimes be seen openly mixing croton oil in the ink with which he writes his charm. Among the Kirghiz, also, Atkinson tells us that the Mullas sell amulets, ' at the rate of a sheep ' for each scrap of paper ; ' ^ and similar charms are ' in great request among the Turkomans,' ^ and in Afghanistan.^ In Africa, the prayers written as medicine or as amulets are generally taken from the Koran. It is admitted that they are no protection from firearms ; but this does not the least weaken the faith in them, because, as guns were not invented in Mahomet's time, he naturally provided no specific against them.^ The science of medicine, indeed, like that of astro- nomy, and like religion, assumes among savages very much the character of witchcraft. Ignorant as they are of the processes by which life ^ Park's Travels, vol. i. p. 357. Asia, p. 50. See also p. 56. Caillie's Travels to ^ Masson's Travels in Balo- Timbuctoo, vol. i.p. 376. Earth, vol. chistan, Afghanistan, &c., vol. i. pp. "• P- 449. 74, 90, 312 ; vol. ii. pp. 127, 302. 2 Siberia, p. 310. 5 Astley's Collection of Voyages, 3 Vambery's Travels in Central vol. ii. p. 35. SAVAGE IDEAS ON DISEASE 26 is iriaintained, of anatomy and of physiology, the true nature of disease does not occur to them. Thus the negroes universally believe that diseases are caused by evil spirits : ^ among the Kaffirs, ' diseases are all attri- ' buted to three causes — either to being enchanted by an ' enemy, to the anger of certain beings whose abode ' appears to be in the rivers, or to the power of evil ' spirits.' ^ So, again, in Guinea, the native doctors paint their patients different colours in honour of the spirit which is supposed to have caused the disease.^ In West Australia, for the same reason, it is the duty of the doctor to run round and round his patient, shouting as he goes, to keep away the evil spirit.^ Similar theories on the origin and nature of disease occur in various parts of the world, as, for instance, in Siberia, among the Kalmucks, the Kirghiz, and Bash- kirs ; ^ in many of the Indian tribes, as the Abors, Kacharis, Kols, &c. ; ^ in Ceylon ; ^ among the Karens ; ^ in the Andamans ; in the Samoan, Harvey, and other Pacific islands ; ^ in Madagascar, among the Caribs,^^ &c. The consequence of this is that cures are effected by ejecting or exorcising the evil spirit. Among the Kal- mucks, this is the business of the so-called ' Priests,' ^ Pritcliard's Natural History of pp. 123, 169. Man, vol. ii. p, 704. "^ Dalton's Des. Etlmology of ^ LichteDstein, toI. i. p. 255. Bengal, pp. 25, 85. Maclean's Kaffir Laws and Customs, "^ Saint-Hilaire, Le Boudha et sa p. 88. Eeligion, p. 387. ^ Astlej's Collection of Voyages, ® The Karens of the Chersonese, voi.ii.p. 439. Cruickshank, Eighteen pp. 123, 354. Years on the Gold Coast, vol. ii- ^ Turner's Nineteen Years pp. 134, 144. Polynesia, p. 224. Gerland's Cont. * Forrest, Jour. Anthrop. Inst. of W^aitz's Anthrop. vol. vi. p. 682. vol. iii. p. 319. ^° Tylor's Primitive Culture, vol. ^ Muller's Des. de toutes les Na- ii. p. 134. tions de I'Empire de Russie, part i. m 26 SAVAGE IDEAS ON DISEASE who induce the evil spirit to quit the body of the patient and enter some other object. If a chief is ill, some other person is induced to take his name, and thus, as is supposed, ' the evil spirit passes into his body.' ■•• In Eome there was an altar dedicated to the Goddess Fever.^ Certain forms of disease, indeed, are now, and, as we know, have long been, regarded, even among the more advanced nations of the East, as caused by the presence of evil spirits. ' The Assyrians and Baby- lonians,' says the Rev. A. H. Sayce, 'like the Jews ' of the Talmud, believed that the world was swarm- ^ ing with obnoxious spirits who produced the various ^ diseases to which man is liable.' ^ Many savage races do not believe in natural death, and if a man, however old, dies without being wounded, conclude that he must have been the victim of magic. Thus, then, when a savage is ill, he naturally attributes his suiFerings to some enemy within him, or to some foreign object, and the result is a peculiar system of treatment, curious both for its simplicity and uni- versality. ' It is remarkable in the Abiponian (Paraguay) phy- * sicians,' says Father DobritzhofFer,^ ' that they cure ' every kind of disease with one and the same medicine. ' Let us examine this method of healing. They apply 'their lips to the part affected, and suck it, spitting ' after every suction. At intervals they blow upon that ^ De Hell's Steppes of the Gas- vol. i, p. 131. plan Sea, p. 256. * History of the Abipones, vol. ii. 2 Epictetus, trans, by Mrs. Carter, p. 249. See also Azara, Voy. dans vol. i. pp. 91, 104. I'Amer. Merid., vol. ii. pp. 25, 117, 3 Kecords of the Past, pub. by 140, 142. the Society of Biblical Literature, MEDIGAL TBEATMENT AMONG SAVAGES 27 'part of the body which is in pain. That blowing and ' sucking are alternately repeated. . . . This method of ' healing is in use amongst all the savages of Paraguay ' and Brazil that I am acquainted with. . . . The ' Abipones, still more irrational, expect sucking and * blowing to rid the body, of whatever causes pain or ' inconvenience. This belief is constantly fostered by Hhe jugglers with fresh artifices; for when they pre - ' pare to suck the sick man, they secretly put thorns, ' beetles, worms, &c., into their mouths, and spitting ' them out, after having sucked for some time, say to ' him, pointing to the worm or thorn, " See here the ' '' cause of your disorder." At this sight the sick man ' revives, when he thinks the enemy that has tormented ' him is at length expelled i' At first one might almost be disposed to think that some one had been amusing himself at the expense of the worthy father, but we shall find the very same mode of treatment among other races. Martins tells us that the cures of the Gruaycurus (Brazil) ' are very sunple, ' and consist principally in fumigating or in sucking ' the part afifected, on which the Paye spits into a pit, ' as if he would give back the evil principle which he ' has sucked out to the earth and bury it.' ^ In British Guiana, Mr. Brett mentions that, ' if the ' sorcerer observes signs of recovery, he will pretend to ' extract the cause of the complaint by sucking the part ' afi'ected. After many ceremonies, he will produce from ' his mouth some strange substance, such as a thorn or ' gravel-stone, a fish bone or bird's claw, a snake's tooth ' or a piece of wire, which some malicious yauhahu is ^ Travels in Brazil, vol. ii. p. 77< 28 MEDICAL TEEATMENT AMONG SAVAGES ' supposed to have inserted in the affected part.' ^ The Mexican doctors pretend to extract a piece of bone or^ some other object, which they then indicate to the patient as having been the cause of his suffering.^ In North America, among the Carohna tribes, ' the ' theory was that all distempers were caused by evil ' spirits.' ° Father Baegert mentions that the Californian sor- cerers blow upon and suck those who are ill, and finally show them some small object, assuring them that it had been extracted, and that it was the cause of the pain. Wilkes thus describes a scene at Wallawalla, on the Columbia River : — ' The doctor, who was a woman, ' bending over the body, began to suck his neck and ' chest in different parts, in order more effectually to ' extract the badsj)irit. She would every now and then ^ seem to obtain some of the disease, and then faint ' away. On the next morning she was still found suck- ling the boy's chest. . . . So powerful was the influence ' operated on the boy that he indeed seemed better. , . . ' The last time Mr. Drayton visited the doctress, she ' exhibited a stone, about the size of a goose's egg, saying ' that she had taken the disease of the boy out of him.' ^ Among the Prairie Indians, also, all diseases are treated alike, being referred to one cause, viz., the presence of an evil spirit, which must be expelled. This the medicine-man ' attempts, in the first place, by ' certain incantations and ceremonies, intended to secure ^ Brett's Indian Tribes of Guiana, Soutliern Indians, p. 31. p. 364. 4 United States Exploring Expe- ^ Bancroft, Native Races of the dition, vol. iv. p. 400. See also Pacific States, vol. ii. p. 602. Jones's Antiquities of the Southern ^ Jones's Antiquities of the Indians, pp. 29, 30. MEDICAL TREATMENT AMONG SAVAGES 29 ' the aid of the spirit or spirits he worships, and then ' by all kinds of frightful noises and gestures, and suck- ' ing over the seat of pain with his mouth.' ^ Speaking of the Hudson's Bay Indians, Hearne says : — ' Here it * is necessary to remark that they use no medicine either ^ for internal or external complaints, but perform all ' their cures by charms — in ordinary cases sucking the ' part affected, blowing and singing/ ^ Again, in the extreme North, Crantz tells us that among the Esquimaux old women are accustomed ^ to ' extract from a swollen leg a parcel of hair or scraps ' of leather ; they do it by sucking with their mouth, ' which they had before crammed full of such stuff.' ^ Passing on to the Laplanders, we are told that if any one among them is ill, a wizard sucks his forehead and blows in his face, thinking thus to cure him. Among the Tunguses the doctor sucks the forehead of his patient. In South Africa, Chapman thus describes a similar custom : — A man having been injured, he says, 'our ' friend sucked at the wound, and then . . . extracted ' from his mouth a lump of some substance, which wa& ' supposed to be the disease.' ^ In New Zealand,^ each disease was regarded as being caused by a particular god ; thus ' Tonga was the ' god who caused headache and sickness : he took up his ' abode in the forehead. Mako-Tiki, a lizard god, was 1 Schoolcraft's Indian Tribes, vol. ^ Travels in Africa, vol. ii. p. 45. i. p. 250. See also Livingstone's Travels in 2 Yojage to the Northern Ocean, South Africa, p. 130. p^ 189. ^ Taylor's New Zealand and 3 History of Greenland, vol. i. its Inhabitants, p. 34. Shortland,. p. 14. P- 114. 30 SAVAGE IDEAS ON THE CAUSES OF DISEASES ' the source of all pains in the breast ; Tu-tangata-kino ^ was the god of the stomach ; Titi-hai occasioned pains ^ in the ankles and feet ; Rongomai and Tuparitapu ' were the gods of consumption ; Koro-kio presided ^ over childbirth.' ' Sickness,' says Yate/ ' is brought on by the ' " Atua," who, when he is angry, comes to them in ' the form of a lizard, enters their inside, and preys * upon their vitals till they die. Hence they use incan- ^ tations over the sick, with the expectation of either ^ propitiating the angry deity or of driving him away ; ' for the latter of which purposes they make use of ^ the most threatening and outrageous language.' The Stiens of Cambodia believe ' in an evil genius, and ' attribute all disease to him. If any one be suiFering ' from illness, they say it is the demon tormenting ' him ; and, with this idea, make, night and day, an ^ insupportable noise around the patient.' ^ ' Among the Bechuana tribes, the name adopted by ^ the missionaries (for God) is Morimo. . . . Morimo, ^ to those who know anything about it, had been ' represented by rain-makers and , sorcerers as a male- * volent being which . . . sometimes came out and ' inflicted diseases on men and cattle, and even caused ' death.' The word did not at first convey to the ' Bechuana mind the idea of a person or persons, but ^ of a state or disease, or what superstition would style ' bewitched. . . . They could not describe who or what ^ Morimo was, except something cunning or mali- ' cious. . . . They never applied the name to a human Yate's New Zealand, p. 141. Mouhot's Travels in the Central Parts of Indo-China, vol. i. p. 250. _ SAVAGE IDEAS ON THE CAUSES OF DISEASES 31 ' being, except in the way of ridicule, or in adulation ^ to those who taught his greatness, wisdom, and * power.' ^ The same idea occurs in Madagascar. Sibree gives the following account : — ' A woman of rank appointed * for the occasion began to dance, while another, seated ' behind the sick persons, began to beat a worn-out ^ spade, suspended by a string, with a hatchet, quite ^ close to their ears, making a horrid din. The idea of ^ this is to drive the angatra (evil spirit) possessing the ^ sick person, into one of those dancing.' ^ * The good spirits of the departed, Azimo or Bazimo, * may be propitiated by medicines, or honoured by ' offerings of beer or meal, or anything tbey loved while ' in the body ; and the bad spirits, " Mchesi," of whom ^ we have heard only at Litte, and therefore cannot be ^ certain that they belong to the pure native faith, may ^ be prevented by medicine from making raids, and mis- ^ chief in the gardens. A man with headache was heard ' to say, " My departed father is now scolding me ; I feel ^ " his power in my head ; " and he was observed to re- ^ move from the company, make an offering of a little ^ food on a leaf, and pray, looking upwards, to where he ' supposed his father's spirit to be. They are not, like * Mohammedans, ostentatious in their prayers.' ^ The Koussa Kaffirs,^ says Lichtenstein, ascribe all their diseases ' to one of three causes : either to being ' enchanted by an enemy ; to the anger of certain beings, ^ whose abode appears to be in the rivers ; or to the ^ power of evil spirits.' Among the Kols of JN'agpore, 1 Moffat's Trayels, p. 260. ^ Livingstone, toI. ii. p. 520. " Folk Lore Record, vol. ii. p. 46. ^ Lichtenstein, vol. ii. p. 255. 32 SUCKING OUT THE EVIL as Colonel E. T. Dalton tells us, ' all disease in men ' and in cattle is attributed to one of two causes : the ' wrath of some evil spirit who has to be appeased, * or the spell of some witch or sorcerer.' ^ The same is the case with the Cinghalese,^ and indeed with the aboriginal races of India generally. In Australia, we are told by ex-Governor Eyre, in his interesting work, that, ' as all internal pains are ' attributed to witchcraft, sorcerers possess the power ' of relieving or curing them. Sometimes the mouth is ' applied to the surface where the pain is seated, the ' blood is sucked out, and a bunch of green leaves ' applied to the part. Besides the blood, which is ' derived from the gums of the sorcerer, a bone is some- ' times put out of the mouth, and declared to have been ' procured from the diseased part. On other occasions ' the disease is drawn out in an invisible form, and ' burnt in the fire or thrown into the water.' ^ Thus, then, we find all over the world this primitive cure by sucking out the evil, which perhaps even with ourselves lingers among nurses and children in the universal nursery remedy of ' Kiss it and make it well.' These misconceptions of the true nature of disease lead to many other singular modes of treatment. Thus, among the Kukis, the doctor, not the patient, takes the remedies. Consequently, food is generally prescribed, and in cases of severe illness a bufirilo is sacrificed, and the doctor gives a feast.^ 1 Trans. Ethn. Soc, N.S., 1868, Trans. Etlm. Soc, N.S., vol. iii. p. 30. p. 243. 2 St.-Hilaire, Boudlia, p. 387. * Dalton's Des. Ethn. of Bengal, 3 Discoveries in Central Australia, p. 46. vol. ii. p. 360. See also Oldfield, LIFE ATTRIBUTED TO INANIMATE OBJECTS 33 Another curious remedy practised by the Austra- lians is to tie a line round the forehead or neck of the patient, while some kind friend rubs her lips with the * other end of the string until they bleed freely ; this ' blood is supposed to come from the patient, passing ^ along the string.' ^ It naturally follows from this, and is, as will be presently shown, the belief of many of the lower races of men, that death also is the work of vile spirits. Some curious ideas prevalent among savages arise from the fact that as their own actions are due to life, so they attribute life even to inanimate objects. Even Plato assumed that everything which moves itself must have a soul, and hence that the world must have a soul. Heame tells us that the^N^orth American Indians pre- fer one hook that has caught a big fish to a handful that have never been tried ; and that they never put two nets together for fear they should be jealous of one another.^ The Esquimaux thought that Captain Lyons 's musical box was the child of his small hand-organ.^ The Bushmen supposed that Chapman's big waggon was the mother of his smaller ones ; they ' despise an ' arrow that has once failed of its mark ; and on the ' contrary consider one that has hit as of double value. ' They will, therefore, rather make new arrows, how ' much time and trouble soever it may cost them, than ' collect those that have missed, and use them again.' * In Mangaia Mr. Gill informs me that a club or spear ^ English Colony in New South ^ Lyons's Journal, p. 140. Wales, pp. 363, 382, ^ Lichtenstein's Travels in South 2 Loc. cit. p. 330. Africa, vol. ii. p. 271 . D 34 KILLING INANIMATE OBJECTS which has not taken human life is not considered fit to go into battle. Hence many an unoffending person is murdered merely to qualify some weapon for use in war. The natives of Tahiti sowed some iron nails given them by Captain Cook, hoping thus to obtain young ones. They also believe that ' not only all animals, but ' trees, fruit, and even stones, have souls, which at death, ' or upon being consumed or broken, ascend to the divi- ' nity, with whom they first mix, and afterwards pass ' into the mansion allotted to each.' The Tongans were of opinion that ' if an animal ' dies,"^ its soul immediately goes to Bolotoo ; if a stone ' or any other substance is broken, immortality is equally ^ its reward ; nay, artificial bodies have equal good luck * with men, and hogs, and yams. If an axe or a chisel ^ is worn out or broken up, away flies its soul for the * service of the gods. If a house is taken down or any ' way destroyed, its immortal part will find a situation ^ on the plains of Bolotoo.' Hence probably the custom of breaking the implements, &c., buried with the dead. This was done to render them useless, for the savage would not dream of violating the sanctity of the tomb ; but because the implements required to be 'killed,' so that their spirits, like those of the wives and slaves, might accompany their master to the land of shadows. Lichtenstein relates that the king of the Koussa Kaffirs, having broken off a piece of the anchor of a stranded ship, died soon afterwards ; upon which all the Kaffirs made a point of saluting the anchor very respectfully whenever they passed near it, regarding it as a vindictive being. ^ Mariner's Tonga Islands, vol. ii. p. 137. SALUTATIONS 35 Some similar accident probabl}^ gave rise to the ancient Mohawk notion that some great misfortune would happen if any one spoke on Saratoga Lake. A strong-minded Englishwoman, on one occasion, while being ferried over, insisted on talking, and, as she got across safely, rallied her boatman on his superstition ; but I think he had the best of it after all, for he at once replied, ' The Great Spirit is merciful, and knows that a ^ white woman cannot hold her tongue.' ^ The forms of Salutation among savages are some- times very curious, and their modes of showing their feelings quite unlike ours, though they can generally be explained without difficulty. Kissing appears to us to be the natural lano-uao-e of affection. ' It is certain,' says Steele, ' that nature was its author, and it began ^ with the first courtship ; ' but this seems to be quite a mistake. In fact, it was unknown to the Australians, the Papouans, the Indians of Guiana, and the Esqui- maux. The Polynesians did not kiss ; they pressed not the lips, but the nose. The African negroes, we are told, do not like it, otherwise I should have thought that, when once discovered, it would have been uni- versally popular. The New Zealanders, and the Hervey Islanders did not know how to whistle ; ^ the West Africans do not shake hands ; ^ the Batonga (one of the tribes residing on the Zambesi) salute their friends by throwing themselves on their backs on the ground, rolling from side to side, and slapping their thighs with their hands. ^ ^ Burton's Abbeokuta, vol. i. p. ^ Burton's Mission to Dahome, 198. vol. i. p. 36. ^ Traditions of the New Zea- '^ Livingstone's Travels in South landers, p. 131. Africa, p. 551. D 2 36 SALUTATIONS Clapping of hands is a high mark of respect in Loango, and occurs also in various other negro tribes ^ the Dahomans and some of the coast negroes snap their fingers at a person as a compliment. In Loango courtiers salute the king by leaping backwards and forwards two or three times, and swinging their arms. The Fuegians show friendship by jumping up and down,, and amongst ourselves 'jumping for joy' has become proverbial. The Bakaa, one of the Zambesi tribes, have a peculiar prejudice against children who cut the upper front teeth before the lower ones ; and ' you cut your ' top teeth first ' is one of the bitterest insults a man can receive.^ I understand that among English nurses also it is considered to indicate a weakly constitution. The Polynesians and the Malays always sit down when speaking to a superior ; a Chinaman puts on his hat instead of taking it oif. Cook asserts that the people of Mallicollo show their admiration by hissings and the same is the case, according to Casalis, among the Kafiirs.^ In some of the Pacific Islands, in parts of Hindostan ^ and some parts of Africa, it is considered respectful to turn your back to a superior. In the Hervey Islands the head is thrown back, instead of for- wards, as a salutation. Doughty ^ tells us that in Arabia, if a beast is ill, they spit into water and then give it to the animal to drink. Parents also often ask their children to spit at them. He is disposed to consider that this is done as a protection against evil spirits, but does not ^ Livingstone, loc. cit. p. 577. ^ Diilaois, loc cit. p. 210. ^ The BasutoS; by the Kev. E. ^ Travels in Arabia Desert a, by Casalis, p. 234. C. M. Doughty, vol. ii. p. 164. SALUTATIONS 37 :seem very clear on the subject. Some of the New Guinea tribes salute a friend by squeezing their own noses ; ^ on the White Nile,^ in Masai-land, and in Ashantee they spit at you,^ and the people of Iddah shake their fist as a friendly greeting.^ The Todas of the Neilgherry Hills are said to show respect by ' raising ' the open right hand to the brow, resting the thumb on ' the nose ; ' on the upper Nile, Dr. Schweinfurth tells us^ that the mode of showing admiration is to open the mouth wide, and then cover it with the open hand ; and it has been asserted that in one tribe of Esquimaux it is customary to pull a person's nose as a compliment, though it is but right to say that Dr. Rae thinks there was some mistake on the point ; on the other hand, Dr. Blackmore mentions that 'the sign of the Arapahoes, ' and from which they derive their name,' consists in -seizing the nose with the thumb and forefinger.^ It is asserted that in China a coffin is regarded as an appropriate present for an aged relative, especially if he be in bad health. ^ Oomrie, Jour. Anthr. Inst. * Allen and Thomson, vol. i. 1876, p. 108. p. 290. 2 Petherick, pp. 424, 441. ^ Heart of Africa, vol. ii. p. 77. .Schweinfurtli, vol. i. p. 204. « Trans. Ethn. Soc. 1869, p. 310. 3 Dupnis, p. 178. 38 ART AS AN CHAPTEE 11. AET AND OKNAMENTS. THE earliest traces of art yet discovered belong to the Stone Age — to a time so remote that the reindeer was abundant in the south of France, and that probably, though on this point there is some doubt, even the mammoth had not entirely disappeared. These works of art are sometimes sculptures, if one may say so, and sometimes drawings or etchings made on bone or horn with the point of a flint. They are of peculiar interest, both as being the most ancient works of art known to us — older than any Egyp- tian statues," or any of the Assyrian monuments — and also because, though so ancient, they show really con- siderable skill. There is, for instance, a certain spirit about the subjoined group of reindeer (fig. 1), copied from a specimen in the collection of the Marquis de Yibraye. The mammoth (PI. I.) represented on the opposite page, though less artistic, is perhaps even more interesting. It is scratched on a piece of mammoth's tusk, and was found in the cave of La Madeleine in the Dordogne. It is somewhat remarkable that while even in the Stone Period we find very fair drawings of animals, yet in the latest part of the Stone Age, and throughout that of Bronze, they are almost entirely wanting, and the >c--<'^^ JETHNOLOGIGAL GHABACTUB 39 ornamentation is confined to various combinations of straight and curved lines and geometrical patterns. This, I believe, will eventually be found to imply a difference of race between the population of Western Europe at these different periods. Thus at present the Esquimaux (see figs. 2-4) are very fair draughtsmen, while the Polynesians, though much more advanced in many ways, and though skilful in ornamenting both themselves and their weapons, have very little idea GKOUr OP KEINDEEil indeed of representing animals or plants. Their tattoo- ings, for instance, and the patterns on their weapons, are, like the ornaments of the Bronze Age, almost in- variably of a geometrical character. Representations of animals and plants are not, indeed, entirely wanting ; but, whether attempted in drawing or in sculpture, they are always rude and grotesque. With the Esquimaux the very reverse is the case : among them we find none of those graceful spirals, and other geometrical patterns, 4P ABT AS AN SO characteristic of Polynesia ; but, on the other hand, their weapons are often covered with representations of animals and hunting scenes. Thus Beechey,^ de- scribing the weapons of the Esquimaux at Hotham's Inlet, says : — ' On the outside of this and other instruments there ^ were etched a variety of figures of men, beasts, birds, ' &c., with a truth and a character which showed the ' art to be common among them. The reindeer were * generally in herds ; in one picture they were pursued ' by a man in a stooping posture, in snow-shoes ; in * another he had approached nearer to his game, and ^ was in the act of drawing his bow. A third repre- ^ sen ted the manner of taking seals with an inflated skin ^ of the same animal as a decoy ; it was placed upon the ' ice, and not far from it was a man lying upon his ' belly, with a harpoon ready to strike the animal when ' it should make its appearance. Another was dragging * a seal home upon a small sledge ; and several baidars ^ were employed harpooning whales which had been * previously shot with arrows ; and thus, by comparing ' one with another, a little history was obtained which ' gave us a better insight into their habits than could be ' elicited from any signs or intimations.' Some of these drawings are represented in figs. 2-4, which are taken from specimens presented by Captain Beechey to the Ashmolean Museum at Oxford. Hooper^ also mentions drawings among the Tuski, especially ' a sealskin tanned and bleached perfectly ' white, ornamented all over in painting and staining ^ Narrative of a Voyage to the Pacific. yoI. i. p. 251. 2 Tents of the Tuski, p. 65. ETHNOLOGICAL CEAEACTEB 43 42 ART AS AN ETHNOLOGICAL GHABAGTEB ' with figures of men, boats, animals, and delineations of ' wliale-fishing, &c.' In the same way we may, I think, fairly hope even- tually to obtain from the ancient drawings of the bone caves a better insight into the habits of our predeces- sors in Western Europe ; to ascertain, for instance, whether their reindeer were domesticated or wild. As yet, however, mere representations of animals have been met with, and nothing has been found to supplement in any way the evidence derivable from the imple- ments, &c. But though we thus find traces of art — simple, in- deed, but by no means contemptible — in very ancient times, and among very savage tribes, there are also other races who are singularly deficient in this respect. Thus, though some Australians are capable of mak- ing rude drawings of animals, &c., others, on the con- trary, as Oldfield ^ tells us, ' seem quite unable to ' realise the most vivid artistic representations. On ' being shown a large coloured engraving of an abo- ' riginal New Hollander, one declared it to be a ship, ^ another a kangaroo, and so on ; not one of a dozen ' identifying the portrait as having any connection with ' himself. A rude drawing, with all the lesser parts ' much exaggerated, they can realise. Thus, to give ' them an idea of a man, the head must be drawn dis- ' proportionately large.' Dr. CoUingwood,^ speaking of the Kibalans of For- ' mosa, to whom he showed a copy of the ' Illustrated * London News,' tells us that he found it ' impossible 1 Trans. Ethn. Soc, N.S., vol. iii. p. 227. - Ibid. vol. vi. p. 139. ABT IN AFBIGA 43 ' to interest them by pointing out the most striking ' illustrations, which they did not appear to compre- ' hend; Denham in his ' Travels in Central Africa,' says that Bookhaloom, a man otherwise of considerable in- telligence, though he readily recognised figures, could not understand a landscape. ' I could not,' he says, ^ make him understand the intention of the print of the ' sand wind in the desert, which is really so well described ' by Captain Lyons's drawing ; he would look at it up- ' side down ; and when I twice reversed it for him he ' exclaimed 5 *' Why ! why ! it is all the same." A ' camel or a human figure was all I could make him *" understand, and at these he was all agitation and ' delight — " Gieb ! gieb ! " — Wonderful ! wonderful ! ' The eyes first took his attention, then the other ' features ; at the sight of the sword he exclaimed, ' " Allah ! Allah ! " and on discovering the guns, ' instantly exclaimed, " Where is the powder ? " ' ^ So also the Kaffir has great difiiculty in understand- ing drawings, and perspective is altogether beyond him. Central and Southern Africa seem, indeed, to be very backward in matters of art. Still, the negroes are not altogether deficient in the idea. Their idols cannot be called, indeed, works of art, but they often not only represent men, but give some of the African character- istics with grotesque fidelity. The Kaffirs also can carve fair representations of animals and plants, and are fond of doing so. The handles of their spoons are often shaped into unmistak- able likenesses of girafi*es, ostriches, and other animals. 1 Denham's Travels in Africa, vol. i. p. 167. 44 THE QUIPPU As to the Bushmen, we have rather different ac- counts. It has been stated by some that they have no idea of perspective, nor of how a curved surface can possibly be represented on a fiat piece of paper ; while, on the contrary, other travellers assert that they readily recognise drawings of animals or flowers. The Chinese, although so advanced in many ways, are, we know, very deficient in the idea of perspective. We may safely conclude that no race of men in the Stone Age had attained the art of communicating facts by means of letters, or even by the far ruder system of picture-writing ; nor does anything, perhaps, surprise the savage more than to find that Europeans can com- municate with one another by means of a few black scratches on a piece of paper. Even the Peruvians had no better means of record- ing events than the Quippu or Quipu, which was a cord about two feet long, to which a number of different coloured threads were attached in the form of a fringe. These threads were tied into knots, whence the name Quippu, meaning a knot. These knots served as cyphers, and the various threads had also conventional meanings attached to them, indicated by the various colours. This singular and apparently very cumbersome mode of assisting the memory reappears in China and in Africa. Thus, ' As to -^ the original of the Chinese characters, ' before the commencement of the monarchy, little cords ' with sliding knots, each of which had its particular ' signification, were used in transacting business. These ' are represented in two tables by the Chinese, called ' Hotu, and Lo-shu. The first colonies who inhabited ^ Astley's Collection of Voyages, vol. iv. p. 194. PIOTUBE-WBITING 45 ' Sechicen had no other literature besides some arith- ' metical sets of counters made with little knotted cords ^ in imitation of a string of round beads, with which ^ they calculated and made up all their accounts in com- ' merce.' Again, in West Africa, we are told that the people of Ardrah ^ ' can neither write nor read. They ' use small cords tied, the knots of which have their ' signification. These are also used by several savage ' nations in America.' It seems not impossible that tying a knot in a pocket-handkerchief may be the direct lineal representative of this ancient and widely-extended mode of assisting the memory. The so-called picture-writing is, however, a great advance. Yet from representations of hunts in general, such as those of the Esquimaux (see figs. 2-4), it is indeed but a step to record pictorially some particular hunt. Again, the Esquimaux almost always places his mark on his arrows, but I am not aware that any Poly- nesian ever conceived the idea of doing so. Thus we get among the Esquimaux a double commencement, as it were, for the representation of ideas by means of signs. This art of pictorial writing was still more advanced among the Eed Skins. Thus Carver tells us that on one occasion his Chipeway guide, fearing that the Nau- dowessies, a hostile tribe, might accidentally fall in with and attack them, ' peeled the bark from a large tree near ' the entrance of a river, and with wood- coal mixed with ' bear's grease, their usual substitute for ink, made in an ' uncouth but expressive manner the figure of the town of ' the Ottagaumies. He then formed to the left a man ^ Astley's Collection of Voyages, vol. iii. p. 71. 46 PIOTURE.WBITING ' dressed in skins, by which he intended to represent a '- Naudowessie, with a line drawn from his mouth to that ' of a deer, the symbol of the Chipeways. After this he ' depicted still farther to the left a canoe as proceeding ^ up the river, in which he placed a man sitting with ' a hat on ; this figure was designed to represent an ^ Englishman, or myself, and my Frenchman was drawn ^ with a handkerchief tied round his head, and rowing ' the canoe ; to these he added several other significant ' emblems, among which the pipe of peace appeared ' painted on the prow of the canoe. The meaning he ^ intended to convey to the Naudowessies, and which I ' doubt not appeared perfectly intelligible to them, was ' that one of the Chipeway chiefs had received a speech ' from some Naudowessie chiefs at the town of the Otta- ' gaumies, desiring him to conduct the Englishman, who ^ had lately been among them, up the Chipeway river ; ' and that they thereby required that the Chipeway, ' notwithstanding he was an avowed enemy, should ' not be molested by them on his passage, as he had the ' care of the person whom they esteemed as one of their ' nation.' ^ An excellent account of the Red Skin pictorial art is given by Schoolcraft in his ' History of the Indian ' Tribes in the United States.' Fig. 5 represents the census-roll of an Indian band at Mille Lac, in the territory of Minnesota, sent in to the United States agent by Nago-nabe, a Chipeway Indian, during the progress of the annuity payments in 1849. The Indians generally denote themselves by their ' totem,' or family sign ; but in this case, as they all had ^ Carver's Travels, p. 418. INDIAN CENSUS-nOLL 47 II Fie. 5 mill (1 13 ^- 20 10 II I I I 1 I I I I 23 •26 C) 28 32 "C) 34 II INDIA2^ CEI^SUS-EOLL 48 INDIAN TOMBSTONES the same totem, he had designated each family by a sign denoting the common name of the chief. Thus number 5 denotes a catfish, and the six strokes indicate that the Catfish's family consisted of six individuals ; 8 is a beaver skin, 9 a sun, 13 an eagle, 14 a snake, 22 a buiFalo, 34 an axe, 35 the medicine-man, and so on. Fig. 6 Fig. 7 IN-DIAN GKAVE-POSTS (Schoolcraft, vol. i. pi. 60) Fig. 6 is the record of a noted chief of the St. Mary's band, called Shin-ga-ba-was-sin, or the Image-stone, who died on Lake Superior in 1828. He was of the totem of the crane, as indicated by the figure. The six strokes on the right, and the three on the left, are marks of honour. The latter represent three important general treaties of peace in which he had taken part at various times.