Sacred Texts  Shinto  Index  Previous  Next 
Buy this Book at

The Kojiki, translated by Basil Hall Chamberlain, [1919], at



The Japanese of the mythical period, as pictured in the legends preserved by the compiler of the "Records of Ancient Matters," were a race who had long emerged from the savage state, and had attained to a high level of barbaric skill. The Stone Age was forgotten by them—or nearly so,—and the evidence points to their never having passed through a genuine Bronze Age, though the knowledge of bronze was at a later period introduced from the neighbouring continent. They used iron for [25] manufacturing spears, swords, and knives of various shapes, and likewise for the more peaceful purpose of making hooks wherewith to angle, or to fasten the doors of their huts. Their other warlike and hunting implements (besides traps and gins, which appear to have been used equally for catching beasts and birds and for destroying human enemies) were bows and elbow-pads,—the latter seemingly of skin, while special allusion is made to the fact that the arrows were feathered. Perhaps clubs should be added to the list. Of the bows and arrows, swords and knives, there is perpetual mention; but nowhere do we hear of the tools

p. xxxiv

with which they were manufactured, and there is the same remarkable silence regarding such widely spread domestic implements as the saw and the axe, We hear, however, of the pestle and fire-drill, of the wedge, of the sickle, and of the shuttle used in weaving.

Navigation seems to have been in a very elementary stage. Indeed the art of sailing was, as we know from the classical literature of the country, but little practised in Japan even so late as the middle of the tenth century of our era subsequent to the general diffusion of Chinese civilization, though rowing and punting are often mentioned by the early poets. In one passage of the "Records" and in another of the "Chronicles," mention is made of a "two-forked boat" used on inland pools or lakes; but, as a rule, in the earlier portions of those works, we read only of people going to sea or being sent down from heaven in water-proof baskets without oars, and reaching their destination not through any efforts of their own, but through supernatural inter-position. 23

To what we should call towns or villages very little [26] reference is made anywhere in the "Records" or in that portion of the "Chronicles" which contains the account of the so-called "Divine Age." But from what

p. xxxv

we learn incidentally, it would seem that the scanty population was chiefly distributed in small hamlets and isolated dwellings along the coast and up the course of the larger streams. Of house-building there is frequent mention,—especially of the building of palaces or temples for sovereigns or gods,—the words "palace" and "temple" being (it should be mentioned) represented in Japanese by the same term. Sometimes, in describing the construction of such a sacred dwelling, the author of the "Records," abandoning his usual fiat and monotonous style, soars away on poetic wings, as when, for instance, he tells how the monarch of Idzumo, on abdicating in favour of the Sun-Goddess's descendant, covenanted that the latter should "make stout his temple pillars on the nethermost rock-bottom, and make high the cross-beams to the plain of High Heaven." 24 It must not, however, be inferred from such language that these so-called palaces and temples were of very gorgeous and imposing aspect. The more exact notices to be culled from the ancient Shintō Rituals (which are but little posterior to the "Records" and in no wise contradict the inferences to be drawn from the latter) having been already summarized by Mr. Satow, it may be as well to quote that gentleman's words. He says: 25 "The palace of the Japanese sovereign was a wooden hut, with its pillars planted in the ground, instead of being erected upon broad fiat stories as in modern buildings. The whole frame-work, consisting of posts, beams, rafters, door-posts and window-frames, was tied together with cords made by twisting the long fibrous stems of climbing plants, such as Pueraria

p. xxxvi

[paragraph continues] Thunbergiana (kuzu) and Wistaria Sinensis (fuji), The floor must have been low down, so that the occupants of the building, as they squatted or lay on their mats, were exposed to the stealthy attacks of venomous snakes, which were probably far more numerous in the earliest ages when the country was for the most part uncultivated, than at the present day . . . There seems some reason to think that the yuka, here translated floor, was originally nothing but a couch which ran round the sides of the hut, the rest [27] of the space being simply a mud-floor, and that the size of the couch was gradually increased until it occupied the whole interior. The rafters projected upward beyond the ridge-pole, crossing each other as is seen in the roofs of modem Shin-tau temples, whether their architecture be in conformity with early traditions (in which case all the rafters are so crossed) or modified in accordance with more advanced principles of construction, and the crossed rafters retained only as ornaments at the two ends of the ridge. The roof was thatched, and perhaps had a gable at each end, with a hole to allow the smoke of the wood-fire to escape, so that it was possible for birds flying in and perching on the beams overhead, to defile the food, or the fire with which it was cooked." To this description it need only be added that fences were in use, and that the wooden doors, sometimes fastened by means of hooks, resembled those with which we are familiar in Europe rather than the sliding, screen-like doors of modem Japan. The windows seem to have been mere holes. Rugs of skins and rush matting were occasionally brought in to sit upon, and we even hear once or twice of "silk rugs" being used for the same purpose by the noble and wealthy.