^ Among the former marks are included his ^ Schoolcraft, Indian Tribes, vol. i. p. 357. FIGTUEE- WRITING 49 Fie. 8 presence under Tecumseh, at the battle of Moravian- town, where he lost a brother. Fig. 7 represents the adjedatig, or tomb-board, of Wabojeeg, a celebrated war-chief, who died on Lake Superior, about 1793. He was of the family or clan of the reindeer. This fact is symbolised by the figure of the deer. The reverse position denotes death. His own personal name, which was the White Fisher, is not noticed. The seven marks on the left denote that he had led seven war parties. The three per- pendicular lines below the totem represent three wounds received in battle. The figure of a moose's head re- lates to a desperate conflict with an en- raged animal of this kind. Fig. 8 is copied from a bark letter which was found above St. Anthony's Falls in 1820. ' It consisted of white birch bark, and the figures had E 50 PIGTUBE'WBITING IN JSfOETH AMERICA ' been carefully drawn. H^o. 1 denotes the flag of tlie * Union : No. 2 the cantonment, then recently established, ' at Cold Spring, on the western side of the cliffs, above ' the influx of the St. Peters : No. 4 is the symbol of the ' (Commanding oflicer (Colonel H. Leavenworth), under ^ w^hose authority a mission of peace had been sent into ^ the Chippewa country : No. 11 is the symbol of ^ Chakope, or the Six, the leading Sioux chief, under ' whose orders the party moved : No. 8 is the second ' chief, called Wabedatunka, or the Black Dog. The ' symbol of his name is No. 10 ; he has fourteen lodges. ^ No. 7 is a chief, subordinate to Chakope, with thirteen ' lodges, and a bale of goods (No. 9), which was devoted ' by the Government to the objects of the peace. The ^ name of No. 6, whose wigwam is No. 5, with thirteen ^ subordinate lodges, was not given.' ^ This was intended to imply that a party of Sioux, headed by Chakope, and accompanied or at least coun- tenanced by Colonel Leavenworth, had come to this spot in the hope of meeting the Chippewa hunters and con- cluding a peace. The Chippewa chief, Babesacundabee, who found this letter, read off its meaning without doubt or hesitation. On one occasion a party of explorers, with two Indian guides, saw, one morning, just as they were about to start, a pole stuck in the direction they were going, and holding at the top a piece of bark, covered with drawings, intended for the information of any other Indians who might pass that way. This is represented in fig. 9. No. 1 represents the subaltern officer in command 1 Schoolcraft's Indian Tribes, vol. i. pp. 352, 353. INDIAN BARK LETTER 51 of tlie party. He is drawn with a sword, to denote his rank. No. 2 denotes the secretary. He is represented as holding a book, the Indians having understood him to be an attorney. No. 3 represents the geologist, ap- propriately indicated by a hammer. Nos. 4 and 5 are attaches ; No. 6 the interpreter. The group of figures marked 9 represents seven infantry soldiers, each of whom, as shown in group No. 10, was armed with a musket. No. 15 denotes that they had a separate fire, and constituted a separate mess. Figs. 7 and 8 repre- sent the two Chippewa guides. These are the only Fig. 9 v: iffiflfflr^ INDIAN BAKE LETTER human figures drawn without the distinguishing symbol of a hat. This was the characteristic seized on by them, and generally employed by the Indians, to distinguish the Red from the White race. Figs. 11 and 12 repre- sent a prairie hen and a green tortoise, which constituted the sum of the preceding day's chase, and were eaten at the encampment. The inclination of the pole was de- signed to show the course pursued, and there were three hacks in it below the scroll of bark, to indicate the esti- mated length of this part of the journey, computing from water to water. The following figure (fig. 10) gives the biography of Wingemund, a noted chief of the E 2 52 INDIAN JBIOGBAPEY Delawares. 1 shows that it belonged to the oldest branch of the tribe, which use the tortoise as their symbol ; 2 is his totem or symbol ; 3 is the sun, and the ten strokes represent ten war parties in which he was engaged. Those figures on the left represent the captives which he made in each of his excursions, the men being distin- guished from the women, and the captives being denoted by having heads, while a man without his head is of course a dead man. The central figures represent three Fi&. 10 INDIAN BIOGRAPHY forts which he attacked ; 8 one on Lake Erie, 9 that of Detroit, and 10 Fort Pitt, at the junction of the Alle- ghany and the Monongahela. The sloping strokes denote the number of his followers.^ Fig. 1 1 represents a petition presented to the Presi- dent of the United States for the right to certain lakes (8) in the neighbourhood of Lake Superior (10). No. 1 represents Oshcabawis, the leader, who is of ^ Schoolcraft, vol. i. p. 353. INDIAN PETITION 53 54 8AVAGE 0ENAMENT8 the Crane clan. The eyes of his followers are all con- nected with his to symbolise unity of views, and their hearts to denote unity of feeling. No. 2 is Wai-mit-tig- oazh, whose totem is a marten ; No. 3 is Ogemagee- zhig, also a marten ; 4 is another marten, Muk-o-mis-ud- ains, the Little Tortoise ; 5 is 0-mush-kose, the Little Elk, belonging, however, to the Bear totem ; 6 belongs to the Manfish totem, and 7 to the Catfish. The eye of the leader has a line directed forwards to the President, and another backwards to the lakes (8). The manner in which such picture-writing would ultimately have led to the use of an alphabet, would probably have been that the drawings would have come to represent, first a word, and then a sound, being at the same time simplified and conventionalised. In some places of Western Europe, rock sculptures have been discovered, to which we cannot yet safely ascribe any meaning, but on which perhaps the more complete study of the picture-writing of modern savages may eventually throw some light. We will now pass to art as applied to the purposes of personal decoration. Savages are passionately fond of ornaments. In some of the very lowest races, indeed, the women are almost undecorated, but that is only be- cause the men keep all the ornaments themselves. As a general rule, we may say that Southerners ornament themselves, Northerners their clothes. In fact, all savage races who leave much of their skin uncovered delight in painting themselves in the most brilliant colours they can obtain. Black, w^hite, red, and yellow are the favourite, or rather, perhaps, the commonest colours. Although perfectly naked, the Australians of Botany SAVAG:E 0BNAMENT8 55 Bay were by no means without ornaments. Tliey painted themselves with red ochre, white clay, and charcoal ; the red was laid on in broad patches, the white generally in stripes, or on the face in spots, often with a circle round each eye ; ^ through the septum of the nose they wore a bone as thick as a man's finger and five or six inches long. This was of course very awk- ward, as it prevented them from breathing freely through the nose, but they submitted cheerfully to the incon- venience for the sake of appearance. 4ir They had also necklaces made of shells, neatly cut and strung together ; earrings, bracelets of small cord, and strings of plaited human hair, which they wound round their waists. Some also had gorgets of large shells hanging from the neck across the breast. On all these things they placed a high value. Spix and Martins ^ thus describe the ornaments of a Coroado woman : — ' On the cheek she had a circle, and ' over that two strokes ; under the nose several marks ' resembling an M ; from the corners of the mouth to ' the middle of the cheek were two parallel lines, and ' below them on both sides many straight stripes ; ' below and between her breasts there were some con- ' nected segments of circles, and down her arms the ' figure of a snake was depicted. This beauty wore no ' ornaments, except a necklace of monkeys' teeth.' In Tanna ' one would have the one half of his face ' smeared with red clay, and the other the plain dark ' copper skin ; another would have the brow and cheeks ' red ; another would have the brow red and cheeks ^ Hawkesworth's Voyages, vol. iii. p. 635. ^ Travels in Brazil, vol. ii. p. 224. 56 SAVAGE ORNAMENTS ' black ; another all the face red, and a round, black, ^ glittering spot on the forehead ; and another would * have his face black all over. The black all over, by the ' way, was the sign of mourning.' ^ The savage also wears necklaces and rings, bracelets and anklets, armlets and legiets — even, if I may say so, bodylets. Round their bodies, round their necks, round their arms and legs, their fingers, and even their toes, they wear ornaments of all kinds. From their number and weight these must sometimes be very inconvenient. Lichtenstem saw the wife of a Beetuan chief wearing no less than seventy-two brass rings. A South African chieftainess, visited by Living- stone,^ wore ' eighteen solid brass rings, as thick as ' one's finger, on each leg, and three of copper under * each knee ; nineteen brass rings on her left arm, and ' eight of brass and copper on her right ; also a large ' ivory ring above each elbow. She had a pretty bead ^ necklace, and a bead sash encircled her waist.' Nor are they particular as to the material : copper, brass, or iron, leather or ivory, stones, shells, glass, bits of wood, seeds, or teeth — nothing comes amiss. In South-East Island, one of the Louisiade Archipelago, M'Grillivray even saw several bracelets made each of a lower human jaw, crossed by a collar bone ; and other travellers have seen brass curtain rings, the brass plates for keyholes, the lids of sardine cases, and other such incongruous objects, worn with much gravity and pride. The Felatah ladies in Central Africa spend several ^ Turner's Nineteen Years in Polynesia, p. 5. ~ Exp. to the Zambesi, p. 284. CHEEK STUDS— LABBETS 57 hours a day over tlieir toilet. In fact they begin over- night by carefully wrapping their fingers and toes in henna leaves, so that by morning they are a rich purple. The teeth are stained alternately blue, yellow, and purple, one here and there being left of its natural colour, as a contrast. About the eyelids they are very particular ; pencilling them with sulphuret of antimony. The hair is coloured carefully with indigo. Studs and other jewellery are worn in great profusion.^ Not content with hanging things round their necks, arms, ankles, and in fact wherever nature has enabled them to do so, savages also cut holes in themselves for the purpose. The Esquimaux from Mackenzie River westward make two openings in their cheeks, one on each side, which they gradually enlarge, and in which they wear an ornament of stone resembling in form a large stud, and which may therefore be called a cheek stud. Brenchley saw the natives of the Solomon Islands decorated by crabs' claws stuck in the cartilage of the nose.^ Throughout a great part of Western America, and again in Africa, we also find the custom of wearing a piece of wood through the central part of the lower lip. A small hole is made in the lip during infancy, and it is then extended by degrees until it is sometimes as much as two inches long. Some races extend the lobe of the ear until it reaches the shoulder ; others file the teeth in various manners. Thus, among the Rejangs of Sumatra, ' both sexes ^ Laird's Expedition into the Interior of Africa, vol. ii. p. 94. ^ Cruise of the 'Curacoa,' p. 250. 58 OBNAMENTATION OF THE SKIN ^ have the extraordinary custom of filing and otherwise ' disfiguring their teeth, which are naturally very white ' and beautiful, from the simplicity of their food. For ^ files they make use of small whetstones of different ^ degrees of fineness, and the patients lie on their backs * during the operation. Many, particularly women of * the Lampong country, have their teeth rubbed down ^ quite even with the gums ; others have them formed ^ in points, and some file off no more than the outer ' coat and extremities, in order that they may the ^ better receive and retain the jetty blackness with ' which they almost universally adorn them.' ^ In Dr. Davis's collection is a Dyak skull in which the six front teeth have each been carefully pierced with a small hole, into which a pin with a spherical brass head has been driven. In this way, the upper lip being raised, the shining knob on each tooth would be dis- played.^ Some of the African tribes also chip their teeth in various manners, each community having a fashion of its own. Ornamentation of the skin is almost universal among the lower races of men. In some cases every individual follows his own fancy ; in others, each clan has a special pattern. Thus, speaking of Abeokuta, Captain Burton ^ says : — ' There was a variety of tattoos and orna- mentation, rendering them a serious difiiculty to strangers. The skin patterns were of every variety, from the diminutive prick to the great gash and the large boil-like lumps. They affected various figures — tortoises, alligators, and the favourite hzard, stars, 1 Marsden's History of Sumatra, ^ Thesaurus Craniorum, p. 289. p. 52. 3 Abeokuta, vol. i. p. 104. TRIBE MABK8 69 * concentric circle, lozenges, right lines, welts, gouts of ' gore, marble or button-like knobs of flesli, and ele- ' vated scars, resembling scalds, which are opened for ' the introduction of fetish medicines, and to €xpel evil ' influences. In this country every tribe, sub -tribe, and ^ even family, has its blazon,^ whose infinite diversifica- ^ tions may be compared with the lines and ordinaries * of European heraldry.' ' The Ardrahs ^ make an incision in each cheek, ' turning up a part of the flesh towards the ears and ^healing it in that position. The Mahees are distin- ' guished by three long oblique cuts on one cheek, and a ' cross on the other.' In South Africa the Nyambanas are characterised by a row of pimples or warts, about the size of a pea, and extending from the upper part of the forehead to the tip of the nose. Among the Bachapin Kaffirs, those who have distinguished themselves in battle are allowed the privilege of marking their thigh with a long scar, which is rendered indelible and of a bluish colour by rubbing ashes into the fresh wound. The tribal mark of the Bunns^ (Africa) consists of three slashes from the crown of the head down the face toward the mouth ; the ridges of flesh stand out in bold relief. This painful operation is performed by cutting the skin, and taking out a strip of flesh ; palm oil and wood ashes are then rubbed into the wound, thus causing a thick ridge. The Bornouese in Central Africa have twenty cuts or lines on each side of the ^ See also Baikie's Exploring ^ Dalzel, History of Dahom, Voyage, pp. 77, 294, 336, and es- p. xviii. pecially 460. ^ Trans. Ethn. Soc, vol. v. p. 86. 60 TATTOOING face, which are drawn from the corners of the mouth towards the angles of the lower jaw and cheekbone. They have also one cut m the centre of the forehead, six on each arm, six on each leg, four on each breast, and nine on each side, just above the hips. This makes 91 large cuts, and the process is said to be extremely painful on account of the heat and flies. ^ The islanders of Torres Straits ornament themselves by a large oval scar, slightly raised and neatly made. It is situated on the right shoulder, but some of them have a second on the left. At Cape York many of the natives also had two or three long transverse scars on the chest. Many had also a two-horned mark on each breast, but these differences seemed to depend on the taste of the individual. The custom of tattooing is found almost all over the world, though, as might be expected, it is most developed in hot countries. In Siberia, however, the Ostiak women tattoo the backs of the hands, the fore- arm and the front of the leg. The men only tattoo, on the wrist, the mark or sign which stands as their signature.^ Among the Tuski ^ ' the faces of the women are tat- ' tooed on the chin in diverging lines ; men only make ' a permanent mark on the face for an act of prowess ' or success, such as killing a bear, capturing a whale, ' &c., and possibly also, in war time, for the death of an ' enemy.' The Aleutian Islanders decorate their hands and faces with figures of quadrupeds, birds, flowers, &c. ^ Denham, vol. iii, p. 175. ^ Hooper, the Tents of the 2 Pallas, vol. iv. p. 56. Tuski, p. 37. TATTOOING 61 Among the Tunguses the patterns are generally formed by straight and curved lines. ^ Among the Arabs ^ ' the Aenezi women puncture ^ their lips and dye them blue ; the Serhhan women ^ puncture their cheeks, breasts, and arms, and the ^ Ammour women their ankles.' The Malagasy do not generally tattoo, but the women of the Betsileo tribes, according to Mr. Camp- bell,^ have their arms ' tattooed all over, some of them * having also a kind of open-work collar tattooed round ^ their necks. The breasts of the men were ornamented ' after the same fashion.' Many of the hill tribes of India tattoo.^ Among the Abors, for instance, the men have a cross on the forehead ; the women a smaller one on the upper lip just below the nose, and seven stripes under the mouth. The Khyens are more extensively tattooed, with figures of animals, &c. ; they admit that it is not ornamental, but allege that they were driven to it because their women were naturally so beautiful that they were con- stantly carried off by neighbouring tribes. The Oraon women have three marks on the brow and two on the temple, while the men burn marks on their forearm. The women of Brumer Island, on the south coast of New Gruinea, were tattooed on the face, arms, and front of the body, but generally not on the back, in vertical stripes less than an inch apart, and connected by zigzag markings. On the face these were more complicated, 1 Miiller's Des. de toutes les ^ Sibree's Madagascar and its. Nat. de I'Emp. de Russie, pt. iii. People, p. 221. pp. 58, 112. ^ Dalton's Des. Ethn. of Bengal^ 2 Burckhardt's Notes on the Be- pp. 27, 114, 251. doiiins and Wahabys, vol. i. p. 51. 62 • TATTOOING and on the forearm and wrist they were frequently so elaborate as to resemble lace-work.^ The men were more rarely tattooed, and then only with a few lines or stars on the right breast. Sometimes, however, the markings consisted of a double series of large stars and dots stretching from the shoulder to the pit of the stomach. Not content with the paint already mentioned, the inhabitants of Tanna have on their arms and chests elevated scars, representing plants, flowers, stars, and various other figures. ' The inhabitants of Tazovan, ' or Formosa, by a very painful operation, impress on ' their naked skins various figures of trees, flowers, ^ and animals. The great men in Guinea have their ^ skin flowered like damask ; and in Decan the women ' likewise have flowers cut into their flesh on the fore- ^ head, the arms, and the breast, and the elevated scars ^ are painted in colours, and exhibit the appearance of ' flowered damask.' ^ In the Tonga Islands ' the men are tattooed from ' the middle of the thigh to above the hips. The women ' are only tattooed on the arms and fingers, and there ' very slightly.'^ In the Fiji Islands, on the contrary, the women are tattooed and not the men. In the Gambier Islands, Beechey says,^ ' tattooing is ' so universally practised, that it is rare to meet a man * without it ; and it is carried to such an extent that * the figure is sometimes covered with small checkered ^ lines from the neck to the ankles, though the breast is 1 M'Gillivray's Voyage of the jd. 588. * Kattlesnake,' vol. i. p. 262. ^ Qook's Voyage towards the 2 Forster's Ohservations made South Pole, vol. i. p. 218. during a Voyage round the World, * Beechey, vol. i. p. 138. TATTOOING 63 ^ generally exempt, or only ornamented with a single * device. In some, generally elderly men, the face is ' covered below the eyes, in which case the lines or net- ' work are more open than on other parts of the body, ' probably on account of the pain of the operation, and ' terminate at the upper part in a straight line from ear ^ to ear, passing over the bridge of the nose. With ^ these exceptions, to which we may add the fashion, ' with some few, of blue lines, resembling stockings, ^ from the middle of the thigh to the ankle, the effect is ^ becoming, and in a great measure destroys the appear- ' ance of nakedness. The patterns which most improve 'the shape, and which appear to me peculiar to this ' group, are those which extend from the armpits to ' the hips, and are drawn forward with a curve which ' seems to contract the waist, and at a short distance ' gives the figure an elegance and outline, not unlike ^ that of the figures seen on the walls of the Egyptian ^ tombs.' Fig. 12 represents a Caroline Islander, after Frey- cinet, and gives an idea of the tattooing, though it cannot be taken as representing the form or features character- istic of those islanders. The tattooing of the Sandwich Islanders is less ornamental, the devices being, according to Arago, ' unmeaning and whimsical, without taste, and in general ' badly executed.' ^ Perhaps, however, the most beau- tiful of all was that of the New Zealanders (see figs. 13 and 14), who were generally tattooed in curved or spiral lines. The process is extremely painful, par- ticularly on the lips ; but to shrink from it, or even to 1 Arago's Letters, pt. ii. p. 147. 64 TATTOOING show any signs of suiFering while under the operation^ would be thought very unmanly. The natives used the ' Moko ' or pattern of their tattooing as a kind of Fig. 12 CAEOLINE ISLANDEE signature. The women have their lips tattooed with horizontal lines. To have red lips is thought to be a great reproach.^ ^ For details of Polynesian tat- ploring Expedition : Ethnography, tooing see Hale's United States Fix- p. 40. ARTIFICIAL ALTERATION OF FORM 65 When tastefully executed, tattooing has been re- garded by many travellers as a real ornament. Thus Laird says that some of the tattooing in West Africa ' in the absence of clothing gives a finish to the skin.' ^ Many similar cases might be given in which savages ornament themselves, as they suppose, in a manner which must be very painful. Perhaps none is more remarkable Fig. 13 Fig. 14 HEAD OP NEW ZEALANDEE HEAD OF NEW ZEALANDEE than the practice which we find in several parts of the world of modifying the human form by means of tight bandages. The small size of the Chinese ladies' feet is a well-known case, but is scarcely less mischievous than the compression of the waist as practised in Europe. The Samoans ^ and some of the American tribes even modified the form of the head. One w^ould have 1 Narrative of an Expedition into ^ burner's Nineteen Years in the Interior of Africa, vol. i. p. 291. Polynesia, p. 175. r QQ EAIBDBE88ING supposed that any such compression would have exer- cised a very prejudicial effect on the intellect ; but, as far as the existing evidence goes, it does not appear to do so. The mode of dealing with the hair varies very much in different races. Some races remove it almost entirely, some leave a ridge along the top of the head ; the Kaffir wears a round rmg of hair ; the North American Indian regards it as a point of honour to leave one tuft, in case he ever has the misfortune of being defeated, for it would be mean to cheat his victor of the scalp, the recognised emblem of conquest. The Islanders of Torres Straits twist their hair into long pipe-like ringlets, and also wear a kind of wig pre- pared in the same fashion. Sometimes they shave the head, leaving a transverse crest of hair. At Cape York the hair is almost always kept short.^ In Tanna the women wear it short, but have it all laid out in a forest of little erect curls, about an inch and a half long. The men wear it twelve and eighteen inches long, and have it divided into some six or seven hundred little locks or tresses. Beginning at the roots, every one of these is carefully wound round by the thin rind of a creeping plant, giving it the appearance of a piece of twine. The ends are left exposed for about two inches, and oiled and curled.^ The Fijians give a great deal of time and attention to their hair, as is shown m PI. II. Most of the chiefs have a special hairdresser, to whom they sometimes devote several hours a day. Their heads of hair are often more than three feet in circumference, and Mr. ^ M'Gillivray's Voyage of the ^ Turner's Nineteen Years in ^ Rattlesnake/ pp. 11, 13. Polynesia, p. 77. FEEJEEAN MODES OF DRESSING THE HAIR FIJI HAIBDBESSES 67 Williains measured one which was nearly five feet round. This forces them to sleep on narrow wooden pillows or neck-rests, which must be very uncomfortable. They also dye the hau\ Black is the natural and favourite colour, but some prefer white, flaxen, or bright red. ' On one head/ says Mr. Williams,^ ' all the hair is ' of a uniform height ; but one-third in front is ashy or ^ sandy, and the rest black, a sharply defined separation ^ dividing the two colours. Not a few are so ingeniously ^ grotesque as to appear as if done purposely to excite * laughter. One has a large knot of fiery hair on his ^ crown, all the rest of his head being bald. Another ^ has the most of his hair cut away, leaving three or four ^ rows of small clusters, as if his head were planted with * small paint-brushes. A third has his head bare except * where a large patch projects over each temple. One, ^ two, or three cords of twisted hair often fall from the ' right temple, a foot or eighteen inches long. Some * men wear a number of these braids, so as to form a ^ curtain at the back of the neck, reaching from one ear Ho the other. A mode that requires great care has ^ the hair brought into distinct locks radiating from the * head. Each lock is a perfect cone about seven inches ^ long, having the base outwards ; so that the surface of * the hair is marked out into a great number of small ^ circles, the ends being turned in in each lock, towards ' the centre of the cone. ' ^ In some of the Pacific Islands the natives wear wigs, or tresses of hair, in addition to their own.^ 1 Fiji and the Fijians, vol. i. p. p. 338, et seq. 158. ^ Hale's United States Expl. ^ See, for many further par- Expedition : Ethnography, p. 12. ticulars, Darwin's Descent of Man, P 2 68 SOUDAN EAIBBEE8S Schweinfurtli describes a dandy, belonging to the Dinkas, a negro tribe of the Soudan, whose hair was dyed red, and trained up into points like tongues of flame, standing stiffly up, all round his head. In fact, the passion for self- ornamentation seems to prevail among the lowest as much as, if not more than^ among the more civilised races of man. I 69 CHAPTER III. MAREIAGE AND RELATIONSHIP. NOTHING, perha23S, gives a more instructive insight into the true condition of savages than their ideas on the subject of relationship and marriage ; nor can the great advantages of civihsation be more conclusively proved than by the improvement which it has already effected in the relation between the two sexes. Marriage, and the relationship of a child to its father and mother, seem to us so natural and obvious, that we are apt to look on them as aboriginal and general to the human race. This, however, is very far from being the case. The lowest races have no institution of mar- riage ; true love is almost unknown among them ; and marriage, in its lowest phases, is by no means a matter of affection and companionship. The Hottentots, says Kolben,^ ' are so cold and in- ' different to one another that you would think there ' was no such thing as love between them.' Among the Koussa Kaffirs, Lichtenstein asserts that there is ' no ' feeling of love in marriage.' ^ In North America, the Tinne Indians had no word for ' dear ' or ' beloved ; ' and the Algonquin language is stated to have contained no verb meaning ' to love | ' so that when the Bible was ^ Kolben's Hist, of the Cape of ^ Travels in .South Africa, vol. i. Good Hope, vol. i. p. 162. p. 261. 70 THE POSITION OF WOMEN AMONG 8AVAGES translated by the missionaries into that language it was necessary to invent a word for the purpose. 'In his native state,' says Mr. Morgan,^ 'the (North ' American) Indian is below the passion of love. It is ' entirely unknown among them, with the exception, to 'a limited extent, of the village Indians.' He men- tions elsewhere a case of an Ahahuelin woman named ' Ethabe,' who had been married for three years to a Blackfoot Indian, yet there was no common articulate language which they both understood. They communi- cated entirely by signs, neither of them having taken the trouble to learn the other's language.^ Though the songs of savages are generally devoted to the chase, war, or women, they can very rarely be called love songs. Dr. Mitchell, for instance, who was for several years chairman of the United States Senate Committee on Indian Affairs, mentions that ' neither ' among the Osages nor the Cherokees could there be ' found a single poetical or musical sentiment, founded ' on the tender passion between the sexes. Though ' often asked, they produced no songs of love.' ^ In Yariba (Central Africa),^ says Lander, 'marriage 'is celebrated by the natives as unconcernedly as pos- ' sible : a man thinks as little of taking a wife as of ' cutting an ear of corn — affection is altogether out of ' the question.' The King of Boussa,^ he tells us in another place, ' when he is not engaged in public affairs, ' usually employs all his leisure hours in superintending ^ Systems of Consanguinity and 317. Affinity of the Human Family, p. * R. and J, Lander's Niger Ex- 207. pedition, vol. i. p. 161. 2 Loc. cit. p. 227. ' " Ibid. vol. ii. p. 106. See also 5 Archseol. Americana, vol. i. p. p. 197. ABSBNGI] OF AFFECTION IN MARRIAGE 71 ^ the occupations of his household, and making his own ' clothes. The Midiki (queen) and he have distinct ^ establishments, divided fortunes, and separate inte- * rests ; indeed, they appear to have nothing in com- ' mon with each other, and yet we have never seen so ^friendly a couple since leaving our native country.' On the Gold Coast, ' not even the appearance of ' affection exists between husband and wife.' ^ Among the Mandingoes marriage is merely a form of regulated slavery. Husband and wife ' never laugh or joke to- * gether.' ' I asked Baba,' says Caillie, ' why he did not ' sometimes make merry with his wives. He replied ' that if he did he should not be able to manage them, ^ for they would laugh at him when he ordered them to ' do anything.' ^ According to Galton, Dammara women ' divorce * themselves as often as they like ; ... in fact, the ' spouse was changed almost weekly, and I seldom knew, ' without inquiry, who the pro tempore husband of each *lady was at any particular time.' ^ In India, the Hill tribes of Chittagong, says Captain Lewin, regard marriage 'as a mere animal and con- * venient connection ; ' as the ' means of getting their * dinner cooked. They have no idea of tenderness, nor ' of chivalrous devotion.' ^ Among the Samoyedes^ of Siberia the husbands show little affection for their wives, and, according to Pallas, ' daignent a peine leur dire une parole de ^ douceur.' Further East, in the Aleutian Islands, the 1 Burton's Mission to the King * Hill Tracts of Ohittagong, p. of Dahomey,, vol. ii. p. 190. 116. 2 Travels, vol. i. p. 350. ^ Pallas's Voyages, vol. iv. p. 94. 3 Tropical Soutli Africa, p. 197. 72 ABSENCE OF AFFECTION IN MAEBIAGE marriages, according to Miiller/ ' meritent a peine le ' nom ; ' and the facts lie mentions go far to justify this statement. Among the Guyacurus of Paraguay ' the bonds of * matrimony are so very slight, that when the parties do ^ not like each other they separate without any further * ceremony. In other respects they do not appear to ' have the most distant notions of that bashfulness so ' natural to the rest of mankind.' ^ The Guaranis seem to have been in a very similar condition.^ In North America the marriage tie was by no means regarded as of a religious character.^ In Australia ' little real affection exists between ^ husbands and wives : and young men value a wife ' principally for her services as a slave ; in fact, when * asked why they are anxious to obtain wives, their ' usual reply is, that they may get wood, water, and ' food for them, and carry whatever property they ' possess.' ^ The position of women in Australia seems indeed to be wretched in the extreme. They are treated with the utmost brutality, beaten and speared in the limbs on the most trivial provocation. Few women, says Eyre, ' will be found, upon examination, to be free ' from frightful scars upon the head, or the marks of ' spear wounds about the body. I have seen a young * woman who, from the number of these marks, appeared ' to have been almost riddled with spear wounds. If ^ Des. de toutes les Nat. de I'Em- vol. ii. p. 60. pire de Russie, part iii. p. 129. * Jones, Antiquities of the ^ Charlevoix, Hist, of Paraguay, Southern Indians, p. 67, vol. i. p. 91. ° Eyre's Discoveries, vol. ii. p. 3 Loc. cit. p. 352. See also Azara, 321. See notes. RELATIONSHIP AMONG SAVAGES 73 ^ at all good-looking, their position is, if possible, even ^ worse than otherwise.' Again, our family system, which regards a child as equally related to his father and his mother, seems so natural that we experience a feeling of surprise on meeting with any other system. Yet we shall find, I think, reason for concluding that a man was first re- garded as merely related to his tribe ; then to his mother but not to his father ; then to his father and not to his mother ; and only at last to both father and mother. Even among the Eomans the family was originally based, not on marriage or on relationship, but on power ; ^ ' le lien seul,' says Ortolan, ' de la ' parente naturelle, de la parente de sang, n'est rien chez les ' Romains ; ' and a man's wife and children only formed a part of his family, not because they were his relatives, but because they were subject to his control ; so that a son who was emancipated — that is to say, made free — had no share in the inheritance, having ceased to belong to the family. In fact, the word, ' family ' is said to be derived from an Osque word, 'famul,' a slave. The fact is, we require a new word for a sort of relationship which we do not ourselves recognise. Savages who have the custom of descent through females do not recognise the family of the father as belonging to the same gens. In one sense they are not relations. They have no right of inheritance, nor does a very near connection (from our point of view) interpose any barrier to marriage. On the other hand, 1 Ortolan's Expl. Hist, des Instituts de TEmp. Justinien, vol. i. pp. 126, 128,180,416. 