p. xxxvii

The habits of personal cleanliness which so pleasantly distinguish the modern Japanese from their neighbours, in continental Asia, though less fully developed than at present, would seem to have existed in the germ in early times, as we read more than once of bathing in rivers, and are told of bathing-women being specially attached to the person of a certain imperial infant. Lustrations, too, formed part of the religious practices of the race. Latrines are mentioned several times. They would appear to have been situated away from the houses and to have generally been placed over a running stream, whence doubtless the name for latrine in the Archaic Dialect, kaha-ya i.e. "river house." A well-known Japanese classic of the tenth century. the "Yamato Tales," 26 tells us indeed that "in older days the people dwelt in houses raised on platforms built out on the river Ikuta," and goes on to relate a story which presupposes such a method of architecture. 27 A passage in the account of the reign of the Emperor Jim-mu which occurs both in the "Records" and in the "Chronicles," and another in the reign of the Emperor Sui-nin occurring in the "Records" only, might be interpreted so as to support this statement. 28 But both are extremely obscure, and beyond the fact that people who habitually lived near the water may have built their houses after the aquatic fashion practised in different parts of the world by certain savage tribes both ancient and modem, the present writer is not aware of any authority for the assertion that they actually

p. xxxviii

did so except the isolated passage in the "Yamato Tales" just quoted.

A peculiar sort of dwelling-place which the two old histories bring prominently under our notice, is the so called "Parturition-house,"—a one-roomed hut without windows which a woman was expected to build and retire into for the purpose of being delivered unseen. 29 It would also appear to be not unlikely that newly-married couples retired into a specially built hut for the purpose [29] of consummating the marriage, and it is certain that

p. xxxix

for each sovereign a new palace was erected on his accession.

Castles are not distinctly spoken of till a period which, though still mythical in the opinion of the present writer, coincides according to the received chronology with the first century B. C. We then first meet with the curious term "rice-castle," whose precise signification is a matter of dispute among the native commentators, but which, on comparison with Chinese descriptions of the Early Japanese, should probably be understood to mean a kind of palisade serving the purpose of a redoubt, behind which the warriors could ensconce themselves. 30 If this conjecture be correct, we have here a good instance of a word, so to speak, moving upward with the march of civilization, the term, which formerly denoted something not much better than a fence, having later come to convey the idea of a stone castle.

To conclude the subject of dwelling-places, it should be stated that cave-dwellers are sometimes alluded to. The legend of the retirement of the Sun-Goddess into a cavern may possibly suggest to some the idea of an early period when such habitations were the normal abodes of the ancestors of the Japanese race. 31 But at the time when the national traditions assumed their present shape, such a state of things had certainly quite passed away, if it ever existed, and only barbarous Ainos and rough bands of robbers are credited with the construction of such primitive retreats. Natural caves

p. xl

[paragraph continues] (it may be well to state) are rare in Japan, and the caves that are alluded to were mostly artificial, as may be gathered from the context.

The food of the Early Japanese consisted of fish and of the flesh of the wild creatures which fell by the hunter's arrow or were taken in the trapper's snare,—an animal diet with which Buddhist prohibitions had not yet interfered, as they began to do in early historical times. Rice is the only cereal of which there is such mention made as to place it beyond a doubt that its [30] cultivation dates back to time immemorial. Beans, millet, and barley are indeed named once, together with silkworms, in the account of the "Divine Age." 32 But the passage has every aspect of an interpolation in the legend, perhaps not dating back long before the time of the eighth century compiler. A few unimportant vegetables and fruits, of most of which there is but a single mention, will be found in the list of plants given below. The intoxicating liquor called sake was known in Japan during the mythical period 33 and so were chopsticks for eating the food with. Cooking-pots and cups and dishes—the latter both of earthenware and of leaves of trees,—are also mentioned; but of the use of fire for warming purposes we hear nothing. Tables are named several times, but never in connection with food. They would seem to have been exclusively used for the purpose of presenting offerings on, and were probably quite small and low,—in fact rather trays than tables according to European ideas.

In the use of clothing and the specialization of garments

p. xli

the Early Japanese had reached a high level. We read in the most ancient legends of upper garments, skirts, trowsers, girdles, veils, and hats, while both sexes adorned themselves with necklaces, bracelets, and head-ornaments of stones considered precious,—in this respect offering a striking contrast to their descendants in modern times, of whose attire jewelry forms no part. The material of their clothes was hempen cloth and paper-mulberry bark, coloured by being rubbed with madder, and probably with woad and other tinctorial plants. All the garments, so far as we may judge, were woven, sewing being nowhere mentioned, and it being expressly stated by the Chinese commentator on the "Shan Hai Ching," 34 who wrote early in the fourth century, that the Japanese had no needles. 35 From the great place which the chase occupied in daily life we are led to suppose that skins also were used to make garments of. There is in the "Records" at least one passage which favours this supposition, 36 and the "Chronicles" in one place mention the straw rain-coat and broad-brimmed hat, which still form the [31] Japanese peasant's effectual protection against the inclemencies of the weather. The tendrils of creeping plants served the purposes of string, and bound the warrior's sword round his waist. Combs are mentioned, and it is evident that much attention was devoted to the dressing of the hair. The men seem to have bound up their hair in two bunches, one on each side of the head, whilst the young boys tied theirs into a topknot, the unmarried girls let their locks hang down over their necks, and the

p. xlii

married women dressed theirs after a fashion which apparently combined the two last-named methods. There is no mention in any of the old books of cutting the hair or beard except in token of disgrace; neither do we gather that the sexes, but for this matter of the headdress, were distinguished by a diversity of apparel and ornamentation.

With regard to the precious stones mentioned above as having been used as ornaments for the head, neck, and arms, the texts themselves give us little or no information as to the identity of the stones meant to be referred to. Indeed it is plain (and the native commentators admit the fact) that a variety of Chinese characters properly denoting different sort of jewels were used indiscriminately by the early Japanese writers to represent the single native word tama which is the only one the language contains to denote any hard substance on which a special value is set, and which often refers chiefly to the rounded shape, so that it might in fact be translated by the word "bead" as fittingly as by the word "jewel." We know, however, from the specimens which have rewarded the labours of archaeological research in Japan that agate, crystal, glass, jade, serpentine, and steatite are the most usual materials, and carved and pierced cylindrical shapes (maga-tama and kuda-tama), the commonest forms. 37

The horse (which was ridden, but not driven), the

p. xliii

barn-door fowl, and the cormorant used for fishing) are the only domesticated creatures mentioned in the earlier traditions, with the doubtful exception of the silkworm, to [32] which reference has already been made. 38 In the later portions of the "Records" and "Chronicles," dogs and cattle are alluded to; but sheep, swine, and even cats were apparently not yet introduced. Indeed sheep were scarcely to be seen in Japan until a few years ago, goats are still almost unknown, and swine and all poultry excepting the barn-door fowl extremely uncommon.

The following enumeration of the animals and plants mentioned in the earlier portion 39 of the "Records" may be of interest. The Japanese equivalents, some few of which are obsolete, are put in parenthesis, together with the Chinese characters used to write them:


Bear, (kuma ).
Boar, (wi ).
Deer, (shika ).
Hare, (usagi ).
Horse, (uma and koma ).
Mouse or Rat (nedzumi ).
"Sea-ass" [Seal or Sea-lion?] (michi ).
Whale, (kujira ).

p. xliv


Barndoor-fowl, (kake ).
Cormorant, (u ).
Crow or Raven, (karasu ).
Dotterel or Plover or Sand-piper, (chidori ).
Heron or Egret (sagi ).
Kingfisher (soni-dori ), Nuye ( ). 40
Pheasant (kigishi ).
Snipe, (shigi ).
Swan, (shiro-tori )
Wild-duck, (kamo ).
Wild-goose, (kari ).


Crocodile, (wani ). 41
Tortoise (kame ).
Toad or Frog, (taniguku, written phonetically).
Serpent, (worochi ).
Snake [smaller than the preceding], (hemi ).


Centipede, (mukade ).
Dragon-fly, (akidzu ).
Fly, (hahi ).
Louse, (shirami ).
Silk-worm, (kahiko ).
Wasp or Bee, (hachi ).


Pagrus cardinalis [probably], (aka-dahi ) [or perhaps the Pagrus cardinalis (tai ) is intended.]
Perch [Percalabrax japonicus] (su-dzuki ).
Beche-de-mer [genus Pentacta] (ko ).
Medusa, (kurage, written phonetically).

p. xlv


Arca Subcrenata [?] (hirabu-kahi, written phonetically).
Cockle [Arca Inflata] (kisa-gahi ).
Turbinidæ [a shell of the family] (shitadami ).