74 BELATIONSHIP AMONG SAVAGES of course no one would assert that they recognised no bond of union between father and son. They have, in fact, three distinct bonds of union : — 1. The tribe ; 2. The gens ; and 3. That actual connection which exists between father and son, even though they are not regarded as belonging to the same gens or family. We shall, however, be better able to understand this part of the question when we have considered the various phases which marriage presents ; for it is by no means of a uniform character, but takes several very distinct forms. In some cases nothing of the sort appears to exist at all ; in others it is essentially temporary, and exists only till the birth of the child, when both man and woman are free to mate themselves afresh. In others, the man buys the woman, who be- comes as much his property as his horse or his dog. The Romans had two forms of marriage. One was created by a religious ceremony, ^ confarreatio.' In this case the wife at once came under the ' manus ' of her husband, and her position was technically almost exactly that of a slave or a child. In the second form of marriage, that by sale, the wife was so closely assimi- lated to property that the full rights of possession could not be acquired until the usual period of prescrip- tion had passed. A title by prescription could only be acquired by a year's continued possession. Accordingly it became quite usual for the wife to return three days every year to her father's house, the result of which was that she never came under the ' manus ' of her husband. She then remained a member of her father's family, and BIFFEBENT KINDS OF MABBIAGE 75 the husband acquired no legal power over her. Her status in the two cases was therefore quite different. In Sumatra there were formerly three perfectly dis- tinct kinds of marriage : the ' Jugur,' in which the man purchased the woman ; the ' Ambel-anak,' in which the woman purchased the man ; and the ' Semando,' in which they joined on terms of equality. In the mode of marriage by Ambel-anak, says Marsden,^ ' the father of ' a virgin makes a choice of some young man for her ' husband, generally from an inferior family, which re- ' nounces all further right to, or interest in, him ; and ' he is taken into the house of his father-in-law, who ' kills a buffalo on the occasion, and receives twenty ^ dollars from his son's relations. After this, the buruk ' baik' nia (the good and bad of him) is invested in the ^ wife's family. If he murders or robs, they pay the ' bangun, or fine. If he is murdered, they receive the ^ bangun. They are liable to any debts he may con- ' tract in marriage ; those prior to it remaining with ' his parents. He lives in the family, in a state between ' that of a son and a debtor. He partakes as a son of ' what the house afibrds, but has no property in himself. * His rice plantation, the produce of his pepper garden, ' with everything that he can gain or earn, belongs ' to the family. He is liable to be divorced at their ' pleasure, and though he has children, must leave all, * and return naked as he came.' In the Jugur marriage the woman became the pro- perty of the man. ' The Semando ^ is a regular treaty between the * parties, on the footing of equality. The adat paid to 1 Marsden's Hist, of Sumatra, p. 262. ^ j^^-^^ p 26.3. IQ DIFFERENT KINDS OF MARRIAGE ' the girl's friends has usually been twelve dollars. ' The agreement stipulates that all effects, gains, or ' earnings are to be equally the property of both ; and, ' in case of divorce by mutual consent, the stock, debts, ' and credits are to be equally divided. If the man ' only insists on the divorce, he gives the woman her ' half of the effects, and loses the twelve dollars he has ' paid. If the woman only claims the divorce, she ' forfeits her right to the proportion of the effects, but ' is entitled to keep hertikar, bantal, and dandan (para- ' phernalia), and her relations are liable to pay back the ' twelve dollars ; but it is seldom demanded.' These three forms of marriage, co-existing in Sumatra, represent, as we shall see, three stages passed through successively by various other races. In Ceylon there were two kinds of marriage — the Deega marriage, and the Beena marriage. In the former the woman went to her husband's hut ; in the latter the man transferred himself to that of the woman. Moreover, according to Davy, marriages in Ceylon were provisional for the first fortnight, at the expiration of which period they were either annulled or confirmed.^ Again, in various parts of Africa we find two kinds of marriage existing together. In Guinea, besides the ordinary wives, a man often buys some slave whom he consecrates to his Bossum or God. The Bossum wife then becomes his in an exceptional sense. She is sacri- ficed at her husband's death, she ranks next to the head wife, and shares his religion. The Hassani3'eh Arabs have a very curious form of marriage, which may be called ' three-quarter ' mar- 1 Davy's Ceylon, p. 286. DIFFEBENT KINDS OF MARBIAGE 11 riage ; that is to say, the woman is legally married for three days out of four, remaining perfectly free for the fourth. In Australia the tribes are divided into clans, and a man may not marry a woman of the same clan. On the other hand, the men are regarded as by birth hus- bands of all the women whom they can legally marry. Besides this, however, a man has, or may have, an individual wife, generally acquired by capture. Among the Eomans, as shown by the Laws of the Twelve Tables, and as already mentioned, there were in reality two kinds of marriage, and, as Ortolan says, ^ il * faut se bien garder de confondre entre eux le mariage * (nuptise, justae nupti^, justum matrimonium) et la ' puissance maritale (manus).' ^ The latter required the performance of ceremonies, which were unnecessary for the former. Among the Karoks, marriage is strictly a matter of purchase : when a young man has paid the price of his bride, she becomes his property ; on the other hand, if he cannot provide the whole sum he is sometimes allowed to pay a portion, and become what is called 'half- ' married.^ In that case, instead of bringing her to his cabin, and making her his slave, he goes to hers and becomes subject to her, or rather to her father. Azara tells us that among the Guanas careful stipulations were made as to the duties and obligations the bride under- took with reference to her husband : how far she was bound to provide him food, whether she was to procure the necessary firewood, whether she was to be the sole wife, whether she was to be free to marry another man * Ortolan's Expl. Hist, des Inst, de I'Emp. Justinieu, p. 127. 78 BIFFEEENT KINDS OF MABBIAGE also, and in that case how much of her time the first husband wished to engage. In Japan, among the higher classes, it is said that the eldest son brings his bride to the paternal home ; but, on the other hand, the eldest daughter does the same, and retains her name, which is assumed by the bridegroom. Thus the wife of an eldest son joins her husband's family ; but, on the other hand, the husband of an eldest daughter enters into that of his wife. Among the Romans, though ' coemptio,' or purchase, was one of the recognised forms of marriage, it would seem that originally this merely gave possession, and a woman who belonged to any man by coemptio might otherwise be married to another.-^ Hence the eldest son of one family cannot marry the eldest daughter of an- other. As regards the younger children, if the husband's father provides the house, the wife takes her husband's name ; while, if the bride's father does so, the bridegroom assumes that of his wife.^ Among the Reddies ^ of Southern India a very singular custom prevails : — ' A young woman of sixteen ' or twenty years of age may be married to a boy of five ' or six years ! She, however, lives with some other ' adult male — perhaps a maternal uncle or cousin — ^biit ' is not allowed to form a connection with the father's ' relatives ; occasionally it may be the boy-husband's ^ father himself — that is, the woman's father-in-law ! ' Should there be children from these liaisons, thev are ' fathered on the boy-husband. When the boy grows ^ Fustal de Coulonges, La Cite Family, p, 428. Antique, p. 376. 3 gbortt. Trans. Ethn. Soc, New '■^ Morgan's System of Consan- Series, vol. vii. p. 194. guinity and Affinity of tlie Human POLYANDRY 79 ^ up, the wife is either old or past child-bearing, when ^ he in his turn takes up with some other " boy's " wife ^ in a manner precisely similar to his own, and procreates ' children for the boy -husband.' Polyandry, or the marriage of one woman to several men at once, is more common than is generally sup- posed, though much less so than polygamy, which is almost universally permitted among the lower races of men. One reason — though I do not say the only one — for this, is obvious when pointed out. Long after our children are weaned, milk remains an important and necessary part of their food. We supply this want with cow's milk ; but among people who have no domesti- cated animals this cannot, of course, be done, and con- sequently the children are not weaned until they are two, three, or even four years old, during all which period the husband and wife generally remain apart. Thus, in Fiji, ' the relatives of a woman take it as a public ' insult if any child should be born before the customary ' three or four years have elapsed, and they consider ^ themselves in duty bound to avenge it in an equally ^ public manner.' ^ It seems to us natural and proper that husband and wife should enjoy as much as possible the society of one another. But this view is by no means universal. On the contrary, among the Turkomans, according to Eraser, for six months or a year, or even sometimes two years, after a marriage, the husband was only allowed to visit his wife by stealth. ' After the wed- ' ding,' says Burnes, ' the bride returns to the house of ^ her parents, and passes a year in preparing the carpets ^ Seemann, A Mission to Fiji, p. 191. 80 SEPARATION OF HUSBAND AND WIFE ' and clothes, which are necessary for a Toorkmun tent ; ^ and on the anniversary of her elopement she is finally ^ transferred to the arms and house of her gallant lover.' ^ Among the Samoyedes the bride and bridegroom are kept apart for a month after their marriage,^ and Klemm states that the same is the case among the Cir- cassians until the first child is born. Martins mentions the existence of a similar custom among some of the Brazilian tribes.^ Among the Fijians, husbands and wives do not usually spend the night together, except as it were by stealth. It is quite contrary to Fijian ideas of delicacy that they should sleep under the same roof. A man spends his day with his family, but absents himself on the approach of night.^ In Chitta- gong (India), although, ' according to European ideas, ' the standard of morality among the Kyoungtha is low,^ yet husband and wife are on no account permitted to sleep together until seven days after marriage.^ Burckhardt^ states that in Arabia, after the wedding, if it can be called so, the bride returns to her mother's tent, but again runs away in the evening, and repeats these flights several times, till she finally returns to her tent. She does not go to live in her husband's tent for some months, perhaps not even till a full year, from the wedding-day. Among the Yotyaks, some weeks after the wedding the bride returns to her father's tent, and lives there for two or three months, sometimes even for ^ Burnes' Travels in Bokhara^ * Seemaiin'sMissiontoViti,p.l9]. vol. ii. p. 56. See also Vambery's ^ Lewin's Hill Tracts of Chitta- Travels in Central Asia, p. 323. gong, p. 51. 2 Pallas, vol. iii. p. 79. ^ Burckhardt's Notes, vol. ii. p. 3 Jour. Roy. Geog. Soc. vol. ii. 269, quoted in M'Lennan's Primitive p. 198. Marriage, p. 302. ABSENCE OF MARRIAGE CEREMONY 81 a year, during which time she dresses and behaves like a girl, and after which she returns to her husband ; making, however, even on the second occasion, a show of resistance.-^ Lafitau informs us that among the North American Indians the husband only visits the wife -as it were by stealth : — ' lis n'osent aller dans les cabanes particulieres, * oil habitent leurs epouses, que durant Tobscurite de la ' nuit ; . . . ce serait une action extraordinaire de s'y * presenter le jour.' ^ In Futa, one of the West African kingdoms, it is said that no husband is allowed to see his wife's face until he has been three years married. In Sparta, and in Crete, according to Xenophon and Strabo, it was the custom that married people for some time after the wedding only saw one another as it were clandestinely ; and a similar custom is said to have existed among the Lycians. So far as I am aware, no satisfactory explanation of this custom has yet been given. I shall, however, presently venture to suggest one. There are many cases in which savages have no such thing as any ceremony in marriage. ' I have said nothing, ' says Metz, ' about the marriage ceremonies of the Bada- ' gas (Hindostan), because they can scarcely be said to ' have any.' The Kurumbas, another tribe of the Neil- gherry Hills, ' have no marriage ceremony.' ^ According to Colonel Dalton.^ the Keriahs of Central India ' have ' no word for marriage in their own language, and the ^ Miiller's Des. de toutes les ^ Loc. cit. vol. i. p. 576. Nations de TEmp. de Kussie, part ii. ^ Trans. Ethn. Soc. vol. vii.p. 276. p. 71. ^ Ibid. vol. vi. p. 25. G 82 ABSENCE OF MARRIAGE CEREMONY, ' only ceremony used appears to be little more than a sort ' of public recognition of the fact.' It is very singular, he adds elsewhere, 'that of the many intelligent observers ' who have visited and written on Butan not one has ' been able to tell us that they have such an institution ' as a marriage ceremony.' The tie between man and woman seems to be very slight, and to be a mere matter of servitude. ' From my own observation,' he continues, *I believe the Butias to be utterly indifferent on the ' subject of the honour of their women.' ^ So also the Spanish missionaries found no word for marriage^ nor any marriage ceremony, among the Indians of Cali- fornia.^ Farther north, among the Kutchin Indians, ' there is no ceremony observed at marriage or birth.' ^ The same is the case among the Aleutians,^ and several other North Pacific tribes. The marital rite, says Schoolcraft, ' among our tribes' (i.e. the Redskins of the United States) 'is no- ' thing more than the personal consent of the parties, ' without requiring any concurrent act of a priesthood, a ' magistracy or witnesses ; the act is assumed by the par- ^ ties, without the necessity of any extraneous sanction.' ^ According to Brett, there is no marriage ceremony among the Arawaks of South America.^ Martins makes the same assertion with reference to the Brazilians generally,^ and it is also the case with some of the Australian tribes.^ ^ Des. Ethn. of Bengal, p. 97. ^ Indian Tribes, pp. 248, 132. ^ Bagaert, Smithsonian Report, *^ Guiana, p. 101. 1863, p. 368. Bancroft, vol. i. p. 565. ' Loc. cit. p. 51. ^ Smithsonian Eeport^ 1866, p. ^ Eyre's Discoveries, vol. ii. p. 326. 319. ^ Bancroft, vol. i. pp. 92, 277. ABSENCE OF MAEBIAGE GBBEMONY 83 There is, says Bruce, ' no such thmg as marriage in Abyssinia, unless that which is contracted by mutual consent, without other form, subsisting only till dis- solved by dissent of one or other, and to be renewed or repeated as often as it is agreeable to both parties, who, when they please, live together again as man and wife, after having been divorced, had children by others, or whether they have been married, or had children with others or not. I remember to have once been at Koscam in presence of the Iteghe (the queen), when, in the circle, there was a woman of great quality, and seven men who had all been her husbands, none of whom was the happy spouse at that time.' ^ Among the Bedouin Arabs there is a marriage ceremony in the case of a girl, but the re-marriage of a Avidow is not thought sufficiently important to deserve one. Speke says, ' there are no such things as marriages in ' Uganda.' ^ Of the Mandingoes (West Africa), Caillie^ says that husband and wife are not united by any ceremony ; and Hutton ^ makes the same statement as regards the Ash- antees. In Congo and xingola ^ ' they use no peculiar ' ceremonies in marriage, nor scarce trouble themselves ' for consent of friends.' Le Vaillant says that there are no marriage ceremonies among the Hottentots.; ^ and the Bushmen, according to Mr. Wood, had in their language no means of distinoTiishins^ an unmarried from a married girl.7 ^ Brace's Travels, vol. iv. p. 487. '" Astley's Ooll. of Voyages, vol. ~ Journal, p. 361. iii. pp. 221, 227. ^ Loc. cit. vol. i. p. 350. ^ Voyages, vol. ii. p. 58. ^ Klemm, Cultur d. Mensclien, '^ Natural History of Man, vol. i. vol. iii. p. 280. p. 269. G 2 84 MAEBIAGE CEBUMONIES In Northern Asia the Tunguses are said to have no marriage ceremony. Yet we must not assume that marriage is necessarily and always lightly regarded where it is unaccompanied by ceremonial. There is a great distinction between what may be called ^ lax ' and ' brittle ' marriages. In some countries the marriage tie may be broken with the greatest ease, and yet, as long as it lasts, is strictly respected ; while in other countries the very reverse is the case. Perhaps on the whole any marriage ceremony is better than none at all, but some races have practices at marriage which are extremely objectionable. Some, also, are very curious, and no doubt symbolical. At Bonabe, one of the Micronesian Pacific Islands, the wife is tattooed with the marks standing for the names of her husband's ancestors.^ One portion of the marriage ceremony among the Mundaris, one of the Bengal Hill tribes, is very suggestive. The bride walks in front of the bridegroom with a pitcher of water on her head, supported by one arm. The bridegroom walks behind, and through the pretty loophole thus formed he shoots an arrow. The girl walks on to where the arrow falls, picks it up with her foot, takes it into her hand, and re- spectfully returns it to her husband.^ In many parts of India, bride and bridegroom are marked with one another's blood, probably to signify the intimate union which has taken place between them. This is the custom, for instance, among the Birhors. Colonel Dal- ton believes this to be ' the origin of the custom now so ' Hale's United States Explor, ^ Dalton's Des. Etlin. of Bengal, Exped.: Ethnography, p. 76. p. 195. MARRIAGE CEREMONIES 85 * universal of marking with red lead.' ^ In other cases the idea symbolised is less obvious. Among some of the Hindoo tribes the bride and bridegroom are respectively married to trees in the first instance, and subsequently to one another. Thus a Kurmi bridegroom is married to a mango, his bride to a malwa tree.^ The idea under- lying this I take to be that they are thus devoted to the deities of the Mango and Malwa, and, having thus be- come respectively tabooed to other men and women, are, with the consent of the deities, espoused to one another. In ancient Russia, as part of the marriage ceremony, the father took a new whip, and after striking his daughter gently with it told her that he did so for the last time, and now presented the whip to the bride- groom, to whose power she then passed.^ Among the Canadian Indians, Carver^ says that, when the chief has pronounced the pair to be married, * the bridegroom turns round, and, bending his body, ' takes his wife on his back, in which manner he carries ' her, amidst the acclamations of the spectators, to his ' tent.' The Western tribes regard it as an important part of the marriage ceremony that the bride should be carried to her husband's dwelling.^ In Mexico also the husband took the bride on his back and carried her a short distance.^ Bruce, in Abyssinia, observed an identical custom. When the ceremony is over, he says, ' the bridegroom takes his lady on his shoulders, and ' carries her off to his house. If it be at a distance he 1 Dalton's Des. Etlin. of Bengal, * Travels, p. 374. pp. 220, 319. ■' Bancroft, vol. i. pp. 411, 703, 2 Ibid. p. 319. 730. 3 Meiners, Vergl. des alt. imd "^ Ihid. Yol.ii, ^. 261. neuer. Rnsslands, vol. ii. p. 167. 86 LIFTING TEE BEIBE ^ does the same thing, but only goes entirely round about ' the bride's house/ ^ In China, when the bridal procession reaches the bridegroom's house, the bride is carried into the house by a matron, and ' lifted over a pan of charcoal at the ' door.' 2 We shall presently see that these are no isolated cases, nor is the act of lifting the bride over the bride- groom's threshold an act without a meaning. I shall shortly mention many allied customs, to the importance and significance of which our attention has recently been called by M'Lennan, in his masterly work on ' Primitive Marriage.' I will now attempt to trace up the custom of mar- riage in its gradual development. There is strong evidence that the lowest races of men live, or did live, in a state of what may perhaps be called ^ Communal Marriage.' In many of the cases above given (pp. TO- TS) there can hardly be said to be any true marriage in our sense of the term, and many other instances might be given. In the Andaman Islands,^ Sir Edward Belcher states that the custom is for the man and woman to remain together until the child is weaned, when they separate as a matter of course, and each seeks a new partner ; but Mr. Man did not find this to be the custom among the families he visited. The Bushmen of South Africa are stated to be entirely with- out marriage. Among the Nairs (India), as Buchanan tells us, ' no one knows his father, and every man looks ' on his sister's children as his heirs.' They may be 1 Vol. vii. p. 67. 285. ^ Davis, The Chinese, vol. i. p. ^ Trans. Ethn. Soc. vol. v. p. 45. BELATI0N8HIPS INDUPENBENT OF MABEIAGE 87 said to have group marriages. A man may marry several women, and a woman may be the wife of several men. The Teehurs of Oude ' live together almost in- ^ discriminately in large communities, and even when ' two people are regarded as married the tie is but ' nominal.' ^ In China, communal marriage is stated to have pre- vailed down to the time of Fouhi,^ in Egypt to that of Menes, and in Greece to that of Cecrops. The Massa- getse,^ and the Auses,^ an Ethiopian tribe, had, according to Herodotus, no marriage — a statement which is con- firmed by Strabo as regards the former. Strabo and Solinus made the same statement as regards the Garamantes, another Ethiopian tribe, and Ammianus Marcellinus with reference to certain Arabian tribes. In California, according to Baegert,^ the sexes met without any formalities, and their vocabulary did not even contain the words ' to marry.' Garcilasso de la Yega asserts that among some of the Peruvian tribes, before the time of the Incas, men had no special wives.^ Mr. Hyde, Principal of the North Pacific Missionary Institute, to whom I am indebted for various valuable suggestions, writes me that among the Pacific Islanders there was an ' utter absence of what we mean by the ' family, the household, and the husband ; the only * thing possible was to keep distinct the line through 1 The People of India, by J. F. ^ Clio, vol. i. p. 216. Watson and J. W. Kaye, published * Melpomene, vol. iv. p. 180. by the Indian Government, vol. ii. ^ Loc. cit. p. 368. pi. 85. ^ Commentaries of the Incas, ^ Goguet, L'Origine des Lois, trans, by C. R. Markham, vol. ii. p. des Arts, et des Sciences, vol. iii. 443. p. 328. 88 AUSTRALIAN SYSTEM OF BELATIONSHIP ' the mother, and enumerate the successive generations * with the several putative fathers.' The original Hawaian word for ' to marry ' meant ' to try,' and the missionaries have been attempting to replace this by our word ' mare ' under a native form. Speaking of the natives of Queen Charlotte Island, Mr. Poole says,^ ' among these simple and primitive ' tribes the institution of marriage is altogether un- ' known.' The women appear to consider almost all the men of their own clan in the light of husbands. They are, on the contrary, very circumspect in their behaviour with other men. According to native legends, communal marriage existed in ancient times among the natives of Australia. Messrs. Fison and Howitt state that the South Aus- tralian tribes^ are divided into two classes or clans, Kumite and Kroki, the feminine equivalents of which are Kumitegor and Krokigor, and every Kumite is theoreti- cally the husband of every Krokigor, every Kroki being in the same way the husband of every Kumitegor. It is not asserted that marital rights are in full force at the present day, but they exist and are still acknow- ledged to a certaui extent. So again the Kamilaroi tribes, near Sydney, are divided into four great clans,^ in which the males are known as Ippai, Murri, Kubbi, and Kumbo ; the females, Ippata, Matha, Kapota, and Butha. ' I. Ipai may marry only Kubitha. ' II. Murri may marry only Butha. ^ Queen Charlotte Islands, p. ^ Pritchard's Nat. Hist, of Man, 312. ^ol. ii. p. 491. Kidley's Journ. 2 See Fison and Howitt, The Anthr. Inst. 1872, p. 263. Lang's Kamilaroi and Kiirnai, p. 50. Queensland, p. 383. AUSTRALIAN SYSTEM OF BELATIONSHIP 89 'III. Kubi may marry only Ipatha. ' ly. Kumbu may marry only Matha. ' Any attempt to infringe these rules would be ^ unanimously resisted, even to bloodshed ; but it ^ seems they never dream of attempting to transgress ' them.' Even if a man has captured a woman in war, he may not marry her if she belongs to a forbidden class. ' I. The children of Ipai by Kubitha are all Muri. ' II. The children of Muri are all Ippai and Ippata. ' III. The children of Kubi are all Kumbu and ' Butha. ' IV. The children of Kumbu are all Kubi and ' Kubitha.' But Mr. Lance first pointed out, and he has since been fully confirmed by subsequent writers, that in a certain sense every Ipai is regarded as married, not by any individual contract, but by organic law, to every Kubitha ; every Kubi to every Ipatha, and so on. If, for instance, a Kubi, says Mr. Lance, ' meet a stranger ' Ipatha, they address each other as spouse. A Kubi ' thus meeting an Ipatha, though she were of another ^ tribe, would treat her as his wife, and his right to do ^ so would be recognised by her tribe.' -^ The idea of marriage in Australia, say Messrs. Fison and Howitt,^ ' is something more than the marriage of group to ' group, loithin a tribe. It is an arrangement, extending ' across a continent, which divides many scattered — ' widely scattered — tribes into intermarrying classes, ^ and gives a man of one class marital rights over ^ Quoted by Fison and Howitt, ^ The Kamilaroi and Kurnai, p. ioc. cit. p. 53. 54, 90 AUSTRALIAN SYSTEM OF RELATION SHIF ^ women of another class in a tribe a thousand miles ' away, and speaking a language other than his own.' It would appear, however, that this right is now dying out, and is in most cases merely nominal. Mr. Bulmer, an English missionary in Australia, not understanding their customs, and wishing to make friends with the natives, allowed himself to be adopted with native ceremonials so as to become the brother of a young native of whom he had formed a high opinion. Next time he met the young man's wife he said to her : ' You know you are my sister now. I am your hus- ' band's brother.' ' Oh, no,' she said, laughing heartily ; ^ you are now my husband.' And he found he had most unintentionally married her, according to native ideas, and every other woman of her name ! The backwardness (until lately) of the Sandwich Islanders in their social relations is manifested in their language. This is shown from the following table extracted from a longer one, given by Mr. Morgan in a most interesting work on the Origin of the Classification System of Eelationship.-^ Hawaian English /Great grandfather Great great uncle Great grandmother T-r- • -n Great OTandaunt Kupuna signifies ^(jr.^aLher Granduncle Grandmother \Grandaunt. ^ Systems of Oonsangumity and Affinity. SOUTH SUA SYSTEM OF RELATIONSHIP 91 Hamaian Makua kana English /Fatlier Father's brother Father's brother-in-law 1 Mother's brother Mother's brother-in-law Grandfather's brother's son. Makua waheena /Mother Mother's sister -< Mother's sister-in-law Father's sister I Father's sister-in-law. Son Sister's son Brother's son Brother's son's son Kaikee kana = < Brother's daughter's son Sister's son's son Sister's daughter's son Mother's sister's son's son mother's brother's son's son. Hunona /Brother's son's wife I Brother's daughter's husband I Sister's son's wife vSister's daughter's husband. Waheena rWife Wife's sister Brother's wife Wife's brother's wife ^ Father's brother's son's wife Father's sister's son's wife Mother's sister's son's wife ^Mother's brother's son's wife. 92 SOUTH SEA SYSTEM OF EELATIONSHIP Hawaian English (Husband Husband's brother Sister's husband. Punalua = Wife's sister's husband (brother-in-law), Kaikoaka = Wife's brother. The key of this Hawaian or Sandwich Island ^ system is the idea conveyed in the word waheena (woman). Thus— Hamaicm English rWife __ , Wife's sister Waheena =\ , 1 Brother s wiie IWife's brother's wife. All these are equally related to each husband. Hence the word — Kaikee = Child, also signifies brother's wife's child; and no doubt the wife's sister's child, and the wife's brother's wife's child. So also, as the sister is wife to the brother-in-law (though not to her brother), and as the brother-in-law is husband to his brother's wife, he is consequently a father to his brother's children. Hence ^ Kaikee ' also means ' sister's son ' and ^ brother's son.' In fact ' Kaikee ' and * Waheena ' correspond to our words ' child ' and ' woman,' and there are apparently no words answering to " son,' ' daughter,' * wife,' or * husband.' That this does not arise from poverty of ^ Morgan, Proceedings of the American Association, 1868. SOUTH SEA SYSTEM OF RELATIONSHIP 93 language is evident, because the same system discri- minates between other relationships which we do not distinguish. Perhaps the contrast is most clearly shown in the terms for brother-in-law and sister-in-law. Thus, when a woman is speaking — Sister-in-law = husband's brother's wife = punalua. Sister-in-law = husband's sister = kaikoaka. But brother-in-law, whether sister's ) . . , _ , , T 11 Tj 1 ,1 h = kana, i.e. husband, husband or husband s brother J ' When, on the contrary, a man is speaking — Sister-in-law — wife's sister = waheena, i.e. wife. Sister-in-law = brother's wife = waheena, i.e. wife. And so — Brother-in-law = wife's brother = kaikoaka. Brother-in-law = wife's sister's husband = punalua. Thus a woman has husbands and sisters-in-law, but no brothers-in-law ; a man, on the contrary, has wives and brothers-in-law, but no sisters-in-law. The same idea runs through all other relationships : cousins, for instance, are called brothers and sisters. So again, while the Romans distinguished between the Father's brother = patruus, and the mother^s brother = avunculus ; Father's sister = amita, and the mother's sister = matertera ; the first two in Hawaian are makua kana, which also signifies father ; and the last two are makua waheena, which also means mother. 94 SOUTH SEA SYSTEM OF BELATIONSHIP In the next chapter I shall enter more at length into the subject of Relationships, but my object at present is to show that the idea of Marriage does not, in fact, enter into the Hawaian system. Uncleship, aunt- ship, cousinship, are ignored ; and we have only — Grandparents Parents Brothers and sisters Children, and Grandchildren. This division into generations was no matter of mere nomenclature ; but, lax as their ideas appear to us in many ways, Mr. Gill, the well-known missionary in the South Seas, tells me that in the Henry group marriage out of one's generation was strictly forbidden ; even when as a mere matter of age it might be quite suitable. Here, moreover, it is clear that the child is related to the group. It is not specially related either to its father or its mother, who stand in the same relation as uncles and aunts ; so that every child has several fathers and several mothers. There are, I think, reasons in the social habits of these islanders which go far to explain the persistence of this archaic nomenclature. From the mildness of the climate and the abundance of food, children soon become independent ; the prevalence of large houses, used as mere dormitories, and the curious prejudice against eating in common, must also have greatly tended to retard the development of special family feelings. Yet the system of nomenclature above mentioned did not TOBA SYSTEM OF BELATI0N8HIP 95 correspond with the actual state of society as found by- Captain Cook and other early voyagers. Among the Todas of the Neilgherry Hills, however, when a man marries a girl she becomes the wife of all his brothers as they successively reach manhood, and they also become the husbands of all her sisters as they become old enough to marry. In this case ' the first- ' born child is fathered upon the eldest brother, the ' next-born on the second, and so on throughout the ' series. Notwithstanding this unnatural system, the ' Todas, it must be confessed, exhibit much fondness ' and attachment towards their offspring, more so than ^ their practice of mixed intercourse would seem to ' foster.' 1 In the Tottiyars of India, also, we have a case in which it is recorded that ' brothers, uncles, and ' nephews hold their wives in common.' ^ So also, according to Nicolaus,^ the Galactophagi had commu- nal marriage, ' where they called all old men fathers, ^ young men sons, and those of equal age brothers.' ^ Among the Sioux and some other North American ' tribes the custom is to buy the eldest of the chiefs ' daughters ; then the others all belong to him, and are ' taken to wife at such times as the husband sees fit.' ^ Such social conditions as these tend to exj^lain the frequency of adoption among the lower races of men, and the fact that it is often considered to be as close a connection as real parentage. Among the Esquimaux, Captain Lyon tells us that ' this curious connection ^ Sliortt, Trans. Etlm. Soc, ^ Bachofen, Das Mutterrecht, N.S., vol. vii p. 240. p. 21. 2 Dubois' Descrip. of the People ^ Ethn. Joui-nal, 1869, p. 286. of India, p. 3. 96 FEEVALENCE OF AFOFTION * binds the parties as firmly together as the ties of ' blood ; and an adopted son, if senior to one by nature, ^ is the heir to all the family riches.' ^ In Central Africa, Denham states that ' the practice ^ of adopting children is very prevalent among the ' Felatahs, and, though they have sons and daughters of ' their own, the adopted child generally becomes heir Ho the whole property.'^ In Madagascar^ also Hhe ' adoption of other children, generally those of relatives,. ' is of frequent occurrence. These children are regarded ^ in every respect as if they were born of their adopted ' parents, and their real father and mother give up all * claim to them.' Ht is a custom,' says Mariner,^ ' in the Tonga * Islands, for women to be what they call mothers to ' children or grown-up young persons who are not their ^ own, for the purpose of providing them, or seeing that ' they are provided, with all the conveniences of life ; ' this is often done even if the natural mother be still living, in which case the adopted mother ' is regarded ' the same as the natural mother.' The same custom also existed in Samoa,^ the Marquesas, and other Pacific Islands.^ Among the Romans, also, adoption was an important feature, and was effected by the symbol of a mock birth, without which it was not regarded as com- plete. This custom seems to have continued down to the time of Nerva, who, in adopting Trajan, transferred 1 Journal, p. 353. See 365. * Mariner's Tonga Islands, vol. ii. ~ Denliam's Travels in Africa, p. 98. vol. iv. p. 131. 5 Nineteen Years in Polynesia, ^ Sibree's Madagascar and its p . 179. People, p. 197. ^ Gerland,Waitz' Anthropologic, vol. vi. p. 216. THE MILK-TIE 97 the ceremony from the marriage-bed to the temple of Jupiter.