Ampelopsis serianæfolia [?] (kaga-mi ).
Aphanarthe aspera, (muku, written phonetically).
Aucuba japonica [probably] aha-gi, written phonetically).
Bamboo, (take ).
Bamboo-grass [Bambusa chino], (sasa ).
Barley [or wheat?], (mugi ).
Beans [two kinds, viz., Soja glycine and Phaselus radiatus (the general name is mame , that of the latter species in particular adzuki ).
Bulrush [Typha japonica] (kama ).
Bush-clover [Lespedeza of various species], (hagi ).
Camellia japonica (tsuba-ki ).
Cassia [Chinese mythical; or perhaps the native Cercidiphyllum japonica], (katsura, variously written).
Chamæcyparis obtusa, (hi-no-ki )
Cleyera japonica [and another allied but undetermined species) (saka-ki ).
Clubmoss, (hi-koge ).
Cocculus thunbergi [probably] (tsu-dzura ).
Cryptomeria japonica, (sugi ).
Eulalia japonica (kaya ).
Evonymus japonica, (masa-ki ).
Ginger, [or perhaps the Xanthoxylon is intended] (hazhikami ).
Halocholoa macrantha [but it is not certain that this is the sea-weed intended] (komo ).
Holly [or rather the Olea aquifolium, which closely resembles holly], (hihira-gi ).
Knot-grass [Polygonum tinctorium] (awi ).
Lily, (sawi written phonetically, yamayuri-gusa , and saki-kusa ).
Madder. (akane ).
Millet [Panicum italicum], (aha ).
Moss, (koke ).
Oak (two species, one evergreen and one deciduous,—Quercus myrsinæfolia, Q. dentata (kashi , kashiwa ).
Peach, (momo ).
Photinia glabra [?], (soba, written phonetically).
Pine-tree, (matsu ).
Pueraria thumbergiana, (kudzu ).
Reed, (ashi ).
Rice, (ine ).
Sea-weed [or the original term may designate a particular species], (me ).
Sedge [Scriptus marintimus], (suge ).
Spindle-tree [Evonymus radians], (masaki no kadzura ).
Vegetable Wax-tree [Rhus succedanea], (hazhi ).
Vine, (yebi-kadsura ). p. xliv
Wild cherry [or birch?], (hahaka ).
Wild chive [or rather the allium and odorum, which closely resembles it], (ka-mira ).
Winter-cherry [Physalis alkekengi] aka-kagachi written phonetically, also (hohodzuki ).

The later portions of the work furnish in addition the following:—


Cow (ushi ).
Dog, (inu ).
Crane, [genus Grus] (tadzu ).
[35] Dove or Pigeon, (hato ).
Grebe, (niho-dori ).
Lark, (hibari ).
Peregrine falcon, (hayabusa ).
Red-throated quail (udzura )
Tree-sparrow (susume ).
Wag-tail, [probably] (mana-bashira), written phonetically.
Wren, (sazaki ).
Dolphin, (iruka ).
Trout, [Plecoglossus altivelis] (ayu ).
Tunny, [a kind of, viz. Thynnus sibi] (shibi ).
Crab, (kani ).
Horse-fly (amu ).
Oyster (kaki) .


Alder [Alnus maritima] (hari-no-ki ).
Aralia (mi-tsuna gashiwa ).
Brasenia peltata (nunaha) .
Cabbage [brassica] (aona ).
Catalpa Kaempfri [but some say the cherry is meant (adzusa ).
Chestnut (kuri ).
Dioscorea quinqueloba (tokoro-dzura ).
Evonymus sieboldianus mayumi )
Gourd (hisago ).
Hedysarum esculentum (wogi ).
Hydropyrum latifolium (komo ).
Kadzura japonica (sen kadzura ).
Livistona sinensis (aji-masa ).
Lotus [nelumbium] (hachisu ).
Musk-melon (hozoahi ).
Oak, [three species, Quercus serrata (kunugi ) and Q. glandifera (nara ), both deciduous; Q. gilva (ichihi ) [evergreen].
Orange (tachibana ).
Podocarpus macrophylla (maki ).
Radish, [Raphanus sativus] (oho-ne ).
Sashibu (written phonetically) [not identified].
Water caltrop, [Trapa bispinosa] (hishi ).
Wild garlic [Allium nipponicum] (nubiru ).
Zelkowa keaki [probably] (tsuki ).

p. xlvii

A few more are probably preserved in the names of places. Thus in Shinano, the name of a province, we seem to have the shina (Tilia cordata), and in Tadetsu the tade (Polygonum japonicum). But the identification in these cases is mostly uncertain. It must also be remembered that, as in the case of all non-scientific nomenclatures, several species, and occasionally even more than one genus, are included in a single Japanese term. The chi-dori (here always rendered "dotterel") is the name of any kind of sand-piper, plover or dotterel. Kari is a general name applied to geese, but not to all the species, and also to the great bustard. Again it should not be forgotten that there may have been, and probably were, in the application of some of these terms, differences of usage between the present day and eleven or twelve centuries ago. Absolute precision is therefore not attainable. 42