^ Diodorus ^ gives a very curious account of the same custom as it existed among the Greeks, men- tioning that Juno adopted Hercules by going through a ceremony of mock birth. To this day, in some Con- tinental codes, adoption gives the right of inheritance.^ In other cases the symbol of adoption represented not the birth, but the milk, tie. Thus, in Circassia, the woman offered her breast to the person she was adopting. In Abyssinia, Parkyn tells us that ' if a man ' wishes to be adopted as the son of one of superior ' station or influence, he takes his hand, and, sucking ' one of his fingers, declares himself to be his " child by ' " adoption," and his new father is bound to assist .him ^ as far as he can.' ^ Among some races marriage between foster children is strictly forbidden. The same idea of adoption underlies, perhaps, the curious Esquimaux habit of licking anything which is presented to them, apparently in token of ownership.^ Dieffenbach^ also mentions the practice of licking a present in New Zealand ; here, however, it is the donor who does so. In the Tonga Islands, Captaui Cook tells us that the natives ^ have a singular custom of putting ' everything you give them to their heads, by way of ^thanks, as we conjectured.'^ Labillardiere observed the same practice in Tasmania.® 1 Miiller,DasMutterrecht,p.254. vol. i. p. 34. 2 IV. 39. See Notes. ^ New Zealand, vol. ii. p. 104. 2 Maine, Early Law and Custom, ^ Voyage towards the South p. 96. Pole, vol. i. p. 221. * Parkyn's Abyssinia, p. 198. ^ Gerland, Waitz' Anthropolo- ^ Franklin's Journeys, 1819-22, gie, vol. vi. p. 812. H 98 OBIGINAL OB GOMMUNAL MABBIAGE Assuming, then, that the communal marriage system shown in the preceding pages to prevail, or have pre- vailed, so widely among races in a low stage of civilisa- tion, represents the primitive and earliest social con- dition of man, we now come to consider the various ways in which it may have been broken up and replaced by individual marriage. Montesquieu lays it down almost as an axiom, that * r obligation naturelle qu'a le pere de nourrir ses ' enfants a fait etablir le mariage, qui declare celui qui ' doit remplir cette obligation.' ^ Elsewhere he states that ' il est arriv6 dans tous les pays et dans tons les ^ temps que la religion s'est melee des manages.' ^ How far these assertions are from the truth will be conclu- sively shown in the following pages. Bachofen,^ M'Lennan,^ and Morgan, the most recent authors who have studied this subject, all agree that the primitive condition of man, socially, was one in which marriage did not exist, ^ or, as we may perhaps for convenience call it, of communal marriage, where all the men and women in a small community were regarded as equally married to one another. Bachofen considers that after a while the women, shocked and scandalised by such a state of things, revolted against it, and established a system of marriage with female supremacy, the husband being subject to the wife, property and descent being considered to go in the female line, and women enjoying the principal share of political power. The first period he calls that ^ Esprit des Lois, vol.ii. p. 186. * Primitive Marriage." * Loc. cit. p. 299. 5 Ibid, xyiii. xix. ' Das Mutterrecht. ORIGIN OF IIABBTAGE 99 of ^ Hetairism/ the second of ' Mutterrecht,' or ' mother- ' right.' In the third stage he considers that the ethereal influence of the father prevailed over the more material idea of motherhood. Men claimed pre-eminence, pro- perty and descent were traced in the male line, sun worship superseded moon worship, and many other changes in social organisation took place — mainly because it came to be recognised that the creative influence of the father was more important than the material tie of motherhood. The father, in fact, was the author of life, the mother a mere nurse. Thus he regards the first stage as lawless, the second as material, the third as spiritual. I believe, however, that communities in which women have exercised the supreme power are rare and exceptional, if indeed they ever existed at all. We do not find in history, as a matter of fact, that women do assert their rights, and savage women would, I think, be peculiarly unlikely to uphold their dignity in the manner supposed. On the contrary, among the lowest races of men, as, for instance, in Australia, the position of the women is one of complete subjection ; and it seems to me perfectly clear that the idea of marriao-e is founded on the riohts, not of the woman, but of the man, being an illustration of the good old plan, That lie should take who has the power, And he should keep who can. Among low races the wife is indeed literally the property of her husband. As Petruchio says of Catherine — LofC: ^^ 100 RELATIONSHIP AMONG THE ROMANS I will be master of what is mine own. Slie is my goods, my chattels ; she is my house^ My household stuff, my field, my barn, My horse, my ox, my ass, my anything. So thoroughly is this the case, that a Roman'^s ^ family ' originally, and indeed throughout classical times, meant his slaves, and the children only formed part of the family because they were his slaves ; so that if a father freed his son, the latter ceased to be one of the family, and had no part in the inheritance. * The mere tie of blood relationship,' says Ortolan^ ^ was of no account among the Romans The ' most general expression and the most comprehensive ' term indicating relationship in Roman Law is cognatio ^ — the cognation, that is to say, the tie between persons ' who are united by the same blood, or those reputed by 'the law as such (cognati ; quasi una communiter nati),. ' But cognation alone, whether it proceeds from legal ' marriage or any other union, does not place the indi- ' vidual within the family, nor does it give any right of ' family.' ^ Even at the present day, in some parts of Africa, a man's property goes, not to his children, as such, but to his slaves. Among the West African tribes of the Gold Coast, under ordinary circumstances the wife was the slave of her husband, purchased of her father by the dowry, but if ' the wife be a woman of free status, who contracts ' a free union with her husband, not only are her children ' not his slaves, but neither she nor they become mem- ' bers of his family.' ^ ^ Ortolan's History of Rom an Law, ^ Foreign Office Despatch, Aug.. tr, by Pricha,rd and Nasmith, p. 129. 21, 1874. WRESTLING FOB WIVES 101 The fact that the wife is regarded literally as the property of the husband explains those cases which seem to us so remarkable, in which great laxity of conduct before, is combined with the utmost strictness after, marriage. Hence, also, the custom, so prevalent among the lower races of men, that on the death of the elder brother the wives belong to the second. This complete subjection of the woman in marriage also explains those cases in which women of rank were considered too great to marry. Livingstone distinctly states this in the case of Mamochisane, daughter of Sebituane, chief of the Bechuanas. Sebituane ' could ^ not look upon the husband except as the woman's ^ lord, so he told her all the men were hers, she might ^ take any one, but ought to keep none.' ^ Hearne tells us, that among the Hudson's Bay Indians ' it has ever been the custom for the men to ' wrestle for any woman to whom they are attached ; ' and, of course, the strongest party always carries off ^ the prize. A weak man, unless he be a good hunter ^ and well-beloved, is seldom permitted to keep a wife ^ that a stronger man thinks worth his notice. . . . ^ This custom prevails throughout all their tribes, and ^causes a great spirit of emulation among their youth, ^ who are upon all occasions, from their childhood, trying ' their strength and skill in wrestling.' ^ Franklin also says that the Copper Indians hold women in the same low estimation as the Chipewyans do, 'looking upon ' them as a kind of property, which the stronger may ^ Travels in South Africa, p. Tuckey's Exp. to the River Zaire, 179. See also Burton's Dahomey, p. 140. vol. i. pp. 107, 366; vol. ii. p. 72. 2 Hearne, p. 104. 102 M'LENNAN'S VIEWS ' take from the weaker ; ' ^ and Richardson^ 'more than ' once saw a stronger man assert his right to take the ' wife of a weaker countryman. Anyone may challenge ' another to wrestle, and, if he overcomes, may carry off ' the wife as the prize.' Yet the women never dream of protesting against this, which, indeed, seems to them perfectly natural. The theory, therefore, of Dr. Bacho- fen, and the sequence of social customs suggested by him, although supported with much learning, cannot, I think, be regarded as correct.^ M'Lennan, like Bachofen and Morgan, starts with a stage of Hetairism or communal marriage. The next stage was, in his opinion, that form of polyandry in which brothers had their wives in common ; afterwards came that of the levirate, i.e. the system under which, when an elder brother died, his second brother married the widow, and so on with the others in succession. Thence he considers that some tribes branched off into endogamy, others into exogamy ; ^ that is to say, some forbade marriage out of, others within, the tribe. If either of these two systems was older than the other, he considers that exogamy must have been the more ancient. Exogamy was based on infanticide,^ and led to the practice of marriage by capture.^ In a further stage the idea of female descent, pro- ducing as it would a division in the tribe, obviated the necessity of capture as a reality and reduced it to a symbol, 1 Journey to tlie Shores of the Tracts of Chittagong, pp. 47, 77, 80, Polar Seas, vol. viii. p. 43. 93, 98, 101. ^ Richardson's Boat Journey, * I^oc. cit. p. 145. vol. ii. p. 24. * Loc, cit. p. 138. ^ See, for instance, Lewin's Hill ^ Loc. dt. p. 140. THE TEUE EXPLANATION 103 In support of this view, Mr. M'Lennan has certainly- brought forward many striking facts ; but, while ad- mitting that it probably represents the succession of events in some cases, I cannot but think that these are exceptional. Exogamy is in fact often associated with polygamy, which under Mr. M'Lennan's system could not well be. Fully admitting the prevalence of infanticide among savages, it will, I think, be found that among the lowest races boys were killed as frequently as girls. Eyre expressly states that this was the case in Australia.^ In fact, the distinction between the sexes implies an amount of forethought and prudence which the lower races of men do not possess. For reasons to be given in the next chapter, I be- lieve that communal marriage was gradually superseded by individual marriage founded on capture, and that this led firstly to exogamy and then to female infanti- cide ; thus reversing M'Lennan's order of sequence. Endogamy and regulated polyandry, though frequent, I regard as exceptional, and as not entering into the normal progress of development. ^ Discoveries, &c., vol. ii. p. 324. 104 CHAPTER lY. THE ORIGIN OF MARRIAGE. THE evidence given in tlie preceding chapter, and which might have been much increased, seems to me to prove, and indeed it is now admitted by most of those who have studied the subject, that there was a time when individual marriage did not exist, and when mankind lived in a state of what I have suggested we might call communal marriage. The curious Australian marriage laws, under which marriage between members of the same clan is strictly forbidden, but on the other hand every man is legally and technically the husband of every woman belonging to some one or more other clans, have been already mentioned on p. 90. It has been supposed by some that an extension of these restrictions might gradually lead up to individual marriage, but this is not so, because a development of the Austrahan rules would always result in the marriage, not of individuals, but of classes — however much the class might be reduced by subdivision, the wives would remain in common within the gens. Such arrangements may be, and in some tribes no doubt are, the nearest approach to what we call marriage, but a husband in this sense is very different from a husband in ours. He has not the exclusive right to a particular woman, which is in our idea the essence of th:e] origin of mabbiage 105 matrimony, and the existence of which is just what we have to account for.-^ Speaking generally, however, we find in Australia, side by side with these class mar- riages, the presence also of individual marriage. Though the same word has been generally used in both cases, it is evident that the relationship is really very different. ' In the following pages,' say Messrs. Fison and Howitt,^ ' the words marriage, husband, wife, and indeed ' all the terms of kinship, are used in a certam accom- ^modated sense. Husband and wife are not neces- ^ sarily man and wife according to our ideas. " My * " husband," for instance, among tribes such as the Aus- * tralian, does not necessarily single out any one man in * particular. A woman may apply it to any one of a ' group of tribal brothers who have the right of taking ^ her to wife.' The question then is. How did individual marriage take its origin ? The theory I have ventured to suggest as regards the former question is, that originally no man could appropriate any woman of his own tribe exclusively to himself, nor could any woman dedicate herself to one man, without infringing tribal rights ; but that, on the other hand, if a man captured a woman belonging to another tribe he thereby acquired an individual and peculiar right to her, and she became his exclusively, no one else having any claim or property in her. Thus, then, the women in such a community would fall into two classes : The one, subject no doubt to the disad- vantage of being aliens, and so to say slaves, but yet ^ See Lubbock on Australian ^ The Kamilaroi and Kurnai, Marriage Customs, J. Anthr. Inst. p. 28. 1885. 106 PROBABLE OBIGIN OF MABBIAGE enjoying the protection, and in many cases having secured the aiFection, of one man. The other, nominally no doubt free, but in the first place subject to the attentions of all their tribesmen — attentions no doubt often very unwelcome, but yet which could not be rejected without giving bitter offence ; and in the second without any claim on any one specially for food, shelter, and protection. It seems to me that under such circumstances many women belonging to the latter class would long to exchange their nominal freedom, and hazardous privi- leges, for the comparative peace and security of the former. On the other hand, many men would desire to appropriate exclusively to themselves some woman of their own tribe by whom they were specially attracted. Hence would naturally arise a desire on the part of many to extend the right of capture, which originally had reference only to women of a different tribe, and to apply it to all those belonging to their own. As a matter of fact, we find in Australia, side by side with the division of the tribes into classes or ' gentes ' and the custom that all the men are regarded as possessing marital rights over all the women of some one, or more, of the other classes, the existence also of individual marriage ; one man and one woman especially connected together as in more civilised communities. The words husband and wife have been usually applied to both of these cases. At the same time, whether we apply the same word in both relationships or not, we must not lose sight of the fact that the two are very different,^ ^ It would be convenient, I think, to say, for instance, tliat a woman to use some sucli term as the New was ' noa ' to a particular gens or Zealand ^noa,' in the former case, and gentes, and wife to a particular man. TWO FORMS OF MABBIAGF IN AUSTRALIA 107 and it is this latter or true marriage to which my sug- gestion refers. It must not, however, be considered that the right to take any woman belonging to another class was originally a concession. The true process was in the reverse order, and the forbidding to take a woman of the man's own class must be regarded as a restriction. There are not wanting traditions of a time when this restriction did not exist. But, however this may be, we have complete and conclusive evidence that in large portions of Australia every man had the privilege of a husband over every woman not belonging to his own gens ; sharing of course those privileges with every other man belonging to the same class or gens as himself. But although we may call this ' marriage ' — and it is a right which in old times was, and to a certain extent still is, recognised as perfectly legal and respectable — it does not help us to the origin of individual marriage. In addition to the 1,000 miles of wives so forcibly described by Messrs. Fison and Howitt, the Australian had his own individual wife. How does he acquire a special right to her ? I have argued that this was originally by right of capture ; Messrs. Fison and Howitt deny this. But let us see what they say them- selves a few pages further on. In describing the habits of the Kiirnai they come to his marriage. How does he procure his wife ? ' The young Kiirnai,' says Mr. Fison,^ ' could, as a rule, acquire a wife in one way 'only. He must run away with her. ... It is no ' use his asking for a wife excepting under the most ^ Kamilaroi and Kurnai, p. 200. 108 AUSTRALIAN MABBIAGE CUSTOMS ' exceptional circumstances, for he could only acquire ^ one in the usual manner, and that was by running off * with her.' As regards the Geawe-gal tribe, they say, ' In the ' case of female captives, they belonged to their captors, 4f of a class from which wives might be legally taken ' by them. If of a forbidden class, then I think that * the captor might make an exchange with some one of ' the proper class who had a woman at his disposal. * In the Wonghi tribe, whose territory was situated on ' the north side of the Lachlan River, for about eighty ' miles above Whealbah, a woman was the property of ' her captor when she was not of a tribe forbidden to ' him,' i.e. if she did not belong to a gens with which it was unlawful for him to intermarry. Speaking of the Turras, another Australian tribe, they say, ' There is individual marriage. Consent of * the woman's parents is necessary before marriage ; if ' this is refused, the pair occasionally elope. Wives are \ also obtained by gift, exchange, or capture. A female * captive belonged to the captor.' Again, the Kamilaroi have ' the right to the female captive, controlled by the ' exogamous rule of marriage.' Indeed, speaking gener- ally, they observe, ' that marriage is brought about ' throughout Australia by capture is quite certain.' It is obvious, indeed, that even under a communal marriage, a warrior who had captured a beautiful girl in some marauding expedition would claim a peculiar right to her, and, when possible, would set custom at defiance. We have already seen that there are other cases of the existence of marriage under two forms side by side in one country ; and there is, therefore, no real TEE TRUE EXPLANATION 10^ difficulty in assuming the co-existence of communal and individual marriage. It is true that under a com- munal marriage system no man could appropriate a girl entirely to himself without infringing the rights of the whole tribe. Such an act would naturally be looked on with jealousy, and only regarded as justifiable under peculiar circumstances. A war- captive, however, was in a peculiar position : the tribe had no right to her ; her capturer might have killed her if he chose ; if he preferred to keep her alive he was at liberty to do so ; he did as he liked, and the tribe was no sufferer. On the other hand, if a marriage system had already existed, it is unlikely that the first wives would have suffered a mere captive to obtain the same station as them- selves.-^ M'Lennan,^ indeed, says that ' it is impossible to ' believe that the mere lawlessness of savages should be ' consecrated into a legal symbol, or to assign a reason ' — could this be believed — why a similar symbol should ' not appear in transferences of other kmds of property. '^ The symbol of capture, however, was not one of law- lessness, but, on the other hand, of — according to the ideas of the times — lawful possession. It did not refer to those from whom the captive was taken, but was- intended to bar the rights of the tribes into which she was introduced. Individual marriage was, in fact, an infringement of communal rights ; the man retaining to himself, or the man and woman mutually appropriat- ing to each other, that which previously belonged to- ^ I am glad to find that Mr. H. accept my sug-gestious as to com- Spencer, in his Principles of Socio- munal marriage, or as to the rights- logy, p. 650 et seq., endorses this of men within the tribe, view, though he does not altogether ^ Loc. cit. p. 44. 110 OEIGIN OF MABBIAGE BY GAPTUBE the whole tribe. Thus, among the Andamaners, any woman who attempted to resist the marital privileges claimed by any member of the tribe was liable to severe punishment.^ ISTor is it, I think, difficult to understand why the symbol of capture does not appear in transferences of other kinds of property. Every generation requires fresh wives ; the actual capture, or at any rate the symbol, needed therefore repetition. This, however, does not apply to land ; when once the idea of landed property arose, the same land descended from owner to owner. In other kinds of property, again, there is an important, though different kind of, distinction. A man made his own bow and arrows, his own hut, his own arms ; hence the necessity of capture did not exist, and the symbol would not arise. McLennan supposed that savages were driven by female infanticide, and the consequent absence or pau- city of women, into exogamy and marriage by capture. He considered that the ' practice of capturing women ' for wives could not have become systematic unless it * were developed and sustained by some rule of law or ^ custom,' and ' that the rule of law or custom which ^ had this effect was exogamy.' ^ I shall presently give my reasons for rejecting this explanation. He also considers that marriage by capture followed, and arose from, that remarkable custom of marrying always out of the tribe, for which he has pro- posed the appropriate name of exogamy. On the con- trary, I believe that exogamy arose from marriage by 1 Trans. Ethn. Soc, N.S., vol. ii. ^ j ^^^^ ^^^^ ^^.^^ ^^^ article in p. 35. the Fortnightly for June 1877. OBIGIN OF MABEIAGE BY CABTUBB 111 capture, not marriage by capture froui exogamy ; that capture, and capture almost alone, could originally give a man the right to monopolise a woman, to the exclu- sion of his fellow- clansmen ; and that hence, even after all necessity for actual capture had long ceased, the symbol remained ; capture having, by long habit, come to be received as a necessary preliminary to marriage. That marriage by capture has not arisen from female modesty is, I think, evident not only because we have no reason to suppose that such a feeling prevails spe- cially among the lower races of man, but also, firstly, because it cannot explain the mock resistance of the relatives ; and, secondly, because the very question to be solved is why it became so generally the custom to win the female not by persuasion but by force, M'Lennan's view throws no light on the remarkable ceremonies of expiation for marriage, to which I shall presently call attention. I will, however, first proceed to show how widely ' capture,' either actual or symboli- cal, enters into the idea of marriage. M'Lennan was, I believe, the first to appreciate its importance. I have taken some of the following instances from his valuable work, with, however, much additional evidence. It requires, no doubt, strong evidence, which, how- ever, exists in abundance, to satisfy us that the origin of marriage was independent of all sacred and social considerations ; that it had nothing to do with mutual afi'ection or sympathy ; that it was invalidated by any appearance of consent ; and that it was symbolised, not by any demonstration of warm aficction on the one side and tender devotion on the other, but by brutal violence and unwilling submission. 112 MABBIAOE BY CAPTUBE OBIGINALLY A BEALITY Yet, as already mentioned, the evidence is over- whelming. So completely, for instance, did the Caribs supply themselves with wives from the neighbouring races, and so little communication did they hold with them, that the men and women actually spoke different languages. So, again, in Australia the men, says Old- field, ' are in excess of the other sex, and, consequently, ' many men of every tribe are unprovided with that ' especial necessary to their comfortable subsistence, a ' wife — who is a slave in the strictest sense of the word, ' being a beast of burden, a provider of food, and a ' ready object on which to vent those passions that the ' men do not dare to vent on each other. Hence, for ' those coveting such a luxury, arises the necessity of ^ stealing the women of some other tribe ; and, in their * expeditions to effect so laudable a design, they will * cheerfully undergo privations and dangers equal to ' those they incur when in search of blood-revenge. * When, on such an errand, they discover an unprotected ' female, their proceedings are not of the most gentle ' nature. Stunning her by a blow from the dowak (to ' make her love them, perhaps), they drag her by the ' hair to the nearest thicket to await her recovery. ' When she comes to her senses they force her to ^ accompany them ; and, as at worst it is but the ex- ' change of one brutal lord for another, she generally ^ enters into the spirit of the affair, and takes as much ' pains to escape as though it were a matter of her own ' free choice.' ^ Collins thus describes the manner in which the na- tives about Sydney used to procure wives : — ' The poor 1 Trans. Ethn. Soc, vol. iii. p. 250. SUBSEQUENTLY A FORM 113 * wretcli is stolen upon in the absence of her protectors. ^ Being first stupefied with blows, inflicted with clubs ' or wooden swords, on the head, back, and shoulders, ^ every one of which is followed by a stream of blood, ^ she is then dragged through the woods by one arm, * with a perseverance and violence that it might be sup- ^ posed would displace it from its socket. This outrage ' is not resented by the relations of the female, who * only retaliate by a similar outrage when they find an ' opportunity.' ^ Marriage by capture is the third form of marriage / specially recognised by ancient Hindoo law.^ In Bali also,^ one of the islands between Java and New Guinea, it is stated to be the practice that girls ' are stolen away by their brutal lovers, who sometimes ' surprise them alone, or overpower them by the way, ' and carry them off w^ith dishevelled hair and tattered * garments to the woods. When brought back from ^ thence, and reconciliation is effected with enraged ' friends, the poor female becomes the slave of her ^ rough lover, by a certain compensation -price being paid ' to her relatives.' So deeply rooted is the feeling of a connection between force and marriage that we find the former used as a form long after all necessity for it had ceased ; and it is very interesting to trace, as Mr. M'Lennan has done, the gradual stages through which a stern reality softens down into a mere symbol. It is easy to see that if we assume the case of a 1 Collins's Englisli Colony in Aryas, p. 127. New South Wales, p. 362. ^ Notices of the Indian Archi- ^ Biihler's Sacred Books of the pelago, p. 90. 114 BINDOSTAN country in which there are four neighbouring tribes^ who have the custom of exogamy, and who trace pedi- grees through the mother, and not through the father a custom which, as we shall presently find, is so common that it may be said to be the usual one among the lower races — after a certain time the result would be that each tribe would consist of four septs or clans, representmg the four original tribes, and hence we should find communities in which each tribe is divided into clans, and a man must always marry a woman of a different clan. But as communities became larger and more civilised the actual ^ capture ' would become inconvenient, and at last impossible. Gradually, therefore, it came to be more and more a mock ceremony, forming, however, a necessary part of the marriage ceremony. Of this many cases might be given. Speaking of the Khonds of Orissa, Major-General Campbell says that on one occasion he ' heard loud cries ^ proceeding from a village close at hand. Fearing some ' quarrel, I rode to the spot, and there I saw a man ' bearing away upon his back something enveloped in ^ an ample covering of scarlet cloth ; he was surrounded ' by twenty or thirty young fellows, and by them pro- ' tected from the desperate attacks made upon him by ' a party of young women. On seeking an explanation ' of this novel scene, I was told that the man had just ' been married, and his precious burden was his bloom- * ing bride, whom he was conveying to his own village. ' Her youthful friends (as it appears is the custom)' * were seeking to regain possession of her, and hurled ^ stones and bamboos at the head of the devoted. CENTRAL INDIA 115 ^ bridegroom, until he reached the confines of his own ' village.' ^ Dalton mentions that among the Kols of Central India, when the price of a girl has been arranged, ' the bridegroom and a large party of his friends ' of both sexes enter wdth much singing and dancing, ' and sham fighting in the village of the bride, where ' they meet the bride's party, and are hospitably enter- ' tained.' ^ Sir W. Elliot also mentions that not only amongst the Khonds, but also in ' several other tribes of Central ' India, the bridegroom seizes his bride by force, either ' affected or real ; ' ^ and the same was customary among the Badagas of the Neilgherry Hills, the Mun- dahs, Hos, Garos, Oraons, Ghonds, and other Hill tribes.^ Among the Garos a young man and woman who wish to marry take some provisions and retire to the Hills for a few days. The girl goes first, and the lover follows after, well knowing, of course, where she will be found. In a few days they return to the village, when the marriage is publicly announced and solemnised, a mock fight taking place, though in this case the pre- tended reluctance is on the part of the bridegroom.^ In this tribe the girls propose to the men, as is also said to be the case among the Bhiuyas.^ In parts of the Punjab,^ 'when the bridegroom's 1 Quoted in M'Lennan's Primi- * Metz, The Tribes of the Neil- tive Marriage, p. 28. gherries, p. 74. See also Lewin's 2 Trans. Ethn. Soc. vol. vi. p. 24. Hill Tracts of Chittagong, pp. 36, 80. See also p. 27 ; the Trihes of India, ^ Dalton's Des, Ethn. of Bengal, vol. i. p. 15 ; and Dalton's Des. p. 64. Ethnology of Bengal, pp. 64, 86, ^ Loe. cit. p. 142. 193, 252, 278, 319. ' Tupper's Punjab Customary 3 Trans. Ethn. Soc. 1869, p. 325. Law. i2 116 MALAY PENINSULA— KALMUCKS * party goes to bring the bride from her father's house, ' they are met by a party of the bride's friends and rela- ^ tions, who stop the path. Hereupon a sham fight of a ' very rough description ensues, in which the bridegroom * and his friends, before they are allowed to pass, are well * drubbed with good thick switches.' M. Bourien ^ thus describes the marriage ceremony among the wild tribes of the Malay Peninsula : — ' When ^ all are assembled, and all ready, the bride and bride- ^ groom are led by one of the old men of the tribe * towards a circle more or less great, according to the ' presumed strength of the intended pair ; the girl ' runs round first, and the young man pursues a short * distance behind ; if he succeeds in reaching her and ' retaining her, she becomes his wife ; if not, he loses * all claim to her. At other times, a larger field is ' appointed for the trial, and they pursue one another * in the forest. The race, according to the words of the ^ chronicle, " is not to the swift, nor the battle to the ' '' strong," but to the young man who has had the good ^ fortune to please the intended bride.' Among the Kalmucks, De Hell tells us that, after the price of the girl has been duly agreed on, when the bridegroom comes with his friends to carry oiF his bride, ' a sham resistance is always made by the people * of her camp, in spite of which she fails not to be borne ^ away on a richly caparisoned horse, with loud shouts ^ and feu de joie.' ^ Dr. Clarke^ gives a romantic account of the cere- 1 Trans. Ethn. Soc. 1865, p. 81. ^ Travels, vol. i. p. 332. See ® Steppes of the Caspian, p. 259. also Vambery's Travels in Central Quoted in M'Lennan's Primitive Asia, p. 323. Burnes' Travels in Marriage, p. 30. Bokhara, pp. 11, 56. TUNGU8E8—KAMGHADALES 1 L 7 mony. ' The girl/ he says, ' is first mounted, who ^ rides off at full speed. Her lover pursues ; if he ' overtakes her, she becomes his wife, and the marriage ^ is consummated on the spot ; after this she returns ^ with him to his tent. But it sometimes happens that ' the woman does not wish to marry the person by ' whom she is pursued ; in this case, she will not suffer ' him to overtake her. We were assured that no ui- * stance occurs of a Kalmuck girl being thus caught, * unless she have a partiality to the pursuer. If she ^ dislikes him, she rides, to use the language of English ' sportsmen, " neck or nought," until she has completely * effected her escape, or until her pursuer's horse be- * comes exhausted, leaving her at liberty to return, ' and to be afterwards chased by some more favoured * admirer.' ' Among the Tunguses and Kamchadales,' says Ernan,^ ' a matrimonial engagement is not definitely ' arranged and concluded until the suitor has got the * better of his beloved by force, and has torn her ^ clothes.' Attacks on women are not allowed to be avenged by blood unless they take place within the yourt or house. The man is not regarded as to blame if the woman ' has ventured to leave her natural place, * the sacred and protecting hearth.' Pallas observes that in his time 'marriage by capture prevailed also ' among the Samoyedes.' ^ At present the custom is for the bridegroom to tap the father and the mother of ^ Travels in Siberia, vol. ii. p. "-^ Vol. iv. p. 97. See also Ast- 442. See also Karnes' History of ley's Collections of Voyages, vol. iv. Man, vol. ii. p. 58. p. 575. 118 MONGOLS— K0BEAN8— ESQUIMAUX the bride on the shoulder with, a small stick — the last trace of an ancient reality.-^ Among the Mongols,'^ when a marriage is arranged, the girl 'flies to some relations to hide herself. The * bridegroom coming to demand his wife, the father-in- ^ law says, " My daughter is yours ; go, take her wher- * " ever you can find her." Having thus obtained his ^ warrant, he, with his friends, runs about searching, ^ and, having found her, seizes her as his property, and ^ carries her home as it were by force.' Marriage by capture, indeed, prevails throughout Siberia. In Kam- skatka, says Mliller, ' attraper une fille est leur ex- * pression pour dire marier.' ^ ' In the Korea, when a man marries, he mounts on * horseback, attended by his friends, and, having ridden ' about the town, stops at the bride's door, where he is ^ received by her relations, who then carry her to his ^ house, and the ceremony is complete.' * Traces of the custom also occur in Japan.^ Among the Esquimaux of Cape York (Smith Sound), according to Dr. Hayes,^ ' there is no marriage cere- * mony further than that the boy is required to carry * off his bride by main force ; for, even among these ^ blubber- eating people, the woman only saves her ^ modesty by a sham resistance, although she knows * years beforehand that her destiny is sealed, and that * she is to become the wife of the man from whose ^ Seebohm, Siberia in Europe, See also pt. i. p. 170 ; pt. iii. pp. p. 74. 38, 71. 