Noticeable in the above lists is the abundant mention of plant-names in a work which is in no ways occupied with botany. Equally noticeable is the absence of some of those which are most common at the present day, such as the tea-plant and the plum-tree, while of the orange we are specially informed that it was introduced from abroad. 43 The difference between the various stones and metals seems, on the other hand, to have attracted [36] very little attention from the Early Japanese. In late

p. xlviii

times the chief metals were named mostly according to their colour, as follows:

Yellow metal


White metal


Red metal


Black metal


Chinese (or Korean) metal


But in the "Records" the only metal of which it is implied that it was in use from time immemorial is iron, while "various treasures dazzling to the eye, from gold and silver downwards," are only referred to once as existing in the far-western land of Korea. Red clay is the sole kind of earth specially named.

Blue (including Green).
Piebald (of horses).

Yellow is not mentioned (except in the foreign Chinese phrase "the Yellow Stream," signifying Hades, and not to be counted in this context), neither are any of the numerous terms which in Modem Japanese serve to distinguish delicate shades of colour. We hear of the "blue (or green) (i.e. black 44) clouds" and also of the "blue (or green), sea"; but the "blue sky" is conspicuous by its absence here as in so many other early literatures, though strangely enough it does occur in the oldest written monuments of the Chinese.

With regard to the subject of names for the different degrees of relationship,—a subject of sufficient interest

p. xlix

to the student of sociology to warrant its being discussed at some length,—it may be stated that in modern Japanese parlance the categories according to which relationship is conceived of do not materially differ from those that are current in Europe. Thus we find father, grandfather, great-grandfather, uncle, nephew. stepfather, stepson, father-in-law, and the corresponding terms for females,—mother, grandmother etc.,—as well as such vaguer designations as parents, ancestors, cousins, and kinsmen, The only striking difference is that brothers and sisters, instead of being considered as all mutually related in the same manner, are divided into two categories, viz:


elder brother(s),


younger brother(s),


elder sister(s),


younger sister(s).

in exact accordance with Chinese usage.

Now in Archaic times there seems to have been a different and more complicated system, somewhat resembling that which still obtains among the natives of Korea, and which the introduction of Chinese ideas and especially the use of the Chinese written characters must have caused to be afterwards abandoned. There are indications of it in some of the phonetically written fragments of the "Records." But they are not of themselves sufficient to furnish a satisfactory explanation, and the subject has puzzled the native literati themselves. Moreover the English language fails us at this point, and elder and younger brother, elder and younger sister are the only terms at the translator's command. It may therefore be as well to quote in extenso Motowori's elucidation of the Archaic usage to be found

p. l

in vol. XIII, p. 63-4 of his "Exposition of the Records of Ancient Matters." 45 He says: "Anciently, when brothers and sisters were spoken of, the elder brother was called se or ani in contradistinction to the younger brothers and younger sisters, and the younger brother also was called se in contradistinction to the elder sister. The elder sister was called ane in contradistinction to the younger sister, and the younger brother also would use the word ane in speaking of his elder sister himself. The younger brother was called oto in contradistinction to the elder brother, and the younger sister also was called oto in contradistinction to the elder sister. The younger sister was called imo in contradistinction to the elder brother, and the elder sister also was called imo in [38] contradistinction to the younger brother. It was also the custom among brothers and sisters to use the words iro-se for se, iro-ne for ane, and iro-do for oto, and analogy forces us to conclude that iro-mo was used for imo." (Motowori elsewhere explains iro as a term of endearment identical with the word iro, "love"; but we may hesitate to accept this view.) It will be observed that the foundation of this system of nomenclature was a subordination of the younger to the elder-born modified by a subordination of the females to the males. In the East, especially in primitive times, it is not "Place aux dames," but "Place aux messieurs."