2 Astley, vol. iv. p. 77. * Ibid. p. 342. ^ Des. de toutes les Nations de ^ Le Japon lUustre, vol. ii. p. I'Empire de Eussie, pt. ii. p. 89. 130. 6 Open Polar Sea, p. 432. NORTH AMERICA 119 * embraces, when the nuptial day comes, she is obliged ^ by the inexorable law of public opinion to free herself, * if possible, by kicking and screaming with might and ' main, until she is safely landed in the hut of her future ^ lord, when she gives up the combat very cheerfully ^ and takes possession of her new abode.' In Greenland, according to Egede, ' when a young ^ man likes a maiden, he commonly proposes it to their * parents and relations on both sides ; and, after he has ' obtained their consent, he gets two or more old women * to fetch the bride (and if he is a stout fellow he will * fetch her himself). They go to the place where the ^ young woman is, and carry her away by force.' ^ We have already seen (p. 101) that marriage by capture exists in full force among the Northern Red- skins. Further south in California, ' when an Oleepa ' lover wishes to marry, he first obtains permission from ^ the parents. The damsel then flies and conceals her- ' self; the lover searches for her, and, should he succeed * in findino; her twice out of three times, she belono's to ' him. Should he be unsuccessful, he waits a few weeks ^ and then repeats the performance. If she again elude ' his search, the matter is decided against him.' ^ Among the Mosquito Indians also, after the wed- ding is all arranged and the presents paid, the bride is arrayed in her best, and the bridegroom on a given signal rushes in, seizes his bride, and carries her oiF, followed by her female relatives, who pretend to try to rescue her.^ ^ History of Greenland, p. 143. ^ Bancroft, Native Races of the Crantz, Hist, of Greenland, vol. i. p. Pacific States, p. 389. 158. • ^ J^oc. cit. p. 733. 120 SOUTH AMERICA— FIJIAN8 The aborigines of the Amazon Valley, says Wallace,^ ' have no particular ceremony at their marriages, except ^ that of always carr3ring away the girl by force, or ' making a show of doing so, even when she and her * parents are quite willing.' M. Bardel, in the notes to D'Urville's Yoyage, mentions that among the Indians round Conception, in South America, after a man has agreed on the price of a girl with her parents, he sur- prises her, and carries her off to the woods for a few days, after which the happy couple return home.'^ In Tierra del Fuego, as Admiral Fitzroy tells us,^ as soon ' as a youth is able to maintain a wife by his ' exertions in fishing or bird- catching, he obtains the ' consent of her relations, and .... having built or •^ stolen a canoe for himself, he watches for an oppor- ' tunity, and carries off his bride. If she is unwilling ^ she hides herself ui the woods until her admirer is Vheartily tired of looking for her, and gives up the ^ pursuit ; but this seldom happens.' Williams mentions that among the Fijians the cus- tom prevails ^ of seizing upon a woman by apparent ' or actual force, in order to make her a wife. On ' reaching the home of her abductor, should she not ' approve of the match, she runs to some one who can ' protect her ; if, however, she is satisfied, the matter is ' settled forthwith ; a feast is given to her friends the ' next morning, and the couple are thenceforward con- * sidered as man and wife.' * Earle ^ gives the following account of marriage in ^ Travels in the Amazons, p. 497. * Fiji and the Fijians, vol. i. p. 2 Vol. iii. pp. 277 and 22. 174. 5 Voyage of the ^Adventure' and ^ Residence in New Zealand, p» 'Beagle,'Tol. ii. p. 182. 244. POLYNESIANS— TSILIFFINE ISLANDERS 121 ISTew Zealand, which he regards as ' most extraordmary/ while in reality it is, as we now see, nothing of the sort : — ' The New Zealand method of courtship and ' matrimony is,' he says, ' most extraordinary ; so much ' so that an observer could never imagine any affection ' existed between the parties. A man sees a woman ' whom he fancies he should like for a wife ; he asks the ' consent of her father, or, if an orphan, of her nearest ' relation ; which, if he obtains, he carries his " intended " 'off by force, she resisting with all her strength ; and, ' as the New Zealand girls are generally pretty robust, ' sometimes a dreadful struggle takes place ; both are ' soon stripped to the skin ; and it is sometimes the ' work of hours to remove the fair prize a hundred ' yards. If she breaks away she instantly flies from her ^ antagonist, and he has his labour to commence again.' Even after a marriage, it is customary in New Zea- land to have a mock scufHe. Mr. Yate ^ gives a good illustration. There was, he says, ' a little opposition to ' the wedding, but not till it was over, as is always the ' custom here. The bride's mother came to me the ' preceding afternoon, and said she was well pleased in ' her heart that her daughter was going to be married ' to Pahau ; but that she must be angry about it with ' her mouth in the presence of her tribe, lest the natives ^ should come and take away all her possessions, and ' destroy her crops. This is customary on all occasions.' Among the Ahitas of the Philippine Islands, when a man wishes to marry a girl, her parents send her before sunrise into the Avoods. She has an hour's start, after which the lover goes to seek her. If he finds her ^ Yate's New Zealand, p. 96. 122 NEGBIT08—AFBICA and brings her back before sunset, the marriage is acknowledged ; if not, he must abandon all claim to her.^ The natives of New Guinea also have a very similar custom.^ Among the Kaffirs marriage is an affair of purchase, notwithstandiug which ' the bridegroom is required to ' carry off his bride by force, after the preliminaries are ' completed. This is attempted by the help of all the ' friends and relatives that the man can muster, and ' resisted by the friends and relatives of the woman ; ' and the contest now and then terminates in the dis- * comfiture of the unlucky husband, who is reduced to ' the necessity of waylaying his wife, when she may ^ be alone in the fields or fetching water from the well.'^ In the West African kingdom of Futa,^ after all other preliminaries are arranged, ' one difficulty yet * remains, viz. how the young man shall get his wife ' home ; for the women-cousins and relations take on ' mightily, and guard the door of the house to prevent ' her being carried away. At last, by the bridegroom's ' presents and generosity, their grief is assuaged. He ' then provides a friend, well mounted, to carry her off ; ' but as soon as she is on horseback the women renew ' their lamentations, and rush in to dismount her. ' However, the man is generally successful, and rides ' off with his prize to the house prepared for her.' Gray mentions^ that a Mandingo (West Africa), ^ Earl's Native Races of the Good Hope, p. 249; and Maclean's Indian Archipelago, p. 133. Kaffir Laws and Customs, p. 52. : ^ Gerland's Waitz' Anthropolo- * Astley's Collection of Voyages, gie, vol. i. p. 633. vol. ii. p. 240. ^ Pritchard's Nat. Hist, of Man, ^ Gray's Travels in Western ii. 403. See also Arbousset's Tour Africa, p. 56. to the North-east of the Cape of AFBIGA 123 wishing to marry a young girl at Kayaye, applied to her mother, who ' consented to his obtaining her in any ^ way he could. Accordingly, when the poor girl was ' employed in preparing some rice for supper, she was ' seized by her intended husband, assisted by three or ^ four of his companions, and carried off by force. She ' made much resistance, by biting, scratching, kick- * ing, and roaring most bitterly. Many, both men and ' women, some of them her own relations, who wit- ' nessed the affair, only laughed at the farce, and con- ^ soled her by saying that she would soon be reconciled * to her situation.' Evidently therefore this was not, as Gray seems to have supposed, a mere act of lawless violence, but a recognised custom, which called for no interference on the part of spectators. Denham,^ de- scribing a marriage at Sockna (North Africa), says that the bride is taken on a camel to the bridegroom's house, ' upon which it is necessary for her to appear greatly ' surprised, and refuse to dismount ; the women scream, ' the men shout, and she is at length persuaded to ^ enter.' Thompson found a similar custom among the Watuta of Masai Land. Among the Arabs of Sinai, when a marriage has been arranged, the girl is waylaid by her lover ' and ^ a couple of his friends, and carried off by force to his ' father's tent. If she entertains any suspicion of their ' designs, she defends herself with stones, and often * inflicts wounds on the young men, even though she ^ does not dislike the lover.' ^ ^ Loc. cit. vol. i. p. 39. douins and Wahabys, vol. i. p. 263. 2 Burckliardt's Notes on the Be- See also pp. 108, 234. 124 GIBGASSIA—EUBOPU—ROMU In Circassia weddings are accompanied by a feast, ^ in the midst of whicli the bridegroom has to rush in, ' and, with the help of a few daring young men, carry ^ off the lady by force ; and by this process she becomes ' the lawful wife.' ^ According to Spencer, another im- portant part of the ceremony consists in the bridegroom drawing his dagger and cutting open the bride's corset. As regards Europe, Plutarch ^ tells us that in Sparta the bridegroom usually carried off his bride by force, evidently, however, of a friendly character. I would venture to suggest that the character of Helen, as portrayed in the ' Iliad,' can only be understood by regarding her marriage with Paris as a case of marriage by capture.^ ' Les premiers Romains,' says Ortolan,* ^ ont ete obliges de recourir a la surprise et a la force * pour enlever leurs premieres femmes,' and he points out that long after any actual violence had ceased it was customary to pass a lance over the head of the bride, ' en signe de la puissance que vaacquerir le mari." Hence also, while a man might be married in his absence, this was not the case as regards the woman. A man might capture a bride for his friend, but the woman could not be captured unless really present.^ In North Friesland, ' a young fellow called the bride- ' lifter lifts the bride and her two bridesmaids upon the ^ waggon in which the married couple are to travel to * their home.' ^ M'Lennan states that in some parts of France, down to the seventeenth century, it was cus- ^ Moser, The Caucasus and its * Expi. Hist, des Inst, de I'Emp. People, p. 31 ; quoted by M'Lennan, Justinien, pp. 81, 82. loc. cit. p. 36. 5 j^QQ ^^Y. p. 127. 2 gge also Herodotus, vi. 65. « M'Lennan, loc. cit. p. 33. ^ See Appendix. POLAND— BUS8IA— BRITAIN^ IBELAND 125 tomary for the bride to feign reluctance to enter the bridegroom's house. In Poland, Lithuania, Russia, and parts of Prussia, according to Seignior Gaya,^ young men used to carry oiF their sweethearts by force, and then apply to the parents for their consent. Lord Kames,^ in his ' Sketches of the History of ^ Man,' mentions that the following marriage ceremony was, in his day, or at least had till shortly before, been customary among the Welsh : — ' On the morning of ' the wedding-day the bridegroom, accompanied by his ' friends on horseback, demands the bride. Her friends, * who are likewise on horseback, give a positive refusal, ' on which a mock scuffle ensues. The bride, mounted ^ behind her nearest kinsman, is carried off, and is pur- ' sued by the bridegroom and his friends, with loud ' shouts. It is not uncommon on such an occasion to ' see 200 or 300 sturdy Cambro-Britons riding at full ' speed, crossing and jostling, to the no small amuse- ' ment of the spectators. When they have fatigued ' themselves and their horses, the bridegroom is suffered ' to overtake his bride. He leads her away in triumph, ' and the scene is concluded with feasting and festivity.' Sir H. Piers says that in Ireland, after a marriage had been arranged, ^ on the day of bringing home, the 'bridegroom and his friends ride out and meet the bride ' and her friends at the place of meeting. Being come ' near each other, the custom was of old to cast short ' darts at the company that attended the bride, but at ' such distance that seldom any hurt ensued. Yet it is ^ Marriage CeremoDies, p. 35. chapter ix. See also Olaus Magnus, vol. xiv. ^ History of Man, vol. ii, p. 59. 126 TURKEY— BENGAL— PEILIPPINES ^ not out of the memory of man that the Lord of Hoath ' on such an occasion lost an eye.' ^ In European Turkey Mr. Tozer tells us that ' the ^ Mirdites never intermarry ; but when any of them, ^ from the highest to the lowest, wants a wife, he carries ' off a Mahometan woman from one of the neighbouring ^ tribes, baptizes her, and marries her. The parents, we ^ were told, do not usually feel much aggrieved, as it ^ is well understood that a sum of money will be paid / in return.' '^ To these instances many others might have been added, as for instance the natives of Sumatra, the Mapuches, Bushmen, &c. In all these cases the girl is carried off by the man ; but among the Garos of Bengal we find a similar custom, only that it is the bridegroom who is carried off. He pretends to be unwilling and runs away, but is caught by the friends of the bride, and then taken by force, ' in * spite of the resistance and counterfeited grief and ^ lamentation of his parents, to the bride's house.' ^ So also among the Ahitas of the Philippine Islands, if her parents will not consent to a love match, the girl seizes the young man by the hair of his head, carries him off, and declares she has run away with him. In such a case it appears that marriage is held to be valid, whether the parents consent or not.^ Thus, then, we see that marriage by capture, either as a stern reality or as an important ceremony, prevails in Australia, among the Malays, in Hindostan, Central ^ Descr. of Westmeath. Quoted ^ Bon wick, The Tasmanians, p. by M'Lennan. 71. ^ The Highlands of Turkey, vol. * Dalton, Descr. Ethn. of Bengal, i. p. 318. p. 64. WIDE BANGS OF MAEEIAGE BY GAPTUBE 127 Asia, Siberia, and Kamskatka ; among the Esquimaux, the Northern Redskins, the Aborigines of Brazil, in Chili, and Tierra del Fuego, in the Pacific Islands, both among the Polynesians and the Fijians, in the Philip- pines, in Tasmania, among the Kaffirs, Arabs, and Negroes, in Circassia, and, until recently, throughout a great part of Europe. I have already referred to the custom of lifting the bride over the doorstep, which we find in such distinct and distant races as the Romans, the Redskins of Canada, the Chinese, and the Abyssinians. Hence, also, perhaps, our honeymoon, during which the bride- groom keeps his bride away from her relatives and friends ; hence even, perhaps, as Mr. M'Lennan supposes, the slipper is, in mock anger, thrown after the departing bride and bridegroom. The curious custom which forbids the father-in-law and the mother-in-law to speak to their son-in-law, and vice versa, which I have already shown (p. 12) to be very widely distributed, but for which no satisfactory explanation has yet been given, seems to be a natural consequence of marriage by capture. When the capture was a reality, the indignation of the parents would also be real • when it became a mere symbol, the parental anger would be symbolised also, and would be continued even after its origin was forgotten.^ According to- statistics collected by Mr. Tylor ^ this avoidance seems to be especially frequent in tribes where the custom is for the husband to live with his wife's relations. At 1 I am glad to see that Mr. Howitt's Kamilaroi and Kurnai, p. Morgan is disposed to adopt this 16. suggestion. Introd. to Fison and ^ jo^^i-jj^^j^thr. Inst. 1889, p.24. 128 MABBIAGJE BY CONFAREEATIO first this seemed to me unfavourable to the view which connects it with marriage by capture, but this is, after all, not so. If the theory is that the wife has been cap- tured, and yet the husband lives with the wife's relations, the natural way of marking the fact would be that the relations would show their displeasure. The separation of husband and wife, to which also I have referred (p. 72), may also arise from the same custom. It is very remarkable, indeed, how persistent are all customs and ceremonies connected with marriage. Thus our ' bride cake,' which so invariably accompanies a wedding, and which must always he cut by the bride, may be traced back to the old Roman form of marriage by ' confarreatio,' or eating together. So also among the Iroquois, bride and bridegroom used to partake to- gether of a cake of ' sagamite,' ^ which the bride offered to her husband. The Fiji Islanders^ have a very similar custom. The marriage ceremony in Samoa, says Turner, ' reminds us of the Roman confarreatio.' ^ ' Confarreatio ' also exists among the Karens and Bur- mese.^ Again, among the Tipperahs, one of the Hill tribes of Chittagong, the bride prepares some drink, * sits on her lover's knee, drinks half, and gives him the * other half ; they afterwards crook together their little ' fingers.' ^ In one form or another a similar custom is found among most of the Hill tribes of India. Among the Ghiliaks (JSTorthEast Asia) the definitive part of the marriage ceremony consists in the bride and bridegroom ^ Lafitau, vol. i. pp. 566, 571. ^ M'Mahon, The Karens of the ^ Fiji and the Fijians, vol. i. p. G. Chersonese, pp. 322, 350. 170. ^ Lewin's Hill Tracts of Chitta- ^ Nineteen Years in Polynesia, gong, pp. 71^ 80. Dalton's Descr. p. 186. Ethn. of Bengal, p. 193. MAEBIAGB BY CAPTURE 129 drinking out of the same cup. A very similar custom occurs in New Guinea ; ^ among the Samoyedes, and in Madagascar also, part of the marriage ceremony con- sists in the bride and bridegroom eating out of one dish.^ The German word ' vermahlen ' points to the same idea. Among the Chuckmas (a tribe residing among the Chittagong hills) the bride and bridegroom are bound together with a muslin scarf, and then eat together.^ Here also I must mention the curious custom of l3oy -marriages, under which a girl is legally married to a mere boy, who is regarded as the father of her children, while she herself lives with someone else, generally the father of her nominal husband. This arrangement is found among some of the Caucasian tribes, in parts of Russia, among the Reddies in South India, and the •Chibchas of New Granada. It has not, I think, been satisfactorily explained. Mr. M'Lennan conceives that marriage by capture arose from the custom of exogamy, that is to say, from the custom which forbade marriage withm the tribe. Exogamy, again, he considers to have arisen from the practice of female infanticide. I have already indicated the reasons which prevent me from accepting this ex- planation, and which induce me to regard exogamy as arising from marriage by capture, not marriage by cap- ture from exogamy. Mr. M'Lennan's theory seems to me quite inconsistent with the existence of tribes which have marriage by capture and yet are endogamous. ' Gerland's Con. of Waitz' An- People, p. 193. throp., vol. vi. p. 633. ^ Lewin, Wild Tribes of South- ^ Sibree's Madagascar and its eastern India, p. 177. K 130 EXPIATION FOB MARRIAGE The Bedouins, for instance, have marriage by capture, and yet the man has a recognised right to marry his cousin, if only he be willing to give the price demanded for her.^ Mr. M'Lennan, indeed, feels the difficulty which would be presented by such cases, the existence of which he seems, however, to doubt ; adding, that if the symbol of capture be ever found in the marriage ceremonies of an endogamous tribe, we may be sure that it is a relic of an early time at which the tribe was organised on another principle than that of exogamy.^ Another objection to his theory is the presence of marriage by capture with polygamy. That marriage by capture has not arisen merely from female coyness is, I think, evident, as already mentioned ; firstly, because it does not account for the resistance of the relatives ; secondly, because it is contrary to all experience that feminine delicacy diminishes with civi- lisation ; and thirdly, because the very question to be solved is why it has become so generally the custom to win the wife by force rather than by persuasion. It leaves moreover entirely unexplained the case men- tioned on p. 126, in which the man, not the bride, is captured. The explanation which I have suggested derives additional probability from the evidence of a general feeling that marriage was an act for which some com- pensation was due to those whose rights were invaded. The nature of the ceremonies by which this was effected makes me reluctant to enter into this part of 1 Klemm, Allgem, Culturg. d. Menscli, vol. iv. p. 146. ^ Zoc. cit. p. 53. EXPIATION FOB MAEBIAGE 131 the subject at length ; and I will here, therefore, merely indicate in general terms the character of the evidence. I will firstly refer to certain details given by Dulaure ^ in his chapter on the worship of Venus, of which he regards these customs merely as one illustration, although they have, I cannot but think, a signification deeper than, and different from, that which he attri- butes to them. We must remember that the better known savage races have, in most cases, now arrived at the stage in which paternal rights are recognised, and hence that fathers can and do sell their daughters into matrimony. The price of a wife is, of course, regulated by the circumstances of the tribe, and every, or nearly every, industrious young man is enabled to buy one for him- self. As long, however, as communal marriage rights were in force this would be almost impossible. That special marriage was an infringement of these com- munal rights, for which some compensation was due, seems to me the true explanation of the offerings which virgins were so generally compelled to make before being permitted to marry.- The same feeling, probably, gave rise to the curious custom existing, according to Strabo,^ among the ( Par- thian) Tapyrians, that when a man had two or three children by one wife, he was obliged to leave her, so that she might marry someone else. There is some reason to suppose that a similar custom once prevailed among the Romans ; thus Cato, who was proverbially austere in his morals, did not think it right permanently ^ Hist, abregee des diiF. Cultes. ^ See Appendix. 3 Strabo, ii. pp. 515, 520. 132 TEMFOBABY WIVES to retain his wife Martia, whom his friend Hortensius wished to marry. This he accordmgly permitted, and Martia lived with Hortensius until his death, when she returned to her first husband. The high character of Cato is sufficient proof that^he would not have permitted this, if he had regarded it as wrong ; and Plutarch ex- pressly states that the custom of lending wives existed among the Romans. Akin to this feeling is that which induces so many savage tribes ^ to provide their guests with temporary wives. To omit this would be regarded as quite inhospitable. The practice, moreover, seems to recognise the existence of a right inherent in every member of the community, and to visitors as temporary members ; which, in the case of the latter, could not be abrogated by arrangements made before their arrival, and, consequently, without their concurrence. The prevalence of this custom brings home to us forcibly the difference existing between the savage and the civilised modes of regarding the relation of the sexes to one another. Perhaps the most striking case of all is that afforded by some of the Brazilian tribes. The captives taken by them in war used to be kept for some time and fatted up ; after which they were killed and eaten. Yet even here, during the time that they had to live, ^ach poor wretch was generally provided with a tem- porary wife.^ This view also throws some light on the remarkable subordination of the wife to the husband, which is so ^ For instance, the Esquimaux, Arabs, Abyssinians, Kaffirs, Mongols, ;Nortli and South. American Indians, Tutski, &c. Polynesians, Australians, Berbers, ^ Lafitau,MoeursdesSauv.Amer., Eastern and Western Negroes, vol. ii. p. 294. UXOGAMY 133. - characteristic of marriage, and so curiously inconsistent with all our avowed ideas ; moreover, it tends to explain those curious cases in which HetairsB were held in greater estimation than those women who were, as we should consider, properly and respectably married to a. single husband.^ The former were originally fellow - countrywomen and relations ; the latter captives and. slaves. And even when this ceased to be the case, the idea would long survive the circumstances which gave rise to it.^ 1 now pass to the curious custom, for which M'Lennan has proposed the convenient term ' exogamy ^ — that, namely, of necessarily marrying out of the tribe. Tylor, who also called particular attention to this custom in his interesting work on ' The Early ^ History of Man,' which was published in the very same year as M'Lennan's ' Primitive Marriage,' thought that ' the evils of marrying near relatives might be the ' main ground of this series of restrictions.' Morgan^ also considers exogamy as * explainable, and only ex- ' plainable, as a reformatory movement to break up the ' intermarriage of blood relations,' and which could only be effected by exogamy, because all in the tribe were regarded as related. We cannot, however, attribute to savages any such farsighted ideas. Moreover, in fact, exogamy afforded little protection against the marriage of relatives, and, wherever it was systematised. it per- mitted marriage even between half brothers and sisters^ either on the father's or mother's side. Where an ob- ^ Bachofen, Das Mutterrecht, ^ See Appendix, pp. xix. 125. Burton's Lake Ke- ^ Proc. Amer. Acad, of Arts and gions of Africa, vol. i. p. 198. Sciences, 1866. 134 OEIGIN OF EXOGAMY jection to the intermarriage of relatives existed, exogamy was unnecessary ; where it did not exist, exogamy, if this view was correct, could not arise. M'Lennan says, ' I believe this restriction on mar- ^ riage to be connected with the practice in early times * of female infanticide, which, rendering women scarce, * led at once to polyandry within the tribe, and the cap- * turing of women from without.' -^ He has not alluded to the natural preponderance of men over women. Thus, throughout Europe, the proportion of boys to girls is as 106 to 100. '^ Here, therefore, even without in- fanticide, we see that there is no exact balance between the sexes. In many savage races, in various parts of the world, it has been observed the men are much more numerous, but it is difficult to ascertain how far this is due to an original difference, and how far to other causes. Moreover, many of the races which are endo- gamous in one sense, as not marrying out of the tribe, are yet exogamous in the true sense, as not marrying within the ' gens.' It is conceivable that the difference between endo- gamous and exogamous tribes may have been due to the different proportion of the sexes : those races tend- ing to become exogamous where boys prevail ; those, on the other hand, endogamous where the reverse is the case.^ I am not, however, aware that we have any statistics which enable us to determine this point, nor do I believe that it is the true explanation of the custom. Infanticide is, no doubt, very prevalent among ^ Loc. cit. p. 138. 2 Waitz' Anthropology, p. 111. ' See Das Mutterrecht, p. 109. ORIGIN OF EXOGAMY 135 savages. As long, indeed, as men were few in number, enemies were scarce and game was tame. Under these circumstances, there was no temptation to mfanticide. There were some things which women could do better than men — some occupations which pride and laziness, or both, induced them to leave to the women. As soon, however, as in any country population became even slightly more dense, neighbours became a nuisance. They invaded the hunting grounds, and disturbed the game. Hence, if for no other reason, Avars would arise. Once begun, they would break out agam and again, under one pretence or another. Men for slaves, women for wives, and the thirst for glory, made a weak tribe always a temptation to a strong one. Under these circumstances, female children became a source of weakness in several ways. They ate, and did not hunt. They weakened their mothers when young, and, when grown-up, were a temptation to surrounding tribes. Hence female infanticide is easily accounted for. Yet I cannot regard it as the true cause of exogamy. It does not appear to have been so general as Mr. McLennan supposes, nor does it specially characterise the very lowest races. I cannot, then, regard as satisfactory any of the explanations which have hitherto been proposed to account for the origin of exogamy. The true solution is, I think, of a diiferent character. We must remem- ber that under the communal system the women of the tribe were all common property. No one could appropriate one of them to himself without infriuging on the sreneral rio^hts of the tribe. Women taken in war were, on the contrary, in a different position. The 136 PEEVALENCU OF EXOGAMY— AUSTRALIA tribe, as a tribe, had no right to them, and men surely would reserve to themselves exclusively their own prizes. These cajDtives, then, would naturally become the wives in our sense of the term. Several causes would tend to increase the importance of the separate, and decrease that of communal, marriage. The im- pulse which it would give to, and receive back from, the development of the affections ; the convenience with reference to domestic arrangements ; the natural vfishes of the wife herself; and, last not least, the inferior energy of the children sprung from ' in and in ' marriages, would all tend to increase the importance of individual marriage. Even were there no other cause, the advantage of crossing, so well known to breeders of stock, would soon give a marked preponderance to those races by whom exogamy was largely practised, and for several reasons, therefore, we need not be surprised to find exo- gamy very prevalent among the lower races of man. When this state of things had gone on for some time^ usage, as M'Lennan well observes, would ' establish a ' prejudice among the tribes observing it — a prejudice ' strong as a principle of religion, as every prejudice ' relating to marriage is apt to be — against marrying ' women of their stock.' ^ We should not, perhaps, have a piiori expected to find among savages any such remarkable restriction, yet it is very widely distributed ; and from this point of view we can, I think, clearly see how it arose. In Australia, where the same family names are com- mon almost over the whole continent, no man may ^ Loc. cit. p. 140. A USTEALIA— AFRICA 137 marry a woman whose family name is the same as his own, and who belongs therefore to the same tribe.^ ' ISo man/ says Mr. Lang, ' can marry a woman of ^ the same clan, though the parties be no way related ' according to our ideas.' ^ In addition to the Australian cases already mentioned (ante, p. 88), the natives of West Australia and Port Lincoln are divided into two great clans, and no man may marry a woman of the same clan.^ So also in New Britain and the Duke of York group of islands the natives are divided into two classes, and marriage be- tween persons of the same clan is thought very dis- reputable.'* In Eastern Africa, Burton ^ says that ' some clans of ' the Somal will not marry one of the same, or even of ' a consanguineous family ; ' and the Bakalari have the same rule.^ Du Chaillu,^ speaking of Western Equatorial Africa, says, ^ the law of marriages among the tribes I have ' visited is peculiar ; each tribe is divided into clans ; * the children in most of the tribes belong to the clan of ' the mother, and these cannot by any possible laws ' marry among themselves, however removed in degree ' they may have been connected : it is considered an ' abomination among them. But there exists no ob- ^jection to possessing a father's or brother's wife. I ^ Eyre's Discoveries in Australia, •* Brown, quoted in Wallace's vol. ii. p. 329. Grey's Journal, p. Australasia, p. 470. 242. 5 First Footsteps, p. 120. 2 The Aborigines of Australia, '^ Trans. Ethn. Soc, N.S., vol. i. p. 10. Taplin's The Narinyeri, p. 1. p. 321. ^ Forrest, Journ. Anthrop. Insti- "^ Ibid. p. 307. tute, vol. V. p. 317. 138 EINDOSTAN ' could not but be struck with the healthful influence ' of such regulations against blood marriages among ' them.' In India the Khasias/ Juangs,^ and Waralis are divided into sections, and no man may marry a woman belonging to his own section. In the Magar tribes these sections are called Thums, and the same rule pre- vails. Colonel Dalton tells us that ' the Hos, Moondahs, ' and Oraons are divided into clans or keelis, and may ' not take to wife a girl of the same keeli.' Again, the Garrows are divided into " maharis,' and a man may not marry a girl of his own ' mahari.' The Munnieporees and other tribes inhabiting the hills round Munniepore — the Koupooees, Mows, Mu- rams, and Murrings, as M'Lennan points out on the authority of M'Culloch — ' are each and all divided into ' four families : Koomrul, Looang, Angom, and King- ' thaja. A member of any of these families may marry ' a member of any other, but the intermarriage of ^ members of the same family is strictly prohibited.' ^ On the contrary, the Todas, says Metz,^ ' are divided ' into five distinct classes, known by the names Peiky, ' Pekkan, Kuttan, Kennae, and Tody ; of which the ^ first is regarded as the most aristocratic. These classes ^ do not even intermarry with each other, and can there- ' fore never lose their distinctive characteristics.' The Khonds, as we are informed by General Campbell, ^ re- ^ gard it as degrading to bestow their daughters in ^ Godwin Austen, Journ. Anthr. ^ Account of the Valley of Mun- Inst., 1871, p. 131. niepore, 1859, pp. 49, 69. 2 Dalton'sDescr.Ethn. of Bengal, ^ Tribes of the Neilgherry Hills, p. 158. p. 21. NEPA UL— CEYLON— CIEGASSIA 139 ^ marriage on men of their own tribe ; and consider it ^ more manly to seek their wives in a distant country.' ^ Major M'Pherson also tells us that they consider mar- riage between people of the same tribe as wicked, and punishable with death. The mountain tribes of Nepaul, before the advent of the Kajpoots, are said to have con- sisted of twelve Thums or clans, and no man was per- mitted to marry a woman of the same Thum.'^ We are indebted to Mr. Brito,^ of Colombo, for a very interesting treatise on the rules of succession among the Mukkuvars of Ceylon. These rules are founded on the custom that no one may marry a person of the same ' kudi,' i.e. anyone who is related on the mother's side. Indeed, all relationship is from the mother, none from the father ; succession is traced through the mother ; land, if inherited, is out of marital power, and is managed by the males for the females. The Kalmucks, according to De Hell, are divided into hordes, and no man can marry a woman of the same horde. The bride, says Bergman, speaking of the same people, is always chosen from another stock ; ' among the Derbets, for instance, from the Torgot ' stock, and araono; the Toro^ots from the Derbet ' stock.' The same custom prevails among the Circassians and the Samoyedes.^ The Ostyaks regard it as a crime to marry a woman of the same family or even of the same name.^ When a Jakut (Siberia) wishes to marry, he must, ' Campbell, p. 142. ^ xhe MukkuTa Law. 2 Hamilton's Account of the King- ^ Pallas, vol. iv. p. 96. dom of Nepaul, p. 27. ^ Ibid. vol. iv. p. 69. 140 CHINA— NORTH AMEUIGA says Middendorf/ choose a girl from another clan. No one is permitted to marry a woman from his own. In China, says Davis, ^ ' marriage between all persons of ' the same surname being unlawful, this rule must of * course include all descendants of the male branch for ^ ever ; and as, in so vast a population, there are not a ' great many more than one hundred surnames through- ' out the empire, the embarrassments that arise from so ' strict a law must be considerable.' Among the Tinne Indians of North- West America, ' a Chit-sangh cannot, by their rules, ^ marry a Chit- * sangh, although the rule is set at naught occasionally ; * but when it does take place the persons are ridiculed ' and laughed at. The man is said to have married his ' sister, even though she may be from another tribe, ' and there be not the slightest connection by blood ' between them. The same way with the other two ^ divisions. The children are of the same colour as ^ their mother. They receive caste from their mother : ' if a male Chit-sangh marry a Nah-tsingh woman, the ' children are Nah-tsingh ; and if a male Nah-tsingh ' marry a Chit-sangh woman, the children are Chit- ^ sangh, so that the divisions are always changing. As ' the fathers die out, the country inhabited by the * Chit-sangh becomes occupied by the Nah-tsingh, and ' so vice versa. They are continually changing coun- ' tries, as it were.' Among the Kenaiyers (N.W. America), ' it was the ^ custom that the men of one stock should choose their ^ Sibirisclie Reise, p. 72. See ^ r^^^^ Chinese, vol. i. p. 282. also Miiller's Descr. de toutes les ^ Notes on the Tinneh. Har- Races de I'Einp. de Russie, pt. ii. disty, Smithsonian Report, 1866^ p. 68. p. 315. EXOGAMY IN NORTH AMEBIGA 141 * wives from another, and the offspring belonged to * the race of the mother. This custom has fallen into ' disuse, and marriages in the same tribe occur ; but the ' old people say that mortality among Kenaiyers has ^ arisen from the neglect of the ancient usage. A man's ' nearest heirs in this tribe are his sister's children.' ^ The Tsimsheean Indians of British Columbia^ are similarly divided into tribes and totems, or ' crests, ^ which are common to all the tribes. The crests are ' the whale, the j^orpoise, the eagle, the coon, the wolf, * and the frog. In connection with these crests, several ' very important points of Indian character and law are ' seen. The relationship existing between persons of ^ the same crest is nearer than that between members of ' the same tribe, which is seen in this, that members of ' the same tribe may marry, but those of the same crest * are not allowed to do so under any circumstances ; ' that is, a whale may not marry a whale, but a whale ' may marry a frog,' &c. Very similar rules exist among the Thlinkeets,^ and indeed, as regards the Northern Redskins generally, it is stated^ in ' Archaeologia Americana' that 'every ' nation was divided into a number of clans, varying in ' the several nations from three to eight or ten, the ' members of which respectively were dispersed indis- ' criminately throughout the whole nation. It has been ' fully ascertained that the inviolable regulations by ' which these clans were perpetuated amongst the ^ Richardson's Boat Journey, p. 6, vol. i. p. 406. See also Smithsonian ^ Bancroft, loe. cit. vol. i. p. 1 09. Report, 1866, p. 326. * Gallatin, loc. cit. vol. xi. p. 109. ^ Metlahkatlah, puhlished by the Lafitau, vol. i. p. 558. Tanner's Church Missionary Society, 1869? Narrative, p. 313. 142 EXOGAMY IN SOUTH AMEBIGA ' southern nations were, first, that no man could marry ^ in his own clan ; secondly, that every child should ' belong to his or her mother's clan.' Among the Mayas of Yucatan, according to Ilerrera, marriage was forbidden between people of the same name. The Indians of Guiana ^ ' are divided into families, * each of which has a distinct name, as the Siwidi, ' Karuafudi, Onisidi, &c. Unlike our families, these all ' descend in the female line, and no individual of either ' sex is allowed to marry another of the same family ' name. Thus, a woman of the Siwidi family bears the ' same name as her mother, but neither her father nor ' her husband can be of that family. Her children and ' the children of her daughters will also be called Siwidi, ' but both her sons and daughters are prohibited from ' an alliance with any individual bearing the same name ; ' though they may marry into the family of their father ' if they choose. These customs are strictly observed, ^ and any breach of them would be considered as ' wicked.' The Brazilian races, according to Martius, differ greatly in their marriage regulations. In some of the very scattered tribes, who live in small families far remote from one another, the nearest relatives often intermarry. In more populous districts, on the contrary, the tribes are divided into families, and a strict system of exogamy prevails.^ In Mangaia, according to Mr. Gill, in olden times, a man was not permitted to marry a woman of his own tribe.^ ^ Brett's Indian Tribes of Guiana, p. 98. - Zoc. cit. p. 63. 2 Savage Life in Polynesia, p. 136. THE CAUSES OF FOLYGAMY 14a Thus, then, we see that this remarkable custom of exogamy exists throughout Western and Eastern Africa, in Circassia, Hindostan, Tartary, Siberia, China, Polynesia, and Australia, as well as in North and South America. The relations existing between husband and wife in the lower races of man, as indicated in the jDreceding^ pages, are sufficient to remove all surprise at the preva- lence of polygamy. There are, however, other causes,, not less powerful, though perhaps less prominent, ta which much influence must be ascribed. Thus in all tropical regions girls become marriageable very young ; their beauty is acquired early, and soon fades, while men, on the contrary, retain their full powers much longer. Hence, when love depends, not on similarity of tastes, pursuits, or sympathies, but entirely on external attractions, we cannot wonder that every man who is able to do so provides himself with a succession of favourites, even when the first wife remains not only nominally the head, but really his confidant and adviser. Another cause has no doubt exercised great influence. Milk IS necessary for children, and in the absence of do- mestic animals it consequently follows that they are not weaned until they are several years old. The effect of this on the social relations has been already referred to.^ Polyandry, on the contrary, is far less common, though more frequent than is generally supposed.. M'Lennan and Morgan, indeed, both regard it as a phase through which human progress has necessarily passed. If, however, we define it as the condition in which ^ Ante, p. 79. 144 POLYANDRY one woman is married to several men, bat (as distin- guished from communal marriage) to them exclusively, then I am rather disposed to regard it as an exceptional phenomenon, arising from the paucity of females. M'Lennan, indeed,^ gives a long list of tribes which he regards as polyandrous, namely, those of Thibet, Cashmeer, and the Himalayan regions, the Todas, Coorgs, Nairs, and various other races in India, in Ceylon, in New Zealand,^ and one or two other Pacific islands, in the Aleutian Archipelago, among the Koryaks, the Saporogian Cossacks, on the Orinoco, in parts of Africa, and in Lancerote. He also mentions the ancient Britons, some of the Median cantons, the Picts, and the Gretes, while traces of it occurred among the ancient Germans. On the other hand, to the in- stances quoted by M'Lennan we may add that of some families among the Australians,^ Nukahivans,^ and Iroquois. If we examine the above instances, some of them will, I think, prove irrelevant. The passage referred to in Tacitus^ does not appear to me to justify us in regarding the Germans as having been polyandrous. Erman is correctly referred to by M'Lennan as mentioning the existence of ' lawful polyandry in the ^ Aleutian Islands.' He does not, however, give his authority for the statement. The account he gives of the Koryaks by no means, I think, proves that poly- andry occurs among them. The case of the Kalmucks, to judge from the account given by Clarke,^ is certainly ^ Loc. cit. p. 180. 4 Ibid. vol. vi. p. 128. 2 Lafitau, loc. cit. vol. i. p. 555. ^ Germania, xx. 3 Gerland's Waitz' Anthropo- « travels, vol. i. p. 241. logie, vol. vi. p. 774. POLYANDBY EXCEPTIONAL 145 one in which brothers, but brothers only, have a wife in common. For Polynesia, M'Lennan relies on the Legend of Rupe, as told by Sir G. Grey.^ Here, however, it is merely stated that two brothers named Ihuatamai and Ihuwareware, having found Hinauri, when she was thrown by the surf on the coast at Wairarawa, ' looked ' upon her with pleasure, and took her as a wife between ' them both.' This seems to me rather a case of com- munal marriage than of polyandry, especially when the rest of the legend is borne in mind. Neither is the evi- dence as regards Africa at all satisfactory. The cus- tom referred to by M'Lennan^ probably originates in the subjection of the woman which is there implied by marriage, and which may be regarded as inconsistent with high rank. Several of the above cases are, indeed, I think, merely instances of communal marriage. Indeed, it is evident that where our information is incomplete, it must often be far from easy to distinguish between communal marriage and true polyandry. Polyandry is no doubt widely distributed in Ceylon, India, and Thibet, and among some of the hill tribes of India. A pretty Dophla girl once came into the station of Luckimpur, threw herself at Colonel Dalton's feet, ' and in most poetical language asked me ' to give her my protection.' She was promised by her father to a man whom she did not love, and had ' eloped with her beloved. This was interesting and ^ romantic' Colonel Dalton sent for the beloved, and, he ^ Polynesian Mythology, p. 81. "^ Eeade's Savage Africa, p. 43. L 146 TEE 8YSTI]M OF LBYIBATJE says, ^ tlie romance was dispelled. She had eloped with ' two young men.' ^ In Ceylon the joint husbands are always brothers,^ and this is also the case among the tribes residing at the foot of the Himalaya ^ Mountains. But, on the whole, lawful polyandry (as opposed to mere laxness of morality) seems to be an exceptional system, generally intended to avoid the evils arising from monogamy where the number of women is less than that of men. The system of L evirate, under which, at a man's death, his wife or wives pass to his brother, is, I think, more intimately connected with the rights of property than with polyandry. This custom is widely distributed. It is found, for instance, among the Mongols ^ and Kaffirs,^ and in Yucatan.^ When an elder brother dies, says Livingstone,^ ' the same thing occurs in re- ^ spect of his wives ; the brother next in age takes them, ^ as among the Jews, and the children that may be born ^ of those women he calls his brothers also.' In India, among the Nairs, ' a man always takes to ' wife, by the custom called Sagai, his elder brother's ^ widow.' ^ Among the Pacific Islanders, Mr. Brenchley mentions that in Erromango ' the wives of deceased '• brothers fall to the eldest surviving brother.' ^ Similar statements have been made also as regards some of the Negro tribes, the Mexicans, Samoans, New Zealanders, and Khyens. 1 Des. Ethn. of Bengal, p. 36. of the Cape of Good Hope, pp. 38, 2 Davy's Ceylon, p. 286. 138. 2 Eraser's Tour to the Himala ^ Bancroft, vol. ii. p. 671. Mountains, pp. 70, 206. 7 Travels in South Africa, p. 185. * Wuttke's Ges. der Menschheit, ^ Dalton's Des. Ethn. of Bengal, Tol. i. p. 223. p. 138. 5 Arhousset's Tour to the N.E. ^ Cruise of the ' Cura^oa,' p. 319. ENDOGAMY 147 Passing on now to the custom of endogamy, I must first observe that there is not the opposition between exogamy and endogamy which Mr. M'Lennan supposed. Some races which are endogamous as re- gards the tribe are yet exogamous as regards the gens. Thus some of the Indian races, as the Abors,^ Kocchs, and Hos, are forbidden to marry excepting within the tribe. The latter at least, however, are not truly endo- gamous, for, as already mentioned, they are divided into ' keelis ' or clans, and ^ may not take to wife a girl ' of their own keeli.' ^ Thus they are in fact exogamous, and it is possible that some of the other cases of endo- gamy might, if we were better acquainted with them, present the same duplex phenomenon. M'Lennan remarks that ' the separate endogamous ' tribes are nearly as numerous, and they are in some ' respects as rude, as the separate exogamous tribes.' ^ So far as my knowledge goes, on the contrary, endogamy is much less prevalent than exogamy, and it seems to me to have arisen from a feeling of race-pride, as, for instance, in Peru,^ and a disdain of surrounding tribes which were either really or hypothetically in a lower condition, though in some cases it may be due to weakness, and a consequent desh-e to avoid offending powerful neighbours. \ Among the Ahts of N. W. America, as mentioned by Sproat, ' though the different^tribes of the Aht nation ' are frequently at war with one another, women are not ' captured from other tribes for marriage, but only to be ^ Dalton's Des. Ethn. of Bengal, ^ Loc. cit. p. 145. p. 28. ■* Wuttke's Ges. der Menschlieit, 2 Ante, p. 115. vol. i. pp. 325, 331. L 2 148 ENDOGAMY ' kept as slaves. The idea of slavery connected with ' capture is so common, that a free-born Aht would hesi- ' tate to marry a woman taken in war, whatever her ' rank had been in her own tribe.' ^ Endogamy also prevails among several of the wild tribes of Central America.^ Amonsf the Yerkalas ^ of Southern India ' a custom ' prevails by which the first two daughters of a family ^ may be claimed by the maternal uncle as wives for his ' sons. The value of a wife is fixed at twenty pagodas. ' The maternal uncle's right to the first two daughters ' is valued at eight out of twenty pagodas, and is carried ' out thus : if he urges his preferential claim, and ' marries his own sons to his nieces, he pays for each ^ only twelve pagodas ; and, similarly, if he, from not ' having sons, or any other cause, forego his claim, ' he receives eight pagodas of the twenty paid to the ' girls' parents by anybody else who may marry them.' Among some of the Karen tribes marriage between near relations is the rule.^ The Doingnaks, a branch of the Chukmas, appear also to have been endogamous, and Captain Lewin mentions that they ' abandoned the parent stem during 'the chiefship of Jaunbux Khan, about 1782. The ' reason of this split was a disagreement on the subject ' of marriages. The chief passed an order that the ' Doingnaks should intermarry with the tribe in general. ' This was contrary to ancient custom, and caused dis- ' content and eventually a break in the tribe.' ^ This ^ Sproat, Scenes and Studies of vol. vii. p. 187. Savage Life, p. 98. ^ M'Mahon, p. 69. 2 Bancroft, vol. i. p. 703. » Lewin's Hill Tracts of Chitta- 3 Sliortt, Trans. Ethn. Soc, N.S., gong, p. 65. ENB0GA2IY 149 is one of the very few cases where we have evidence of a change in this respect. The Kalangs of Java are also endogamous, and when a man asks a girl in marriage he must prove his descent from their peculiar stock.^ The Alantchu Tartars for- bid marriages between those whose family names are different.^ Among the Bedouins, ^ a man has an ex- ' elusive right to the hand of his cousin,'^ and it is the custom of the Karens that ' marriages must always be ' contracted by relations.' ^ Livingstone also mentions that in South Africa the women of the Akombwi ' never ' intermarry with any other tribe.' ^ In Guam brothers and sisters used to intermarry, and it is even stated that such unions were preferred as being most natural and proper.^ Endogamy would seem to have prevailed in the Sandwich Islands,^ and in New Zealand, where, as Yate mentions, ' great opposition is made to anyone ^ taking, except for some political purpose, a wife from ' another tribe, so that such intermarriages seldom ' occur.' ^ Barrow mentions that the Hottentots seldom married out of their own kraal. ^ On the whole, how- ever, endogamy seems a far less common custom than exogamy. The idea of relationship as existing amongst us, founded on marriage, and implying equal connection of a child to its father and mother, seems so natural and ^ Raffles' History of Java, vol. i. ^ Exp. to the Zambesi, p. 39. p. 328. *^ Arago's Letters. Freycinet's •-^ M'Lennan, loc. cit. p. 146. Voyage, vol. ii. p. 17. 3 Burckhardt's Notes on the ^ Ibid. p. 94. Bedouins and Wahabys, vol. i. pp. ^ New Zealand, p. 99. 113, 272. ^ Travels in South Africa, vol. i. * Morgan, Syst. of Cons, and AfF. p. 144. of the Human Family, p. 444. 150 THU MILK-TW obvious that there are, perhaps, many to whom the possibility of any other system has not occurred. The facts already recorded will, however, have pre23ared us for the existence of peculiar ideas on the subject of rela- tionship. The strength of the foster -feeling, the milk- tie, among the Scotch Highlanders is a familiar instance of a mode of regarding relationship very diiFerent from that prevalent amongst us. We have also seen that, under the custom of com- munal marriage, a child was regarded as related to the tribe, but not specially to any particular father or mother. Such a state of things, indeed, is only pos- sible in very small communities. It is evident that under communal marriage — and little less so wherever polygamy prevailed, and men had many wives — the tie between and father and son must have been very slight. Among agricultural tribes, and under settled forms of government, the chiefs often have very large harems, and their importance even is measured by the number of their wives, as in other cases by that of their cows or horses. This state of things is in many ways very preju- dicial. It checks, of course, the natural affection and friendly intercourse between man and wife. The King of Ashantee, for instance, always had 3,333 wives ; but no man can love so many women, nor can so* many women cherish any personal affection for one man. Even among hunting races, though men were un- able to maintain so many wives, still, as changes are of frequent occurrence, the tie between a mother and child is much strono'er than that which binds a child to its. BELATIONSHIP THROUGH FEMALES 151 father. Hence we find that among many of the lower races relationship through females is the prevalent cus- tom, and we are thus able to understand the curious practice that a man's heirs are not his own, but his sister's children. By some it has been regarded as indicating the high respect paid to women. Thus Plutarch tells us that ^ when Bellerophon slew^ a certain wild boar, which ' destroyed the cattle and fruits in the province of the ' Xanthians, and received no due reward of his services,, ' he prayed to Neptune for vengeance, and obtained ' that all the fields should cast forth a salt dew and be ' universally corrupted, which continued till he, conde- ^ scendingly regarding the women suppliants, prayed to ' Neptune and removed his wrath from them. Hence ^ there was a law among the Xanthians that they should ' derive their names in future, not from the fathers, but ' from the mothers.' ^ Montesquieu^ regarded relationship through females as intended to prevent the accumulation of landed pro- perty in few hands — an explanation manifestly inap- plicable to many, nay, the majority, of cases in which the custom exists — and the explanation above suggested is, I have no doubt, the correct one. Thus, when a rich man died in Gruinea, his property, excepting the armour, descended to the sister's son, expressly, according to Smith, on the ground that he must certainly have been a relative.^ Battel mentions that the town of Longo (Loango) ' is governed by four ^ Plutarch, Concerning the Vir- 143. See also Pinkerton's Voyages, tues of Women. vol. xv. pp. 417, 421, 528 ; Astley's ^ Esprit des Lois, vol. i. p. 70. Collection of Voyages, vol. ii. pp. 3 Smith's Voyage to Guinea, p. 63, 256. 152 BELATIONSEIP THROUGH FUMALBS ' chiefs, who are sons of the king's sisters ; for the ' king's sons never come to be kings.' ^ Quatremere men- tions that ' chez les Nubiens, dit Abou Selah, lorsqu'un ' roi vient a mourir et qu'il laisse un fils et un neveu du ' cote de sa soeur, celui monte sur le tr6ne de preference ' a I'heritier naturel.' ^ In Central Africa, CaiUie ^ says that ' the sovereignty ' remains always in the same family, but the son never ' succeeds his father ; they choose in preference a son of ' the king's sister, conceiving that by this method the ' sovereign power is more sure to be transmitted to one ' of the blood royal ; a precaution which shows how ' little faith is put in the virtue of the women of this * country.' In South Africa, among the Bangalas of the Cassange valley, ^ the sons of a sister belong to her ' brother ; and he often sells his nephews to pay his ' debts ; ' ^ the Banyai ' choose the son of the deceased ' chiefs sister in preference to his own offspring.' In Northern Africa we find the same custom among the Berbers ; ^ Burton records it as existing in the North- East ; and on the Congo, according to Tuckey, the chieftainships ' are hereditary, through the female line, ' as a precaution to make certain of the blood royal in ' the succession.'^ Sibree mentions that the same is the case in Madagascar, where the custom is defended ex- pressly on the ground ' that the descent can be proved ^ Pinkerton's Voj^ages, vol. xvi. p. 273. p. 331. * Livingstone's Travels in South ^ M^m. Geogr. sur I'Egypte et Africa, pp. 434, 617. sur quelques contrees voisines, Paris, ^ La Mere chez certains peuples 1811. Quoted in Bachofen's Mutter- de I'Antiquite, p. 45. recht, p. 108. ^ Tuckey's Exp. to the Kiver 3 Oaillie's Travels, vol. i. p. 153. Zaire, p. 365. Barth's Travels, vol. i. p. 337 ; vol. ii. CAUSES AND WIDE DISTBIBUTION OF TEE CUSTOM 153 * from the mother, while it is often impossible to know ' the paternity of a child.' ^ Herodotus ^ supposed that this custom was peculiar to the Lycians : they have, he says, ' one custom pecu- * liar to themselves, in which they differ from all other ' nations ; for they take their name from their mothers, ' and not from their fathers ; so that if anyone asks ' another who he is, he will describe himself by his ' mother's side, and reckon up his maternal ancestry in ' the female line.' Poly bins makes the same statement as regards the Locrians ; and on Etruscan tombs descent is stated in the female line. In Athens, also, relationship through females pre- vailed down to the time of Cecrops. Tacitus,^ speaking of the Germans, says, ' Children are regarded with equal affection by their maternal uncles as by their fathers ; some even consider this as the more sacred bond of consanguinity, and prefer it in the requisition of hostages.' He adds, ' A person's own children, however, are his heirs and successors ; no wills are made.' From this it would appear as if female inheritance had been recently and not universally abandoned. Among the Picts also the throne until a late period was always held by right of the female. In the Irish Legends it is stated that this was a con- dition imposed by Eremon, who when the Picts were about to invade Scotland supplied them with wives on this condition.^ In India the Kasias, the Kocchs, and the Nairs have ^ Madagascar and its People, ^ De Mor. Germ. xx. p. 192. * Ferguson, The Irish before the 2 Olio, 173. Conquest, p. 129. 154 NEGLECT OF PATERNAL BELATION the system of female kinsliip. BQchanan^ tells us that among the Bantar in Tulava a man's property does not descend to his own children, but to those of his sister. Sir W. Eliot states that the people of Malabar ' all ' agree in one remarkable usage — that of transmitting ' property through females only.' ^ He adds, on the authority of Lieutenant Conner, that the same is the case in Travancore, among all the castes except the Ponans and the Namburi Brahman s. As Latham states, 'no Nair son knows his own ' father ; and, vice versa, no Nair father knows his own ^ son. What becomes of the property of the husband ? ' It descends to the children of his sister.' ^ Among the Limboos (India), a tribe near Darjee- ling,* the boys become the property of the father on his paying the mother a small sum of money, when the child is named and enters his father's tribe : girls re- main with the mother, and belong to her tribe. Marsden tells us ^ that among the Battas of Sumatra ' the succession to the chiefships does not go, in the ^ first instance, to the son of the deceased, but to the ' nephew by a sister ; and that the same extraordinary ' rule, with respect to the property in general, prevails ' also amongst the Malays of that part of the island, ' and even in the neighbourhood of Padang. The ' authorities for this are various and unconnected with ' each other, but not sufficiently circumstantial to ' induce me to admit it as a generally established ' practice.' ^ Vol. iii. p. 16. 4 Campl)ell, Trans. Ethn. Soc, 2 Trans. Ethn. Soc, 1869, p. 119. N. S., vol. vii. p. 155. 3 Descriptive Ethnology, vol. ii. ^ Marsden's History of Sumatra^ p. 463. p. 376. EELATIONSHIF THROUGH FEMALES 155 Among the Kenaiyers at Cook's Inlet, according to Sir Jolm Richardson, property descends, not to a man's own children, but to those of his sister.^ The same is the case with the Kutchin,^ and it is said generally, though not always, among the Columbian Indians.^ Carver'* mentions that among the Hudson's Bay Indians the children 'are always distinguished by the ' name of the mother ; and if a woman marries several ' husbands, and has issue by each of them, they are all ' called after her. The reason they give for this is, that ' as their offspring are indebted to the father for their ' souls, the invisible part of their essence, and to the ' mother for their corporeal and apparent part, it is. ' more rational that they should be distinguished by the ' name of the latter, from whom they indubitably derive ' their being, than by that of the father, to which a ' doubt might sometimes arise whether they are justly ' entitled.' ' Descent amongst the Iroquois is in the ' female line, both as to the tribe and as to nationality. ' The children are of the tribe of the mother. If a ' Cayuga marries a Delaware woman, for example, his * children are Dela wares and aliens, unless formally ' naturalised with the forms of adoption ; but if a ^ Delaware marries a Cayuga woman, her children are ' Cayugas, and of her tribe of the Cayugas. It is the ' same as if she marries a Seneca.' ^ In fact, among the North American Indians gene- rally, as we shall see more particularly in the next ^ Boat Journey, vol. i. p. 406. 259; also ante, p. 112. 2 Smithsonian Keport, 1866, p. ^ Morgan's Syst. of Cons, and 326. Aff. of the Human Family, p. 165. ^ Bancroft, vol. i. p. 193. Hunter's Captivity among the North * Carver, p. 378. See also p. American Indians, p. 249. 156 SUBOBDINATION OF FATEENAL UFLATION chapter, the relationship of the uncle, that is to say, the mother's brother, is more important than any other. He is practically the head of his sister's family. Among the Choctas, for instance, even now, if a boy is to be placed at school, his uncle, and not his father, takes him to the mission and makes the arrangement.^ A similar rule prevailed in Haiti and Mexico.^ According to Gromara, among the Peruvians, except as regards the Incas, nephews inherited, not sons. As regards Polynesia, Mariner states that in the Friendly or Tonga Islands ' nobility descends by the ' female line ; for when the mother is not a noble, the ' children are not nobles.' ^ The same custom, or traces of it, exist throughout Polynesia, but it would seem that these islanders were passing from the stage of relation- ship through females to that through males. The existence of inheritance through females is clearly indi- cated in the Fijian custom known as Vasu. In some of the Carolines and Mariannes the highest honour passed in the female line.^ In the Hervey Islands, children belong either to the tribe of the father or to that of the mother, according to arrangement ; gener- ally, however, to that of the father.^ In Western Australia, ' children of either sex always ^ take the family name of their mother.' ^ In other districts, however, as, for instance, on the Lower 1 Morgan, loc. cit. p. 158. Antlir. vol. v. pt. ii. pp. 108, 114, ^ Miiller, Gescli. d. American. 117. Urreligionen, pp. 167, 539. ^ Qi^^ Myths of the South ^ Tonga Islands, vol. ii. pp. 89, Pacific, p. 36. 91. 6 Eyre, loc. cit. p. 330. Eidley, * Hale, United States Ex. Exp., Journal Anthrop. Institute, 1872, p. 83. Gerland, Con. of Waitz' p. 264. OBIGIN OF BELATIONSHIP IN THE MALE LINE 157 Murray, a man's children belong to his tribe, and not to that of the mother.^ Among the ancient Jews, Abraham married his half sister, Nahor married his brother's daughter, and Amram his father's sister ; this was permitted because they were not regarded as relations. Tamar also evi- dently might have married Amnon, though they were both children of David : ' Speak unto the king,' she said, ' for he will not withhold me from thee ; ' for, as their mothers were not the same, they were no relations in the eye of the law. Solon also permitted marriage with sisters on the father's side, but not on the mother's. Here, therefore, we have abundant evidence of the second stage, in which the child is related to the mother, and not to the father ; whence a man's heir is his nephew on the sister's side — not his own child, who is in some cases regarded as no relation to him at all. When, however, marriage became more respected, and the family affections stronger, it is easy to see that the rule under which a man's property went to his sister's children would become unpopular, both with the father, who would naturally wish his children to inherit his property, and not less so with the children themselves. This change is even now in process among the more civilised North American Indians.^ M. Girard Teulon, indeed, to whom we are indebted for a very interesting memoir on this subject,^ regards the first recognition of his parental relationship as an 1 Taplin, Tlie Narinyeri, p. 10, ^ La Mere chez certains peuples 2 Report of the Peabody Museum, de 1' Antiquity. vol. iii. p. 214. 158 CHANGE FEOM FEMALE TO MALE KINSHIP act of noble self-devotion on the part of some great genius in ancient times. ' Le premier/ he says, ' qui ' consentit a se reconnaitre pere fut un homme de genie ^ et de coeur, un des grands bienfaiteurs de I'humanite. ^ Prouve en efFet que I'enfant t'appartient. Es-tu sur ' qu'il est un autre toi-meme, ton fruit ? que tu I'as ' enfante ? ou bien, a I'aide d'une genereuse et volon- ' taire credulite, marches-tu, noble inventeur, a la con- ' quete d'un but superieur ? ' Bachofen also, while characterising the change from female to male relationship as the ' wichtigsten Wende- ^ punktin der Geschichte des Geschlechts-Yerhaltnisses,' explains it, as I cannot but think, in an altogether erro- neous manner. He regards it as a liberation of the spirit from the deceptive appearances of nature, [an elevation of human existence above the laws of mere matter ; as a recognition that the creative power is the most important ; and, in short, as a subordination of the material to the spiritual part of our nature. By this step, he says, ' man durchbricht die Bander des Tellu- ' rismus, und erhebt seinen Blick zu den hohern Eegionen ' des Kosmos.' ^ These seem to me, I confess, very curious notions, and I cannot at all agree with them. The recognition of paternal responsibility grew up, I believe, gradually through the impulses of natural alFection. The adoption of relationship through the father's line, instead of through the mother's, was probably effected by the natural wish which everyone would feel that his property should go to his own children. It is true that we have not many cases like that of Athens, in which there is a record of ^ Bachofen, Das Mutterrecht; p. 27. CHANGE FROM FJBMALJEJ TO MALE KINSHIP 159 this change ; but as it is easy to see how it might have been brought about, and difficult to suppose that the opposite step can ever have been made ; as, moreover, we find relationship through the father very general, not to say universal, in civilised races, while the opposite system is very common among savages, it is evident that this change must frequently have been effected. Taking all these facts, then, into consideration, when- ever we find relationship through females only, I think we may safely look upon it as the relic of an ancient barbarism. As soon as the change was made, the father would take the place held previously by the mother, and the father, instead of the mother, would be regarded as the parent. Hence, on the birth of a child, the father would natur- ally be very careful what he did, and what he ate, for fear the child should be injured. Thus, I believe, arises the curious custom of the Couvade to which I referred in my first chapter. Relationship to the father at first excludes that to the mother, and, from having been regarded as no relation to the former, children came to be looked on as none to the latter. In some parts of South America, where it is cus- tomary to treat captives well in every respect for a certain time, giving them clothes, food, a wife, &c., and then to kill and eat them, any children they may have are killed and eaten also.^ As a general rule inherit- ance and relationship go together ; but in some parts of Australia, while the old rule of tracing descent through 1 Lafitau, vol. ii. p. 307. 