Another important point to notice is that, though in a few passages of the "Records" we find a distinction drawn between the chief and the secondary wives, perhaps nothing more than the favorite or better-born,

p. li

and the less well-born, are meant to be thus designated,—yet not only is this distinction not drawn throughout, but the wife is constantly spoken of as imo, i.e. "younger sister." It fact sister and wife were convertible terms and ideas; and what in a later stage of Japanese, as of Western, civilization is abhorred as incest was in Archaic Japanese times the common practice. We also hear of marriages with half-sisters, with stepmothers, and with aunts; and to wed two or three sisters at the same time was a recognized usage. Most such unions were naturally so contrary to Chinese ethical ideas, that one of the first traces of the influence of the latter in Japan was the stigmatizing of them as incest; and the conflict between the old native custom and the imported moral code is seen to have resulted in political troubles. 46 Marriage with sisters was naturally the first to disappear, and indeed it is only mentioned in the legends of the gods; but unions with half-sisters, aunts, etc., lasted on into the historic epoch. Of exogamy, such as obtains in China, there is no trace in any Japanese document, nor do any other artificial impediments seem to have stood in the way of the free choice of the Early Japanese man, who also (in some cases at least) received a dowry with his bride or brides.

*          *          *

If, taking as our guides the incidental notices which are scattered up and down the pages of the earlier [39] portion of the "Records" we endeavour to follow an Archaic Japanese through the chief events of his life

p. lii

from the cradle to the tomb, it will be necessary to begin by recalling what has already been alluded to as the 'parturition-house' built by the mother, and in which, as we are specially told that it was made windowless, it would perhaps be contradictory to say that the infant first saw the light. Soon after birth a name was given to it,—given to it by the mother,—such name generally containing some appropriate personal reference. In the most ancient times each person (so far as we can judge) bore but one name, or rather one string of words compounded together into a sort of personal designation. But already at the dawn of the historical epoch we are met by the mention of surnames and of what, in the absence of a more fitting word, the translator has ventured to call "gentile names," bestowed by the sovereign as a recompense for some noteworthy deed. 47

It may be gathered from our text that the idea of calling in the services of wet-nurses in exceptional case had already suggested itself to the minds of the ruling class, whose infants were likewise sometimes attended by special bathing-women. To what we should call education, whether mental or physical, there is absolutely no reference made in the histories. All that can be inferred is that, when old enough to do so, the boys began to follow one of the callings of hunter or fisherman,

p. liii

while the girls staid at home weaving the garments of the family. There was also a great deal of fighting, generally of a treacherous kind, in the intervals of which the warriors occupied themselves in cultivating patches of ground. The very little which is to be gathered concerning the treatment of old people would seem to indicate that they were well cared for.

We are nowhere told of any wedding ceremonies except the giving of presents by the bride or her father, the probable reason being that no such ceremonies [40] existed. Indeed late on into the Middle Age cohabitation alone constituted matrimony,—cohabitation often secret at first, but afterwards acknowledged, when, instead of going round under cover of night to visit his mistress, the young man brought her back publicly to his parents' house. Mistress, wife, and concubine were thus terms which were not distinguished, and the woman could naturally be discarded at any moment. She indeed was expected to remain faithful to the man with whom she had had more than a passing intimacy, but no reciprocal obligation bound him to her. Thus the wife of one of the gods is made to address her husband in a poem which says:

"Thou . . . indeed, being a man, probably hast on the various island-headlands that thou seest, and on every beach-headland that thou lookest on, a wife like the young herbs. But I, alas! being a woman, have no spouse except thee," etc., etc. 48

In this sombre picture the only graceful touch is the custom which lovers or spouses had of tieing each other's girdles when about to part for a time,—a ceremony by

p. liv

which they implied that they would be constant to each other during the period of absence. 49 What became of the children in cases of conjugal separation does not clearly appear. In the only instance which is related at length, we find the child left with the father; but this instance is not a normal one. 50 Adoption is not mentioned in the earliest traditions; so that when we meet with it later on we shall probably be justified in tracing its introduction to Chinese sources.

Of death-bed scenes and dying speeches we hear but little, and that little need not detain us. The burial rites are more important. The various ceremonies observed on such an occasion are indeed not explicitly detailed. But we gather thus much: that the hut tenanted by the deceased was abandoned,—an ancient custom to whose former existence the removal of the capital at the commencement of each reign long [41] continued to bear witness,—and that the body was first deposited for some days in a "mourning-house," during which interval the survivors (though their tears and lamentations are also mentioned) held a carousal, feasting perhaps on the food which was specially prepared as an offering to the dead person. Afterwards, the corpse was interred, presumably in a wooden bier, as the introduction of stone tombs is specially noted by the historian as having taken place at the end of the reign of the Emperor Sui-nin, and was therefore believed by those who handed down the legendary history to have been a comparatively recent innovation, the date assigned to this monarch by the author of the "Chronicles" coinciding