160 SYSTEM OF KINSHIP TEEOUGH MALES the mother still exists, property is inherited in the male line/ though it appears that the division is made during the father's life. How completely the idea of relationship through the father, when once recognised, might replace that through the mother, we may see in the very curious trial of Orestes. Agamemnon, having been murdered by his wife Clytemnestra, Avas avenged by their son Orestes, who killed his mother for the murder of his father. For this act he was prosecuted before the tribunal of the gods by the Erinnyes, whose function it was to punish those who shed the blood of relatives. In his defence, Orestes asks them why they did not punish Clytemnestra for the murder of Agamemnon ; and when they reply that marriage does not constitute blood relationship, — ' She was not the kindred of the ' man whom she slew,' — he pleads that by the same rule they cannot touch him^ because a man is a relation to his father, but not to his mother. This view, though it seems to us so unnatural, was supported by Apollo and Minerva, and being adopted by the majority of the gods, led to the acquittal of Orestes. Hence we see that the views prevalent on relation- ship — views by which the whole social organisation is so profoundly affected — are by no means the same among different races, nor uniform at the same histori- cal period. We ourselves still confuse affinity and con- sanguinity ; but into this part of the question it is not my intention to enter : the evidence brought forward in the preceding pages is, however, I think, sufficient to show that children were not in the earliest times ^ Grey's Australia, vol. ii. pp. 226, 236. THE FEESENT SYSTEM 161 regarded as related equally to their father and their mother, but that the natural progress of ideas is, first, that a child is related to his tribe generally ; secondly, to his mother, and not to his father ; thirdly, to his father, and not to his mother ; lastly, and lastly only, that he is related to both. M 162 CHAPTER Y. ON THE DEVELOPMENT OE EELATIONSHIPS. IN the previous chapter I have discussed the ques- tion of marriage as it exists among the lower races of men, and the relation of children to their parents. In the present, I propose to consider the question of rela- tionships in general, and to endeavour to trace up the ideas on this subject from their rudest form to that in which they exist amongst more civilised races. For the facts on which this chapter is based we are mainly indebted to Mr. Morgan, who has collected a great mass of information on the subject, which has recently been published by the Smithsonian Institution. Though I dissent from Mr. Morgan's main conclusions, his work appears to me one of the most valuable contributions to ethnological science which has appeared for many years. ^ It contains schedules, most of which are very complete, giving the systems .of relationships of no less than 139 races or tribes ; and we have, therefore (though there are still many lamentable deficiencies — the Sibe- rians, South Americans, and true Negroes being, for instance, as yet unrepresented), a great body of evidence illustrating the ideas on the subject of relationships which prevail among different races of men. ^ Systems of Consanguinity and Affinity of the Human Family, by L. H. Morgan, 1870. ON TEE DEVELOPMENT OF BELATI0N8HIPS 163 Oar own system of relationships naturally follows from the marriage of single pairs ; and it is, in its gene- ral nomenclature, so mere a description of the actual facts, that most persons tacitly regard it as necessarily general to the human race, with, of course, verbal and unimportant differences in detail. Hence but little in- formation can- be extracted from dictionaries and voca- bularies. They generally, for instance, give words for uncle, aunt, and cousin ; but an uncle may be either a father's brother or a mother's brother, and an aunt may be either a father's sister or a mother's sister ; a first cousin, again, may be the child of any one of these four uncles and aunts ; but practically, as we shall see, these cases are in many races distinguished from one another ; and I may add, in passing, it is by no means clear that we are right in regarding them as identical and equiva- lent. Travellers have, on various occasions, noticed with surprise some special peculiarity of nomenclature which came under their notice ; but Mr. Morgan was the first to collect complete schedules of relationships. The special points which have been observed have, indeed, been generally regarded as mere eccentricities ; but this is evidently not the case, because the principle or prin- ciples to which they are due are consistently carried out, and the nomenclature is reciprocal generally, though not quite without exceptions. Thus, if the Mohawks call a father's brother, not an uncle, but a father, they not only call his son a brother and his grandson a son, but these descendants also use the correlative terms. We must remember that our ideas of relationships are founded on our social system, and that, as other 164 DIFFERENT 8Y8TEM8 OF BELATI0N8HIPS races have very different habits and ideas on this sub= ject, it is natural to expect that their systems of rela- tionship would also differ from ours. I have in the previous chapter pointed out that the ideas and customs with reference to marriage are very dissimilar in dif- ferent races, and we may say, as a general rule, that, as we descend in the scale of civilisation, the family diminishes, and the tribe increases, in importance. Words have a profound influence over thought, and true family-names prevail principally among the highest races of men. Even m the less advanced portions of our own country, we know that collective names were those of the tribe, rather than the family. I have already mentioned that among the Romans the ' family ' was not a natural family in our sense of the term. It was founded,^ not on marriage, but on power. The family of a chief consisted, not of those allied to him by blood, but of those over whom he exercised control. Hence, an emancipated son ceased to be one of the family, and did not, except by will, take any share in his father's ]3roperty ; on the other hand, the wife introduced into the family by marriage, or the stranger converted into a son by adoption, became regularly recognised members of the family, though no blood tie existed. Marriage, again, in Rome, was symbolised by cap- ture or purchase, as among so many of the lower races at the present day. In fact, the idea of marriage among the lower races of men generally is essentially of a different character from ours ; it is material, not spiritual ; it is founded on force, not on love ; the wife ^ See Ortolan's Justinian, p. 126, et seq. CLASSIFICATION OF SYSTFMS 165 is not united with, but enslaved to, her husband. Of such a system traces, and more than traces, still exist in our own country : our customs, indeed, are more advanced, and wives enjoy a very different status in reality, to that which they occupy in law. Among the Redskins, however, the wife is a mere servant to her husband, and there are cases on record in which husband and wife, belonging originally to different tribes, have lived together for years without either caring to acquire the other's language, satisfied to com- municate with one another entirely by signs. It must, however, be observed that, though the Redskin family is constituted in a manner very unlike ours, still the nomenclature of relationships is founded upon it, such as it is, and has no relation to the tribal system, as will presently be shown. Mr. Morgan divides the systems of relationships into two great classes, the descriptive and the classifica- tory, which he regards as radically distinct. The first, he says (p. 12), ' which is that of the Aryan, Semitic, ^ and Urahan families, rejecting the classification of * kindred, except so far as it is in accordance with the ' numerical system, describes collateral consanguinei, ' for the most part, by an augmentation or combination * of the primary terms of relationship. The second, ^ which is that of the Turanian, American Indian, and ^ Malayan families, rejectmg descriptive phrases in every ^instance, and reducing consanguinei to great classes ' by a series of apparently arbitrary generalisations, ' applies the same terms to all the members of the same ' class. It thus confounds relationships, which, under ■' the descriptive system, are distinct, and enlarges the 166 NATURE OF THE EVIDENCE ' signification both of the primary and secondary terms ' beyond their seemingly appropriate sense.' While, however, I fully admit the immense diifer- ence between, say, our English system and that of the Kingsmill Islanders, as shown in Table I.^ opposite, they seem to me to be rather the extremes of a series than founded on different ideals. Mr. Morgan admits that systems of relationships have undergone a gradual development, following that of the social condition ; but he also attributes to them great value in the determination of ethnological affini- ties. I am not sure that I exactly understand his views as to the precise bearing of these two conclusions in relation to one another ; and I have elsewhere ^ given m}^ reasons for dissenting from his interpretation of the facts in reference to social relations. I shall, therefore, now confine myself to the question of the bearing of systems of relationships on questions of ethnological affinity, and to a consideration of the manner in which the various systems have arisen. As might naturally have been expected, Mr. Morgan's information is most full and complete with reference to the North American Indians. Of these, he gives the terms for no less than 268 relationships in about seventy different tribes. Of these relationships, some are, for our present j^urposes, much more important than others. The most signifi- cant are the following : — 1. Brother's son and dauofhter. 2. Sister's son and daughter. ^ I have constructed this table in a manner which seems to me from Mr. Morgan's schedules, select- more instructive than that adopted ing the relationships which are the by Mr. Morgan, most significant, and arranging them ^ Jour. Anthr. Inst. vol. i. UPS. 12 Kappir Uncle Cousin Son ? Grandctiild Father Brother, B. or Y. Son ? G-randchild Uncle Brother, B. or Y. Son ? Grandchild Aunt Brother, B. or Y. Son ? G-randchild Grandfather Grandmother Son Son Son Son Grandchild Grandchild 13 MOHIGA Uncle Stepbrother Stepchild Stepchild Grandson Stepmother Stepbrother Stepchild Stepchild Grandchild Stepfather Stepfather Stepchild Grandchild Mother Brother, E. Stepchild Stepchild Grandchild Grandfather Grandmothe Stepson Stepson Nephew Sou Grandchild Grandchild the Tamil. HIPS. ^ Eighteen 10 11 Hare Omaha Mother's brother Uncle Cousin Uncle Son Uncle ? Uncle Son Brother, E. i Son Brother, E. Grandson Uncle Aunt Aunt Cousin Nephew Son Grandchild Son Grandchild Son Grandchild Son Grandchild Table I.— SYSTEMS OF RELATIONSHIPS. sister's son, SLS. Brothor's sou's soi Sister's sou's son TWO-MOONTAI> Secuiid littlti ! . ' Brother, E. ur Y. Grfiudsou I Little Motlier, or I Uncle iil*miile 1 Graudsou I Motlier , E. or Y. Brother, B. or Y. Brother, E. or Y. Nephew Graiidehild Little father Brother, E. or Y. Nephew Grandmother Grimdchild mole i Grandson Mother, I Brother. E. or "i ' Grandmother j Neiihew Grandson Srandsin Grandson Mother Mother Mother Brother, E. or T. SSn Brother, E. or y. Brother Bo°y _ Grandfather Grandfather GrandfatuSr Grandmother Grandmother Grandmother Brother, E. or Y. ? Grfuidohild ' Brother, E. or T. 13 U 15 10 17 MOllIOAS H..„. Oa.. Ojibwa (Lake Mieliigim) Kirks Uncle Uncle Uncle Unele ■ Uuele Sf Nephew Gmndson GrandohUd Coujln iSiud §?ptw ; Stepmother Annt ... Amit ' Aunt Stepchild Stepchild Grandson ■s s. ■ Gnvmlsou Uncle stepfather «»» Undo Stepfather Stepchild Brother Nephew Stepson ^■"Tild sr Gmudohild :°N7ptw , • Grandson Mother Aunt Stepmother Stepmother 1 Annt Brother, E. or T. Brother Brother, E. or Y. Stepbrother Co»f . d! N. I The Tehigii aud Caimrese s Tamil nud Fijiflii Table II.— SYSTEMS OF EELATIONSHIPS. ' REDKxnTSS MIOMAOS 6 6 Cow «?:— Gi....dLw... CllEhOKEE HABE OMAHA SAWK ASD F..X OXEIbA OTAWA Mother's brother . . . . Uncle Uncle Uncle Elder broth Uncle Mother's brother Uncle Uncle Uncle Unele Cousin Uncle Stepbrother Grandchild Nephew Nephew Stepbrother Grandson Grandchild UratWr. E. or Y. Nephew Grandchild Grandchild Grandchild M.,tlu-r Aunt Aunt Nei;!;;;;; :; :; d„;.ht^i;^tM.s; loS Kephew Nephew Nephew llew :; Brother Brother iz GraSdehlld Grandchild Nephew ''•^' Son Father Brother Eather Son GmnddnUl ADDRESSING PERSONS BY RELATIONSHIPS 167 3. Motlier's brother. 4. Mother's brother's son. 5. Father's sister. 6. Father's sister's son. 7. Father's brother. 8. Father's brother's son. 9. Mother's sister. 10. Mother's sister's son. 11. Grandfather's brother. 12. Brothers' and sisters' grandchildren. Now let me call attention to the Wyandot system as shown in column 8 of Table I. It will be observed that a mother's brother is called an nncle ; his son a cousin ; his grandson a son when a male is speaking, a nephew when a female is speaking ; his great-grandson a grandson. A father's sister is termed an aunt ; her son a cousin ; her grandson a son ; her great-grandson a grandson. A father s brother is a father ; his son a brother, distinguished, however, by different terms ac- cording as he is older or younger than the speaker ; his grandson a son ; his great-grandson a grandson. A mother's sister is a mother ; ^ her son is a brother, dis- tinguished as before ; her grandson a son when a male is speaking, a nephew when a female is speaking. A grandfather's brother is a grandfather ; and a grand- father's sister is a grandmother. A brother's son is a son when a male is speaking, but a nephew when a * In Madagascar ' first cousins ' that we learn the exact degree of ' are usually termed brother and ' relationship. These secondary fa- * sister, and uncles and aunts father ' thers and mothers seem often to be ^ and mother respectively ; and it is ' regarded with little less affection ' only by asking distinctly of persons ' than the actual parents.' — Sibre^'s ' whether they are " of one father " Madagascar and its People, p. 192. ' or are " uterine brother and sister," 168 ADDRESSING PERSONS BY RELATIONSHIPS female is speaking ; while a sister's son is a nephew when a male is speaking, but a son when a female is speaking. Lastly, brothers' grandchildren and sisters' o-randchildren are called ^grandchildren. This system, at first, strikes one as illogical and in- consistent. How can a person have more than one mother? How can a brother's son be a son, or an uncle's great-grandson a grandson ? Again, while classing together several relationships which we justly separate, it distinguishes between elder and younger brothers and sisters ; and in several cases the relation- ship depends on the sex of the speaker. Since, how- ever, a similar system prevails over a very wide area, it cannot be dismissed as a mere arbitrary or accidental arrangement. The system is, moreover, far from being merely theoretical, in every- day use. Every member of the tribe knows his exact relationship to each other, according to this system ; and this knowledge is kept up by the habit, general among the American tribes, and occurring also elsewhere — as, for instance, among the Esquimaux, the Tamils, Telugus, Chinese, Japanese, Fijians, &c. — of addressing a person, not by his name, but by his relationship. Among the Telugus and Tamils an elder may address a younger by name, but a younger must always use the term for relationship in speaking to an elder. This custom is, probably, connected with the curious superstitions about names ; but, however it may have arisen, the result is that an Indian addresses his neighbour as ' my father,' ' my son,' or ' my brother,' as the case may be : if not related, he says ^ my friend.' Thus the system is kept up by daily use ; nor is 8IMILABITIE8 OF SYSTEM 169 it a mere mode of expression. Although, in many respects, opposed to the existing customs and ideas, it is, in some, entirely consonant with them : thus, among many of the Redskin tribes if a man marries the eldest girl in a family he can claim in marriage all the others as they successively come to maturity ; this custom exists among the Shyennes, Omahas, lowas, Kaws, Osages, Blackfeet, Crees, Minnitarees, Crows, and other tribes. I have already mentioned that among the Redskins, generally, the mother's brother exercises a more than paternal authority over his sister's children. I shall have occasion to refer again to this remarkable exaggeration of avuncular authority. Mr. Morgan was much surprised to find that a system more or less like that of the Wyandots was very general among the Redskins of North America ; but he was still more astonished to find that the Tamil races of India have one almost identical. A com- parison of columns 8 and 9 in Table I. will show that this is the case, and the similarity is even more striking in Mr. Morgan's tables, where a larger number of rela- tionships is given. How, then, did this system arise? How is it to be accounted for ? It is by no means consonant, in all respects, to the present social conditions of the races in question ; nor does it agree with tribal affinities. The American Indians generally follow the custom of exo- gamy, as it has been called by Mr. M'Lennan — that is to say, no one is permitted to marry within the clan ; and as descent goes in the female line, a man's brother's son, though called his son, belongs to a difi'erent clan ; while his sister's son does belong to the clan, though he 170 BED SKIN 8 AND TAMIL RACES is regarded as a nephew, and consequently as less closely connected. Hence a man's nephew belongs to his clan, but his son belongs to a different clan. Mr. Morgan discusses, at some length/ the conclu- sions to be drawn from the wide extension of this system over the American continent, and its presence also in India. ' The several hypotheses/ he says, ' of accidental ' concurrent invention, of borrowing from each other, ' and of spontaneous growth, are entirely inadequate.' '^ With reference to the hypothesis of independent develop- ment in disconnected areas, he observes that it pos- ^ sesses both plausibility and force.' It has, therefore, he adds, 'been made a subject of not less careful study ' and reflection than the system itself. Not until after ' a patient analysis and comparison of its several forms ^ upon the extended scale in which they are given in ' the tables, and not until after a careful consideration ' of the functions of the system, as a domestic institu- ' tion, and of the evidence of its mode of propagation ' from age to age, did these doubts finally give way, and ' the insufficiency of this hypothesis to account for the ' origin of the system many times over, or even a second ' time, became fully apparent.' And again, ' if the two families — i.e. the Redskin * and the Tamil — commenced on separate continents in ' a state of promiscuous intercourse, havmg such a ' system of consanguinity as this state would beget, of ' the character of which no conception can be formed, ' it would be little less than a miracle if both should ' develop the same system of relationship.' ^ He con- 1 See, for instance, pp. 157, 392, 394, 421, 456, &c. 2 Loc. cit. p. 495. 3 j^Qc^ ^it p. 505. IlEDSKINS AND TAMIL EACES 171 eludes, then, that it must be due to ' transmission with ' the blood from a common original source. If the four 'hypotheses named cover and exhaust the subject, and ' the first three are incapable of explaining the present ' existence of the system in the two families, then ' the fourth and last, if capable of accountmg for its ' transmission, becomes transformed into an established ' conclusion.' ^ That there is any near alliance between the Eedskin and Tamil races would be an ethnological conclusion of great importance. It does not, however, seem to me to be borne out by the evidence. The Pijian system, with which the Tongan is almost identical, is very instructive in this respect, and scarcely seems to have received from Mr. Morgan the consideration which it merits. Now, columns 9, 10, and 11 of Table I. show that the Fijian, Tongan, and Australian systems are almost identical with the Tamil. "'^ If, then, this similarit}^ is, in the case of the Tamil, proof of close ethnological affinity between that race and the Redskins, it must equally be so in reference to the Fijians, the Tongans, and the Australians. It is, however, well known that these races belong to very distinct divisions of mankind, and any facts which prove similarity between these races, however interesting and important they may be as proofs of identity in human character and history, can obviously have no bearing on special ethnological affinities. Moreover, it seems clear, as I shall attempt presently to show, that the Tongans have not used their present system ever since their ancestors "first * Ibid. See also p. 497. Australia the system appears to be ^ In some parts, at any rate, of very similar. 172 MALAYANS— FIJIANS landed on the Pacific Islands, but that it has subse- quently developed itself from a far ruder system, which is still in existence in many of the surrounding islands. I may also observe that the Two- Mountain Iroquois, whose close ethnological affinity with the Wyandot s no one will question, actually agree, as shown by columns 3 and 4 of Table L, more nearly with this ruder Pacific, or, as Morgan calls it, ' Malayan ' system than they do with that of the neighbouring American tribes. For these and other reasons, I think it impossible to adopt Mr. Morgan's views, either on the causes which have led to the existence of the Tamil system, or as to the ethnological conclusions which follow from it. How, then, have these systems arisen, and how can we account for such remarkable similarities between races so distinct, and so distant, as the Wyandots, Tamils, Fijians, and Tongans ? In illustration of my views on this subject, I have constructed the preceding table (Table L), in which I have given the translation of the native words, and, when one word is used for several relationships, have translated it by the simplest. Thus, in Fijian, the word ' Tamanngu ' — literally ' Tama my,' the suffix ' nngu ' meaning ' my ' — is applied, not only to a father, but to a father's brother ; hence, as the father is the more important, we say that they call a father's brother a father. In many cases the origins of the terms for relation- ships are undeterminable ; I shall discuss some in a subsequent chapter. Others, however, have so far withstood the wear and tear of daily use as to be still traceable. Thus, in Polish, the word for my great-uncle is. NOMENGLATCfBE OF RELATIONSHIPS 173 literally, ' my cold grandfather ; ' the word for ' wife ' among the Crees is ' part of myself ; ' that for husband, among the Choctas, is ' he who leads me ; ' a daughter- in-law among the Delawares is called ' Nah-hmn,' literally, ' my cook ; ' for which ungracious expression, however, they make amends by their word for husband or wife, ' Wee-chaa-oke,' which is, literally, ' my aid ' through life.' It might, a priori^ be supposed that the nomencla- ture of relationships would be greatly affected by the question of male or female descent. This, however, does not appear to be the case. Under a system of female descent, combined with exogamy, a man must marry out of his tribe ; and, as his children belong to their mother's tribe, it follows that a man's children do not belong to his tribe. On the other hand, a woman's children, whomsoever she may marry, belong to her tribe. Hence, while neither a man's nor his brother's children belong to the same tribe as himself, his sister' & children must do so, and are, in consequence, often regarded as his heirs. In fact, for all practical purposes, among many of the Redskin and other tribes, a man's sister's sons are regarded as his children. As we have already seen, this remarkable custom prevails, not only among the Redskins, but also in various other parts of the world. As regards the native tribes of North America, it may also be laid down as a general proposition that the mother's brother exercises more authority over his sister's children than does their father. He has a recognised right to any property they may acquire, if he choose to exercise it ; he can sive orders which a true father would not 174 NOMENGLATUBE OF EELATIONSHIPS venture to issue ; he arranges the marriages of his nieces, and is entitled to share in the price paid for them. The same custom prevails even among the semi- civilised races ; for instance, among the Choctas the uncle, not the father, sends a boy to school. Yet among these very tribes a man's sister's son is called his nephew, while his brother's son is called his son. Thus, although a man's mother's brother is called an uncle, he has, in reality, more power and responsi- bility than the true father. The true father is classed with the father's brother and the mother's sister ; but the mother's brother stands by himself, and, although he is called an uncle, he exercises the real parental power, and on him rests the parental responsibility. In fact, while the names of relationships follow the mar- riage customs, the ideas are guided by the tribal organisation. Hence we see that not only do the ideas of the several relationships, among the lower races of men, diiFer from ours ; but the idea of relationship, as a whole, is, so to say, embryonic, and subsidiary to that of the tribe. In fact, the idea of relationship, like that of mar- riage, was founded, not upon duty, but upon power. Only with the gradual elevation of the race has the latter been subordinated to the former. I have endeavoured to illustrate the various sys- tems of relationships by Table I. (opposite p. 166), which begins with the Hawaiian, or Sandwich Island system. The Hawaiian language is rich in terms for rela- tionships. A grandparent is ' Kupuna,' a parent is TBE HAWAIIAN SYSTEM 175 ' Makua,' a child ' Kaikee/ a son-in-law, or daughter-in- law, is ' Hunona,' a grandchild ' Moopuna ; ' brothers in the plural are ' Hoahanau ; ' a brother-in-law, or sister- in-law, is addressed as ' Kaikoeke ; ' there are special words for brother and sister according to age and sex ; thus, a boy speaking of an elder brother, and a girl speaking of an elder sister, use the term ' Kai-kuuana ; ' a boy speaking of a younger brother, or a girl of a younger sister, uses the word ' Kaikaina ; ' a boy speak- ing of a sister calls her ' Kaikuwahine,' while a sister calls a brother, whether older or younger, ' Kai-kuaana.' They also recognise some relationships for which we have no special terms ; thus, an adopted son is ' Hanai ; ' the parents of a son-in-law, or daughter-in-law, are ' Puluna ; ' a man addresses his brother-in-law, and a woman her sister-in-law, as ' Punalua ; ' lastly, the word ' Kolea ' has no corresponding term in English. It will be observed that these relationships are con- ceived in a manner entirely unlike ours ; we make no difference between an elder brother and a younger brother, nor does the term used depend on the sex of the speaker. The contrast between the two systems is, however, much more striking when we come to con- sider the deficiencies of the Hawaiian system, as indi- cated in the nomenclature. Thus, there is no word for cousin, none for uncle or aunt, nephew or niece, son or daughter ; nay, while there is a word indicating parent, there is said to be none for father or even for mother. The principal features of this interesting system, so elaborate, yet so rude, are indicated in the second column of Table I. I have already mentioned that there is no word for father or mother : for the latter 176 TEE HAWAIIAN SYSTEM they say ' parent female,' for the former, ' parent male ; ' but the term ' parent male ' is not confined to the true parent, but is applied equally to the father's brother and mother's brother ; while the term ' parent female ' denotes also father's sister and mother's sister. Thus, uncleships and auntships are ignored, and a child may have several fathers and several mothers. In the suc- ceeding generation, as a man calls his brother's and sister's children his children, so do they regard him as their father. Again, as a mother's brother and a father's brother are termed ' parents male,' a mother's sister and father's sister, ' parents female,' their sons are regarded as brothers, and their daughters as sisters. Lastly, a man calls the children of these constructive brothers and sisters, equally with those of true brothers and sisters, his children ; and their children, his grand- children. The term ' parent male,' then, denoted not only a man's father, but also his father's brother and mother's brother ; while the term ' parent female ' in the same way denotes not only a man's mother, but also his mother's sister and father's sister. There are, in fact, six classes of parents : three on the male side, and three on the female. The term, ' my elder brother,' or ' younger brother,' as the case may be,^ stands also for my ^ Among tlie Australians, near ^ for brother and sister always involve Sydney, ' brothers and sisters speak ' the distinction of elder or younger/ ' of one another by titles that indicate — Ridley, Journ. Anthr. Inst., vol. ' relative age ; that is, their words xxvi. p. 266. TH:E] HAWAIIAN SYSTmi 177 Mother's brother's son, Mother's sister's son, Father's brother's son, Father's sister's son, while their children, again, are all my grandchildren. Here there is a succession of generations, but no family. We find here no words for true fathers and mothers, uncles or aunts, nephews or nieces, but only Grandparents, Parents, Brothers and sisters, Children and Grandchildren. This nomenclature is actually in use, and so far from having become obsolete, being in Fiji combined with inheritance through females, and the custom of im- mediate inheritance, gives a nephew the right to take his mother's brother's property : a right which is frequently exercised, and never questioned, although apparently moderated by custom. It will very likely be said, that though the word ^son,' for instance, is used to include many who are really not sons, it by no means follows that a man should regard himself as equally related to all his so-called ' sons.' And this is true, but not in the manner which might have been a priori expected. For, as many among the lower races of men have the system of inheritance through females, it follows that they consider their sister's children to be in reality more nearly related to them, not only than their brother's children, but even than their very own children. Hence we see that these terms, son, father, mother, &c., which to us imply relationship, have not 178 THE HAWAIIAN SYSTEM strictly, in all cases, this significance, but rather imply the relative position in the tribe. Additional evidence of this is afforded by the re- strictions on marriage which follow the tribe, and not the terms. Thus the customs of a tribe may, and con- stantly do, forbid marriage with one set of constructive sisters or brothers, but not with another. The system shown in column 2 is not apparently confined to the Sandwich Islands, but occurs also in other islands of the Pacific. Thus, the Kingsmill system, as shown in column 3, is essentially similar, though they have made one step in advance, having devised words for father and mother. Still, however, the same term is applied to a father's brother and a mother's brother as to a father : and to a father's sister and a mother's sister as to a mother ; consequently, first cousins are still called brothers and sisters, and their children and grandchildren are called children and grandchildren. The habits of the South Sea Islanders, the entire absence of privacy in their houses, their objection to sociable meals, and other points in their mode of life, have probably favoured the survival of a very rude system, though the nomenclature is not in accordance with their present social and family relations, but in- dicates a time when these were less developed than at present. We know as yet no other part of the world where the nomenclature of relationships is so primitive. Yet a near approach is made by the system of the Two-Mountain Iroquois, which is, perhaps, the lowest yet observed in America. In this tribe a brother's children are still regarded as sons, and a woman calls AMEBIC AN SYSTEMS 179 her sister's children her sons ; a man, however, does not regard his sister's children as his children, but dis- tinguishes them by a special term ; they become his nephews. This distinction between relationships, which we regard as identical, has its basis in, and is in accord- ance with, American marriage customs. Unfortunately I have no means of ascertaining whether these rules prevail among the tribes in question, but they are so general among the Indians of North America that in al] probability it is the case. One of these customs is that if a man marries a girl who has younger sisters, he thereby acquires a right to those younger sisters as they successively arrive at maturity.^ This right is w^idely recognised, and frequently acted upon. The first wife makes no objection, for the work which fell heavily on her is divided with another, and it is easy to see that, when polygamy prevails, it would be uncom- plimentary to refuse a wife who legally belonged to you. Hence a woman regfards her sister's sons as her sons : they may be, in fact, the sons of her husband : any other hypothesis is uncomplimentary to the sister. Throughout the North American races, therefore, we shall find that a woman calls her sister's children her children ; in no case does she term them nephews or nieces, though in some few tribes she distinguishes them from her own children by calling them step- children. Another general rule in America, as elsewhere, is that no one may marry within his own clan or family. It has been shown in the previous chapter that this rule is not only general in North America, but widely ^ Arcli£eol. Amer. vol. ii. p. 109. n2 180 AMEEIOAN SYSTEMS prevalent elsewhere. The result is, that as a woman and her brother belong to one family, her husband must be chosen from another. Hence, while a man's father's brother and sister belong to his clan, and his mother's sister, being one of his father's wives, is a member of the family — one of the fire- circle, if I may so say — the mother's brother is necessarily neither a member of the fire- circle nor even of the clan. Hence, while a father's sister and mother's sister are called mother, and a father's brother father, in most of the Eedskin tribes the marriage rules exclude the mother's brother, who is accordingly distinguished by a special term, and in fact is recognised as uncle. Thus we can understand how it is that of the six classes of parents mentioned above, the mother's brother is the first to be distin- guished from the rest by a special name. It will, how- ever, be seen by the table that among the Two-Mountain Iroquois a mother's brother's son is called brother, his grandson son, and so on. This shows that he also was once called ' father,' as in Polynesia, for in no other man- ner can such a system of nomenclature be accounted for. All the other relationships, as given in the table, are, it will be seen, identical with those recognised in the Hawaiian and Kingsmill systems. Thus, in two re- spects only, and two, moreover, which can be satisfac- torily explained by their marriage regulations, do the Two-Mountain Iroquois difi'er from the Pacific system. It is true that these two points of difi'erence involve some others not shown in the table. Thus, while a woman's father's sister's daughter's son is her son, a man's father's sister's daughter's son is his nephew, because his father's sister's daughter is his sister, and THE MICMAC SYSTEM 181 Ms sister's son, as already explained, is his nephew. It should also be added that the Two- Mountain Iroquois show an advance, as compared with the Hawaiian system, in the terms relating to relationships by marriage. The Micmac system, as shown in column 5, is in three points an advance on that of the Two- Mountain Iroquois. Not only does a man call his sister's son his nephew, but a woman applies the same term to her brother's son. Thus, men term their brother's sons ^ sons,' and their sister's sons ' nephews ; ' while women, on the contrary, call their brother's sons ' nephews,' and their sister's sons ' sons ; ' obviously because there was a time when, though brothers and sisters could not marry, brothers might have their wives in common, while sisters, as we know, habitually married the same man. It is remarkable also that a father's brother and a mother's sister are also distinguished from the true father and mother. In this respect the Micmac system is superior to that prevailing in most other Redskin races. For the same reason, not only is a mother's brother termed an uncle, but the father's sister is no longer called a mother, being distinguished by a special term, and thus becomes an aunt. The social habits of the Eedskins, which have already been briefly alluded to, sufficiently explain why the father's sister is thus distinguished, while the father's brother and mother's sister are still called respectively father and mother. Moreover, as we found among the Two-Mountain Iro- quois that although the mother's brother is recognised as an uncle, his son is still called brother, thus pointing back to a time when the father's brother was still called 182 BUBMESE AND JAPANESE 8Y8TEM8 father ; so here we see that though the father's sister is called aunt, her son is still regarded as a brother ; indicating the existence of a time when, among the Mic- macs, as among the Two-Mountain Iroquois, a father's sister was termed a mother. It follows as a consequence that, as a father's brother's son, a mother's brother's son, a father's sister's son, and a mother's sister's son, are considered to be brothers, their children are termed sons by the males ; but as a woman calls her brother's son a nephew, so she applies the same term to the sons of the so-called brothers. If the system of relationship be subject to gradual growth, and approach step by step towards perfection, we should naturally expect that, from differences of habits and customs, the various advances would not among all races follow one another in precisely the same order. Of this the Micmacs and Wyandots afford us an illustration. While the latter have, on the whole, made most progress, the former are in advance on one point ; for though the Micmacs have distinguished a father's brother from a father, he is among the Wyandots still termed a father ; on the other hand, the Wyandots call a mother's brother's son a cousin, while among the Micmacs he is still termed a brother. Here we may conveniently consider two Asiatic nations — the Burmese and the Japanese — which, though on the whole considerably more advanced in civilisation than any of the foregoing races, yet appear to be singu- larly backward in their systems of family nomenclature. I will commence with the Burmese. A mother's brother is called either father (great or little) or uncle ; his son is regarded as a brother ; his grandson as a nephew ; THE WYANDOT 8Y8TEM 183 his great-grandson as a grandson. A father's sister is an aunt ; but her son is a brother, her grandson is a son, and her great-grandson a grandson. A father's brother is still a father (great or little) ; his son is a brother; his grandson a nephew ; and his great-grand- son a grandson. A mother's sister is a mother (great or little) ; her son is a brother ; her grandson a nephew ; and her great-grandson a grandson. Grandfathers' brothers and sisters are grandfathers and grandmothers. Brothers' and sisters' sons and daughters are recognised as nephews and nieces, whether the speaker is a male or female ; but their children again are still classed as grandchildren. Among the Japanese a mother's brother is called a ^ second little father ; ' a father's sister a ' little mother ' or ' aunt ; ' a father's brother a ' little father ' or ' uncle ; ' and a mother's sister a ' little mother ' or ' aunt.' The other relationships shown in the table are the same as among the Burmese. The Wyandots, descendants of the ancient Hurons, are illustrated in the eighth column. Their system is somewhat more advanced than that of the Micmacs. While, among the latter, a mother's brother's son, and a father's sister's son, are called brothers, among the Wyandots they are recognised as cousins. The children of these cousins, however, are still by males called sons, thus reminding us that there was a time when these cousms were still regarded as brothers. A second mark of progress is, that women regard their mother's brother's grandsons as nephews, and not as sons, though the great-grandsons of uncles and aunts are still, in all cases, termed grandsons. 