p. lv

with the latter part of our first, and the first half of our second centuries. To a time not long anterior is attributed the abolition of a custom previously observed at the interments of royal personages. This custom was the burying alive of some of their retainers in the neighbourhood of the tomb. We know also, both from other early literary sources and from the finds which have recently rewarded the labours of archaeologists, that articles of Clothing, ornaments, etc., were buried with the corpse. it is all the more curious that the "Records" should nowhere make any reference to such a custom, and is a proof (if any be needed) of the necessity of not relying exclusively on any single authority, however respectable, if the full and true picture of Japanese antiquity is to be restored. A few details as to the abolition of the custom of burying retainers alive round their master's tomb, and of the substitution for this cruel holocaust of images in clay will be found in Sect. LXIII, Note 23, and in Sect. LXXV, Note 4, of the following translation. 51 If the custom be one which is properly included under the heading of human sacrifices, it is the only form of such sacrifices of which the earliest recorded Japanese social state retained any trace. The absence of slavery is another honourable feature. On the other hand, the most cruel punishments were dealt out to enemies and wrongdoers. Their nails were extracted, the sinews of their knees were cut, they were buried up to the neck so that their eyes burst, etc. Death, too, was inflicted for the [42]

p. lvi

most trivial offences. Of branding, or rather tattooing, the face as a punishment there are one or two incidental mentions. But as no tattooing or other marking or painting of the body for any other purpose is ever alluded to, with the solitary exception in one passage of the painting of her eyebrows by a woman, it is possible that the penal use of tattooing may have been borrowed from the Chinese, to whom it was not unknown.

The shocking obscenity of word and act to which the "Records" bear witness is another ugly feature which must not quite be passed over in silence. It is true that decency, as we understand it, is a very modern product, and is not to be looked for in any society in the barbarous stage. At the same time, the whole range of literature might perhaps be ransacked in vain for a parallel to the naïve filthiness of the passage forming Sect. IV of the following translation, or to the extraordinary topic which the hero Yamato-Take and his mistress Miyazu are made to select as the theme of poetical repartee. 52 One passage likewise would lead us to suppose that the most beastly crimes were commonly committed. 53

To conclude this portion of the subject, it may be useful for the sake of comparison to call attention to a few arts and products with which the early Japanese were not acquainted, Thus they had no tea, no fans, no porcelain, no lacquer,—none of the things, in fact, by which in later times they have been chiefly known. They did not yet use vehicles of any kind. They had no accurate method of computing time, no money, scarcely any

p. lvii

knowledge of medicine. Neither, though they possessed some sort of music, and poems a few of which at least are not without merit, 54 do we hear anything of the art of drawing. But the most important art of which they were ignorant is that of writing. As some misapprehension has existed on this head, and scholars in [43] Europe have been misled by the inventions of zealous champions of the Shintō religion into a belief in the so-called "Divine Characters," by them alleged to have been invented by the Japanese gods and to have been used by the Japanese people prior to the introduction of the Chinese ideographic writing, it must be stated precisely that all the traditions of the "Divine Age," and of the reigns of the earlier Emperors down to the third century of our era according to the received chronology, maintain a complete silence on the subject of writing materials, and records of every kind. Books are nowhere mentioned till a period confessedly posterior to the opening up of intercourse with the Asiatic continent, and the first books whose names occur are the "Lun Yü" and the "Ch'ien Tzŭ Wên," 55 which are said to have been brought over to Japan during the reign of the Emperor Ō-jin,—according to the same chronology in the year 284 after Christ. That even

p. lviii

this statement is antedated, is shown by the fact that the "Ch'ien Tzŭ Wên," was not written till more than two centuries later,—a fact which is worthy the attention of those who have been disposed simply to take on trust the assertions of the Japanese historians. It should likewise be mentioned that, as has already been pointed out by Mr. Aston, the Japanese terms fumi "written document," and fude "pen," are probably corruptions of foreign words. 56 The present, indeed, is not the [44] place to discuss the whole question of the so-called "Divine Characters" which Motowori, the most patriotic as well as the most learned of the Japanese literati, dismisses in a note to the Prolegomena of his "Exposition of the Records of Ancient Matters" with the remark that they "are a late forgery over which no words need be wasted." But as this mare's nest has been imported into the discussion of the Early Japanese social state, and as the point is one on which the absolute silence of the early traditions bears such clear testimony, it was impossible to pass it by without some brief allusion.


xxxiv:23 A curious scrap of the history of Japanese civilization is preserved in the word kaji, whose exclusive acceptation in the modern tongue is "rudder." In archaic Japanese it meant "oar," a signification which is now expressed by the term ro, which has been borrowed from the Chinese. It is a matter of debate whether the ancient Japanese boats possessed such an appliance as a rudder, and the word tagishi or iaishi has been credited with that meaning. The more likely opinion seems to be that both the thing and the word were specialized in later times, the early Japanese boatmen having made any oar do duty for a rudder when circumstances necessitated the use of one.