184 THE WYANDOT SYSTEM I crave particular attention to this system, which may be regarded as the typical system of the Redskins/ althousrh, as we have seen, some tribes have a ruder nomenclature, and we shall presently allude to others which are rather more advanced. A mother's brother is termed uncle ; his son is a cousin ; his grandson is termed nephew when a woman is speaking, son in the case of a male. In either case, his grandson is termed grandson. A father's sister is an aunt, and her son a cousin ; but her grandson and great-grandson are termed, respectively, son and grandson, thus reminding us that there was a time when a father's sister was re- garded as a mother. A father's brother is called father ; his son, brother ; his grandson, son ; and his great- grandson, grandson. A mother's sister is a mother, her son is a brother, her grandson is called nephew by a female, son by a male ; her great-grandson is, in either case, called grandson. A grandfather's brother and sister are called grandfather and grandmother respectively. A brother's son is called son by a male, and nephew by a female, while a sister's son is called nephew by a male, and son by a female, the reasons for which have been already explained. Lastly, brothers' son's sons and daughters, sisters' son's sons and daughters, are all called grandsons and granddaughters. Thus we see that in every case the third generation returns to the direct line. The two following columns represent the Tamil and ^ The Peruvian system appears, been very similar, in some of its from the vocabularies given in most essential features, to that of Mr. Clements Markham's Quichua the Wyandots. Grammar and Dictionary, to have TEE TAMIL AND FIJIAN SYSTEMS 185 rijian systems, with wliich also that of the Tonga Islands very closely agrees. I have already called attention to this, and given my reasons for being unable to adopt the explanation suggested by Mr. Morgan. It will be observed that the only differences shown in the table between the system of these races and that of the Wyandots, are, firstly, that the mother's brother's grandson is regarded among the Wyandots as a nephew by males, and as a son by females ; while in the Tamil and Fijian system the reverse is said to be the case, " -and he is termed son by males, and nephew by females. Secondly, that the father's sister's grandson is regarded -as a son among the ^^yandots, while in the Tamil and Fijian system he is, when an uncle is speaking, recognised as a nephew. The latter difference merely indicates that the Tamil and Fijian systems are slightly more advanced than the Wyandot. The other difference is more diffi- cult to understand. But though the Redskin, Tamil, and Fijian sys- tems, differing as they do from ours in many ways, which at first seem altogether arbitrary and unaccount- able, agree so remarkably with one another, we find, also, in some cases, remarkable differences among the Redskin races themselves. These differences affect principally the lines of the mother's brother and father's sister. This is natural. They are the first to be dis- tinguished from true parents, and new means have, therefore, to be adopted to distinguish the relationships thus recognised. In several cases other old terms were tried, with very comical results. These modes of over- coming the difficulty were so unsatisfactory, that, by the time a father's sister's son was recognised as a 186 BEMAEKABLE TEEMS IN USE cousin, tlie necessity for the creation of new terms seems to have been generally felt. Table II. shows, as regards fourteen tribes, the re- sult of the attempt to distinguish these relationships. Taking, for instance, the line which gives the terms in use for a mother's brother's grandson, we find the following, viz. son, stepbrother, grandson, and grand- child, stepson, and uncle ; in the case of a father's sister's grandson (male speaking), we have grandchild, son, stepson, brother, and father ; when a female is speaking, grandchild, son, nephew, brother, and father. Thus, for this single relationship we find six terms in use, and a difference of three generations, viz. from grandfather to son. At first the use of such terms seems altogether arbitrary, but a further examination will show that this is by no means the case. Column 2 gives the system of the Kedknives, one of the most backward tribes on the American continent as regards their nomenclature of relationships. Here, though a mother's brother and a father's sister are, respectively, uncle and aunt, their children are regarded as brothers, their grandchildren as sons, and their great-grandchildren as grandsons. The Munsee system shows a slight advance. Here, though the women call their sister's sons their sons, the males, on the contrary, term them nephews, and, consequently, apply the same term to their mother's brother's daughter's son, and their father's sister's daughter's son ; because, as in the preceding case, mother's brother's daughters, and father's sisters daughters, are termed sisters. The Micmacs (column 3) show another step in advance. Here, not only does a man call his sister's son nephew, EXPLANATION OF THE TEBM8 18/ but, in addition, a woman applies the same term to her brother's son ; consequently, not only a mother's brother's daughter's son, if a male is speaking, but a mother's brother's son's son, if a female is speaking, and the corresponding relations on the side of the father's sister, are termed nephews. Among the Delawares a mother's brother's son, and father's sister's son, are distinguished from true brothers by a term corresponding to ' stepbrother.' They appear to have also felt the necessity of distinguishing a step- brother's son from a true son ; but, having no special term, they retain the same word, thus calling a step- brother's son a stepbrother. This principle, as we shall see, is followed by several other tribes, and has produced the most striking inconsistencies shown in the table. We find it again among the Crows, where a father's sister is called mother, her daughter again, mother ; but as her son cannot of course be a mother, he is called ' father.' The same system is folio v^ed by the Pawnees, as shown in columns 7 and 8 ; and the Grand Pawnees carry it a generation lower, and call their father's sister's grandson on the male side ' father ; ' a father's sister's daughter's son is, however, called a brother. Among the Cherokees we find this principle most thoroughly carried out, and a father's sister's grandson is also called a father. This case is the more interesting, because the circumstance which produced the system is no longer in existence ; for, as will be seen, a father's sister is called an aunt. It is not at first obvious that a father's sister being called a mother would account for her son being called a father ; but, with the Crow and Pawnee systems before us, we see that the Cherokees 188 SYSTEMS OF TEE OMAHAS, ONEIDAS, OTAWAS could not call their father's sister's sons ' fathers,' unless there had been a time when a father's sister was regarded as a mother. The Hare Indians supply us with a case in which, mother's brothers and father's sisters being distin- guished from fathers and mothers, their children are no longer termed brothers, but are distinguished as cousins ; while their grandchildren and great-grandchildren, on the contrary, are still termed sons and grandsons. So far as the relationships shown in the table are concerned, the system of the Omahas, and of the Sawks and Foxes, is identical. A mother's brother is an uncle, and, for the reason already pointed out in the case of the Delawares, his sons and son's sons, and even son's grandsons, are also termed grandsons. His daughter's sons, on the contrary, retain the old name of brother. A father's sister is an aunt, her children are nephews, and the descendants of these nephews are grandchildren. Among the Oneidas, a father's brother is an uncle, and his son is a cousin ; his son's sons, however, are still sons. His daughter's son is a son, when a female is speaking ; but, for the reason already explained in the case of the Munsees, males term them nephews. The relationships connected with a father's sister are dealt with in a similar manner, except that a father's sister is still called mother. The Otawa system resembles the Micmac, and is formed on the same plan, being, however, somewhat more advanced, inasmuch as the children of uncles and aunts are recognised as cousins, and a man calls his cousin's son, not his son, but his stepson. The Ojibwa THE KAFFIB 8Y8TFM 189 system is the same, except that a woman also calls her mother's brother's daughter's son, and father's sister's daughter's son, her stepson, instead of her son. In some of the relationships by marriage the same causes have led to even more striking differences. Thus, a woman generally calls her father's sister's daughter's husband her brother-in-law ; but among the Missouri and Mississippi nations her son-in-law ; among the Minnitarees, the Crows, and some of the Chocta clans, her father ; among the Cherokees, her step-parent ; the Eepublican Pawnees, and some of the Choctas, her grandfather ; and among the Tukuthes, her grandson ! Having thus pointed out the curious results to which some of the lower races have been led in their attempts to distinguish relationships, and endeavoured to explain those shown in Table II., I will now return to the main argument. The Kaffir (Amazulu) system is given in column 12, Table I. Here, for the first time, we find the father's brother regarded as an uncle, and the mother's sister as an aunt. In other respects, however, the system is not more advanced than the Tamil, Fijian, or Wyandot. The mother's brother is called uncle ; ^ his son, cousin ; his grandson, son ; and his great-grand- son, grandchild. A father's sister, quaintly enough, is called father, the Kaffir word for which, uhaba, closely resembles ours. His son, however, is called brother ; his grandson, accordingly, son ; his great-grandson^ grandchild. A father's brother, as already mentioned, is uncle ; but, as before, his son is called brother ; his ^ It iS; howeyer, significant that he calls his sister's sons ^sons/ and not nephews. 190 TED M0EEGAN8 grandson, son ; and his great-grandson, grandson. So, also, a mother's sister is an aunt, but lier son is a brother ; her grandson, a son ; and her great-grandson, a grandson. As in all the preceding cases, grand- fathers' brothers and sisters are considered as, re- spectively, grandfathers and grandmothers. Brothers' sons and sisters' sons are called sons, and, lastly, their sons again are grandsons. ■ Excepting in the case of nephews, this system, there- fore, closely resembles the Tamil, Fijian, and Wyandot ; the other principal differences being a more correct nomenclature of uncles and aunts. Column 13, Table I., exhibits the nomenclature in use among the Mohegans, whose name signifies ' sea- side people,' from their geographical position on the Hudson and the Connecticut. They belong to the great Algonkin stock. Here, for the first time, a dis- tinction is introduced between a father and a father's brother. The latter, however, is not recognised as an uncle ; that is to say, a father's brother and a mother's brother are not regarded as equivalent relationships, but the former is termed stepfather. This distinguish- ing prefix is the characteristic feature ; and, as will be seen, we find the terms stepmother, stejDbrother, and stepchild (to the exclusion of cousin), as natural con- sequences of the stepfather ship. Still, the mother's sister remains a mother, and her son a brother ; and the derivation of this system from one similar to those already considered is, moreover, indicated by the fact that the members of the third generation are still regarded as grandchildren. The Crees and the Ojibwas, or Chippewas (of Lake I M0HEGAN8~GBUE8—GHIPPEWAS 191 Michigan), who also belong to the great Algonkin stock, resemble the Mohegan in the use, though with some minor differences, of the prefix ' step-', a device which occurs also in a more complicated form among the Chinese. In some points, however, they are rather more advanced, and, in fact, these tribes possess the highest system of relationship yet recorded among the Redskins of North America. A mother's brother is an uncle, and his son is a cousin ; as regards his grandson, the tendency to the use of different terms, according as the speaker is a male or female, shows itself in the use by the former of the term stepson, where the latter say nephew as in some of the ruder tribes. In both cases, mothers' brothers' great-grandchildren are called grand- children. A father's sister is an aunt, and the nomen- clature with reference to her descendants is the same as in the case of the mother's brother. A father's brother is a stepbrother ; his son is still called a brother by males among the Crees, but is called stepson by the Ojibwas ; the other relationships in this line being the same as in the case of the mother's brother and father's sister. No Redskin regards his mother's sister as an aunt ; but the Crees and Ojibwas distinguish her from a true mother by the term stepmother, and her descendants are addressed by the same terms as those of the father's brother. The grandfather's brothers and sisters are called grandfathers and grandmothers. As before, brothers' sons, when a female is speaking, and sisters' sons, when a male is speaking, are called nephews ; while brothers' sons, when a male is speaking, and sisters' sons, when a female is speaking, are no longer 192 SUMMARY OF BEBSKIN SYSTEMS regarded as true sons, but are distinguished as stepsons. The grandchildren of these nephews and stepsons are^ however, all termed grandchildren. If, now, we compare this system with that of the Two-Mountain Iroquois, we find that out of twenty- eight relationships given in the table only ten have remamed the same. Of these, two are indicative of progress made by the Two- Mountain Iroquois — namely, the term for mother's brother and sister's son ; the other eight are marks of imperfection still remaining in the Ojibwa nomenclature : points, moreover, not by any means characteristic of American races, but common, also, as we have seen, to the Hawaiian, Kingsmill, Burmese, Japanese, Tongan, Fijian, Kaffir, and Tamil systems ; as we shall also find, to the Hindi, Karen, and Esquimaux ; in fact, to almost all, if not all, barbarous peoples, and even to some of the more advanced races. Column 14, Table L, shows the system of nomen- clature as it exists in Hindi, and it may be added that the Bengali, Marathi, and Gujerathi are essentially the same, although the words differ. All these languages are said to be Sanskrit as regards their words ; abori- ginal, on the contrary, in their grammar. Hindi con- tains 90% of Sanskrit words, Grujerathi as much as 95%. With three or four exceptions, it appears that the terms for relationship may be all of Sanskrit origin. Here, for the first time, we find that a brother's son and a sister's son are termed nephews, whether the speaker is a male or a female. Yet nephews' children are still termed grandchildren. Again, for the first time, the mother's brother, father's brother, mother's HINDOO SYSTEMS^KABENS 198 sister, and father's sister are regarded as equivalent, and the terms for their descendants are similar. The two former — i.e. mother's brother and father's brother, are termed ' uncles ; ' the two latter — i.e. mother's sister and father's sister, are called ' aunts.' Yet, as reo'ards the next generations, the system is less advanced than the Ojibwa, for uncles' sons and aunts' sons are termed brothers ; their grandsons, nephews ; and their great- grandsons, grandsons. It should, however, be observed that, in the first three languages, viz. the Hindi, Ben- gali, and Marathi, besides the simple term ' brother,' the terms * brother through paternal uncle,' ' brother through paternal aunt,' ' brother through maternal uncle,' and ' brother through maternal aunt,' are also in use, and are less cumbersome than our English literal translation would indicate. The system, therefore, is transitional on this point. Lastly, a grandfather's brother is called ' grandfather ; ' a grandfather's sister, ' grandmother.' The Karens are a rude, but peaceful and teachable race, inhabiting parts of Tenasserim, Burmah, Siam, and extending into the southern parts of China. They have been encroached upon and subjected by more powerful races, and are now divided into different tribes, speaking different dialects, of which three are given in Mr. Morgan's tables. Though rude and savage in their mode of life, they are described as extremely moral in their social relations — praise which seems to be corroborated by their system of relation- ships, as shown in column 17, Table I. Column 18 shows the system of another rude people, belonging to a distinct family of the human o 194 KABENS— ESQUIMAUX race, and inhabiting a distant and very different part of the world. Like the Karens, the Esquimaux are a rude people, but, like them, they are a quiet, peaceable, and moral race. No doubt on some points their ideas differ from ours ; their condition does not admit of much refinement — of any great advance in science or art. They cannot be said to have any religion worthy of the name, yet there is, perhaps, no more moral people on the face of the earth ; none among whom there is less crime ; and it is, perhaps, not going too far to say that there is, as far as I can judge, no race of men which has more fully availed itself of its opportunities. It is most remarkable to find that these two races of men, so distinct, so distant, so dissimilar in their modes of life, without a word in common, yet use systems of relationshijD which, in their essential features, are identical, although by no means in harmony with the existing social condition : in both, uncles and aunts are correctly recognised, and their children regarded as cousins ; their grandchildren, however, are termed nephews, and the children of these so-called nephews are classed, as in all the previous cases, as grand- children. Thus, out of the twenty-eight relationships indicated in the table, the Karens and Esquimaux agree with us in twelve, and differ in sixteen. As regards every one, however, of these sixteen they agree with one another, while in eight they follow the same system as every other race which we have been considering. These facts cannot be the result of chance ; there is one way, and, as it seems to me, one way only, of accounting for them, and that is by regarding them as the outcome of a progressive development, such as INDICATIONS OF PE0GBES8 195 that which I have endeavoured to sketch. An examin- ation of the several cases will, I think, confirm this view. The Karen -Esquimaux system is inconsistent with itself in three respects, and precisely where it differs from ours. The children of cousins are termed nephews, which they are not ; the children of nejDhews are re- garded as grandchildren, and a grandfather's brothers and sisters are termed, respectively, grandfathers and grandmothers. The first fact— namely, that a mother's brother's grandsons, and a mother's sister's grandsons, a father's sister's grandsons, and a father's brother's grandsons, are all termed ' nephews ' — clearly points to the existence of a time when a mother's brother and a father's brother were regarded as fathers, a mother's sister and a father's sister as mothers, and their children, consequently, as brothers. The second — namely, that the great-grand- children of uncles and aunts are regarded as grand- children — similarly points to a time when nephews and nieces were termed, and regarded as, sons and daughters, and their children, consequently, as grandchildren. Lastly, why should grandfathers' brothers and grand- fathers' sisters be called grandfathers and grandmothers, unless there was a time when fathers' brothers and sisters were respectively called ' fathers' and 'mothers : ' unless the Karens and Esquimaux once had a system of relationship similar to that which still prevails among so many barbarous tribes, and which, to all appearance, has been gradually modified ? Hence, though the Karens and Esquimaux have now a far more correct system of nomenclature than that of many other races, o 2 196 INCOMPLETENESS OF SYSTEMS we find, even in this, clear traces of a time when these peoples had not advanced in this respect beyond the lowest stage. As already mentioned, the European nations follow, almost without exception, a strictly descriptive system, founded on the marriage of single pairs. The principle is, however, departed from in a few rare cases, and in them we find an approach to the Karen-Esquimaux system. Thus, in Spanish, a brother's great-grandson is called ' grandson.' Again, in Bulgarian, a brother's grandson and sister's grandson are called ' Mai vnook mi,' literally ' httle grandson my.' A father's father's sister is termed a grandmother, and a father's father's brother a grandfather, as is also the case in Russian. The French and Sanskrit, alone, so far as I know, among the Aryan languages, have special words for elder and younger brother. Among Ar3^an races the Romans and the Germans alone developed a term for cousin,^ and we ourselves have, even now, no word for a cousin's son. The history of the term ' nephew ' is- also instructive. The word ' nepos,' says Morgan,^ ^ among the Romans, as late as the fourth century, was ' applied to a nephew as well as a grandson, although ' both " avus " and " avunculus " had come into use. ' Eutropius, in speaking of Octavianus, calls him the 'nephew of Caesar, " Csesaris nepos." (Lib. vii. c. i.) ' Suetonius speaks of him as " sororis nepos " (Caesar, ' c. Ixxxiii.), and afterwards (Octavianus, c. vii.) de- ' scribes Caesar as his great-uncle, '' major avunculus," ^ So that of many nations it may be said, literally as well as figuratively, that ' les nations n'ont pas de cousins.' ^ Log. cit. p. 35. EYIBENGB OF FBOGBESS 197 ' in which he contradicts himself. When " nepos " was ^ finally restricted to grandson, and thus became a strict ' correlative of " avus," the Latin language was with- ^ out a term for nephew, whence the descriptive phrase, * " Fratris vel sororis filius." In English, "nephew" ' was applied to grandson, as well as nephew, as late as ^ 1611, the period of King James's translation of the ' Bible. Niece is so used by Shakspeare in his will, in ' which he describes his granddaughter, Susannah Hall, ' as " my niece." ' So that even among the most advanced races we find some lingering confusion about nephews, nieces, and grandchildren. Thus, then, we have traced these systems of relation- ships from the simple and rude nomenclature of the Sandwich Islanders up to the far purer and more correct terminology of the Karens and Esquimaux. I have endeavoured to show that the systems indicated are explicable only on the theory of a gradual improvement and elevation, and are incompatible with degradation ; that as the valves indicate the course of the blood in our veins, so do the terms applied to relationships point out the course of past history. In the first place, the moral condition of the lower races, wherever we can ascertain it, is actually higher than that indicated by the phraseology in use ; and, secondly, the systems themselves are, in almost all cases, inexplicable, except on the hypothesis that they were themselves preceded by still ruder ones. Take, for instance, the case of the Two- Mountain Iroquois ; they call a mother's brother an uncle, but his son they regard as a brother. This is no accident, for 198 EXISTING SYSTEMS INCOMPATIBLE the idea is carried out in the other relationships, and occurs also in other races. On the theory of progress it is easily accounted for : if a father's brother was pre- viously called a father, his son would, of course, be a brother ; and when the father's brother came to be dis- tinguished as an uncle some time would, no doubt, often elapse before the other changes, consequent on this step, would be effected. But how could such a system be accounted for on the opposite theory ? How could a father's brother's son come to be regarded as a brother, if a father's brother had always been termed an uncle ? The sequence of terms for the relationships connected with a father's sister, on the two hypotheses of progress on the one hand, and degradation on the other, may be illustrated as in the Table III. (p. 204). In the first, or lowest stage, the sequence is mother, brother, son, grandson, as in the Sandwich and Two- Mountain Iroquois system. In the next stage, the mother's sister being recognised as an aunt, and the other relationships remaining the same, we have the sequence, aunt, brother, son, grandson, as among the Micmacs. When a brother's son becomes a nephew we have aunt, brother, nephew, grandson, as in the Burmese, Japanese, and Hindi systems. In the next stage, an aunt's son being distinguished as a cousin, we have aunt, cousin, nephew, grandson, as among the Tamils and Fijians. The last two stages would be aunt, cousin, aunt's grandson, grandson ; and, lastly, aunt, cousin, aunt's grandson, aunt's great-grandson. Thus, out of these six stages, five at least actually exist. On the other hand, on the theory of retrogression, WITH TEE THEORY OF DEGEADATION 199 we should commence with the highest system : namely, aunt, cousin, aunt's grandson, and aunt's great-grand- son. The second stage would be, mother, cousin, aunt's grandson, aunt's great-grandson. The third, mother, brother, aunt's grandson, aunt's great-grandson. The fourth, mother, brother, nephew, aunt's great-grand- son. The fifth, mother, brother, son, aunt's great- grandson. And the last, mother, brother, son, grand- son. Thus, it will be observed that, except, of course, the first and last, they have not a stage in common ; and, though there may be some doubt whether the sequence suggested on the second hypothesis is the one which would be followed, it cannot be maintained that we could ever have the systems which would occur in the case of progress as shown in Table IIL, and the first four of which are actually in existence. Whenever, then, the son or daughter of an uncle, or aunt, is termed a brother, as in the case of seven of the races referred to in the table, we may be sure that there was once a time when that uncle, or aunt, was termed a father or mother ; whenever a cousin's son is termed a son, as again in seven races, we must infer, not only that those cousins were once regarded as brothers, but that brothers' sons were once termed sons. Ao;ain, when great-uncles and aunts are termed grandfathers and grandmothers — when great-nephews and nieces are termed grandchildren, as in the case of all the races we have been considering — we have, I submit, good reason to infer that those races must once have had a system of nomenclature as rude as that of the Hawaiians or Kingsmill Islanders. But it may be asked : admitting that the seventeen 200 EVIDENCE OF PB0GBE8S races, illustrated in Table I., are really advancing, are there not cases of the contrary ? The answer is clear : out of the 139 races whose systems of relationship are more or less completely given by Mr. Morgan, there is not one in which evidence of degradation is thus indi- cated. To show this clearly and concisely, I have pre- pared the following table (p. 201). It will be seen that, taking merely the relation of uncles and aunts with reference to their children, there are 207 cases indicating progress. On the other hand, there are four cases, the Cayuda, Onondaga, Oneida, and Mohawks, among whom, while a father's sister is called a mother, her son is called a cousin. These cases, however, are neutra- lised by the fact that the sons of these cousins are called sons. We have, therefore, a very large body of evidence indicating progress, and collected among very different races of men, while there appear to be none which favour the opposite hypothesis. In the preceding chapter, I have endeavoured to show that relationship is, at first, regarded as a matter, not of blood, but of tribal organisation ; that it is, in the second stage, traced through the mother ; in the third, through the father ; and that only in the fourth stage is the idea of family constituted as amongst ourselves. To obtain clear and correct ideas on this subject, it is necessary to know the laws and customs of various races. The nomenclature alone would, in many cases, lead us into error, and, in fact, has often done so. When checked by a knowledge of the tribal rules and customs, it is, however, most interesting and instructive. From this point of view especially, Mr. Morgan's work is of great value. It has been seen, however, that I differ greatly NO EVIBBNGE OF BEGUADATION 201 ;^ oo <7q c o iN oo i> O O lO — -^ -^ - — — ir-; ri ?n . is H q; § iz! ^ ? 8 II OJ n C ^ i g I^ i i rt o 'H •?n 'H ai ^ 13 vt; rS^ ^ ^ i8 US P-< cS 3 Ph ce ;=i s^ ^^3 •T3 & 8 too O o s ^ ^ PI gg II II II II II II II II II be So 6 d 6 d d d ^ 4i ^ ■t:) 13 n=! 13 13 g ^ o d O o O o too toe ^ 13 13 -O 13 13 6 6 6 6 6 6 6 d d 6 6 d d d d d •ns n:3 13 ■73 t:3 ■73 ■-d 13 13 '^ 13 ni 13 13 13 13 o 6 6 6 6 d 6 d d d d d d d d d •ft 50 P p o fi P 1 — 1 P P o P Q o P P o P P O P p o P 1 ^ t (N G^ 1 '^ , , , 1 a p^ § .S .9 .S p S m P) 5 fl d •E3 P 1 o o o o o o o • i-i o OQ o «2 o CQ o t>. 11 II II II II II II o J=° pj PI PI f:j PI P! p f^H o S o o o o <» 02 OQ m 02 02 JIJ r^ 13 13 13 -^3 13 13 ■*± PI rt fl fl PI , p c3 ci ce 03 e3 Cw eg ^ !Lh i^ ^ ;-t rS2 CD hb QC too be be toe •v ^OQ ^ ^ •V •. •s •N c •X ^ •N ^ •N •^ d %H d d d d d d d d d d d d d d * -d 13 ^ HJ 13 13 13 13 13 13 ■^ 13 13 13 13 d o d 6 d d d d d d d d d d d d p ^ P P P P P P P P P P P P P P -;t^ o I— 1 Ti^ Tt< O ^ O Oi , I— 1 CM II ^ ' ' ' '" ' '~ ^ ^ Th 0) •i 1 CO i g o 00 5 ^ rQ 1 o u .9 o o II 1 1:3 13 o o 1 PI "i^ o •s 1 o -i-T a i PI O 0) o .9 8 i 13 II •5 o Pi g o !l PI II 1 11 o CO i 1 1 p" ■| o o II i • sl^la •^ ^ -M pq !l 6 § II 5 -S CO CO PI 7-t 4^ 5 1 g- CQ <1 :§ c3 ilg J3 1 iM -2 CO ^ P ^ (D 2 ^ 1 ^1 S 1 05 •1^ *0Q P .2 ? "oo g §'£ ^ pq § -s 1 1 ^02 « ^ CQ o H J^ 02 OQ % ^02 ^1 1 o 1 O P 1 ^ 1 1 ^( 1 s o -(J o 1^ P O o 1^ 1 202 CONCLUSION from him as to the conclusions to be drawn from the facts which he has so diligently collected. Of course, I do not deny that these facts may, in some cases, indicate ethnological affinities ; but the}^ have not, I think, so great an importance in solving ques- tions of ethnological relationships as he supposes. I do not, however, in any way, undervalue their importance ; they afford a striking evidence in favour of the doctrine of development, and are thus a very interesting and impor- tant contribution to the great problem of human history. Mainly from the materials which he has so laboriously collected, and for which ethnologists owe him an im- mense debt of gratitude, I have endeavoured to show : Firstly, that the terms for, what we call, relation- ships, are, among the lower races of men, mere ex- pressions for the results of marriage customs, and do not comprise the idea of relationship as we understand it ; that, in fact, the connection of individuals inte?' se, their duties to one another, their rights, and the descent of their property, are all regulated more by the rela- tion to the tribe than by that to the family ; that, when the two conflict, the latter must give way. Secondly, that the nomenclature of relationships is, in all the cases yet collected, explainable in a clear and simple manner on the hypothesis of progress. Thirdly, that while two races in the same state of social condition, but of which the one has risen from the lowest known system, the other sunk from the highest, would, necessarily, have a totally different system of nomenclature for relationships, we have not a single instance of such a system as would result from the latter hypothesis. CONCLUSION 203 Fourthly, that some of those races which approxi- mate most nearly to our European system differ from it upon points only explainable on the hypothesis that they were once in a much lower social condition than they are at present. 204 SYSTEMS OF RELATIONSHIP m m w O o o o w o P o P5 1 . ■>-' s ^ fl g R fj o 2 o M ^^^ H .9 CO PI CO PI -|j ^ «3 -M Cv >< ?3 1^ Q rt ?H p ^
  • p! fi. ?e o Zl O ® S Pq ^ o ^ O 4-4- H d3 :»1 fl 02 f^ ^ g o 3 ^1 PI 1 12; O -t— H O -^ 5 ■ . § a •+j q3 -ip CO o P! o 1 a «3 o '^ CB -11 W GQ O « H C5 ■< H PI m ^ o 1 o 1 02 • Pl o OQ 1 ^ • r^ s ^ ^ 03 ^ d) h2 ;^ Is JJ CQ ^02 'jh O) ^ •» ». ^ •N •S .N ■4-3 ee P=H 03 P-i W 02 .5 rP ^ "" ri PI -^ 0? bo .9 ^ a § a-T3 O l>^ P^ O K Eh o o PIH o w H 02 ;>^ w CD -4 , H PJ 02 ?H PI m ^ K c^ o H 1 . 5 f^ si i W o O) j» Pl H ^ rd -M C3 5:^ ■*e fl ^ S o g § s 5^3 g W ^ <1 H o +!» . ■^ 03 p. . ^ ^ g=l o 0) rP O 1 2 S j» Pl g^ %^ Ph ^ m ^ -^ H • ^ . H m ^ G iJS fl Q