xxxv:24 See the end of Sect. XXXII.

xxxv:25 See Vol. IX, Pt. II, pp. 191-192, of these "Transactions."

xxxvii:26 Yamato Monogatari.

xxxvii:27 For a translation of this story see the present writer's "Classical Poetry of the Japanese," pp. 42, 44.

xxxvii:28 See Sect. XLIV, Note 12 and Sect. LXXII, Note 29.

xxxviii:29 Mr. Ernest Satow, who in 1898 visited the island of Hachijo, gives the following detail concerning the observance down to modern times in that remote corner of the Japanese Empire of the custom mentioned in the text: "in Hachijo women, when about to become mothers, were formerly driven out to the huts on the mountain-side, and according to the accounts of native writers, left to shift for themselves, the result not unfrequently being the death of the newborn infant, or if it survived the rude circumstances under which it first saw the light, the seeds of disease were sown which clung to it throughout its after life. The rule of non-intercourse was so strictly enforced, that the woman was not allowed to leave the hut even to visit her own parents at the point of death, and besides the injurious effects that this solitary confinement must have had on the wives themselves, their prolonged absence was a serious loss to households, where there were elder children and large establishments to be superintended. The rigour of the custom was so far relaxed in modern times, that the huts were no longer built on the hills, but were constructed inside the homestead. It was a subject of wonder to people from other parts of Japan that the senseless practice should still be kept up, and its abolition was often recommended, but the administration of the Shoguns was not animated by a reforming spirit, and it remained for the Government of the Mikado to exhort the islanders to abandon this and the previously mentioned custom. They are therefore no longer sanctioned by official authority and the force of social opinion against them is increasing, so that before long these relics of ancient ceremonial religion will in all probability have disappeared from the group of islands." (Trans. of the Asiat. Soc. of Japan, Vol. VI, Part III,, pp. 455-6.)

xxxix:30 See Sect. LXX, Note 6. The Japanese term is ina-ki, ki being an Archaic term for "castle."

xxxix:31 See Sect. XVI. Mention of cave-dwellers will also be found in Sects. XLVIII, and LXXX.

xl:32 See the latter part of Sect. XVII.

xl:33 See Sect. XVIII, Note 16.

xli:34 .

xli:35 See, however, the legend in Sect. LXV.

xli:36 See beginning of Sect XXVII.

xlii:37 For details on this subject and illustrations, see Mr. Henry von Siebold's "Notes on Japanese Archaeology," p. 15 and Table XI, and a paper by Professor Milne on the "Stone Age in Japan," read before the Anthropological Society of Great Britain on the 25th May, 1880, pp. 10 and 11.

xliii:38 The tradition preserved in Sect. CXXIV, shows that in times almost, if not quite, historical (the 4th century of our era) the silkworm was a curious novelty, apparently imported from Korea. It is not only possible, but probable, that silken fabrics were occasionally imported into Japan from the mainland at an earlier period, which would account for the mention of "silk rugs" in Sects XL and LXXXIV.

xliii:39 The (necessarily somewhat arbitrary) line between earlier and later times has been drawn at the epoch of the traditional conquest of Korea by the Empress Jin-go at the commencement of the third century of our era, it being then, according to the received opinions, that the Japanese first came in contact with their continental neighbours, and began to borrow from them. (See however the concluding Section of this Introduction for a demonstration of the untrustworthiness of all the so-called history of Japan down to the commencement of the fifth century of the Christian era).

xliv:40 See Sect. XXIV, Note 4.

xliv:41 Mr. Satow, in his translation of a passage of the "Records of Ancient Matters" forming part of a note to his third paper on the "Rituals" in Vol. IX, Pt. II of these "Transactions," renders wani by "shark." There is perhaps some want of clearness in the old historical books in the details concerning the creature in question, and its fin is mentioned in the "Chronicles." But the accounts point rather to an amphibious creature, conceived of as being somewhat similar to the serpent, than to a fish, and the Chinese descriptions quoted by the Japanese commentators unmistakably refer to the crocodile. The translator therefore sees no sufficient reason for abandoning the usually accepted interpretation of wani ( ) as "crocodile." It should be noticed that the wani is never introduced into any but patently fabulous stories, and that the example of other nations, and indeed of Japan itself, shows that myth-makers have no objection to embellish their tales by the mention of wonders supposed to exist in foreign lands.

xlvii:42 Sect. CXXVIII preserves a very early ornithological observation in the shape of the Songs composed by the Emperor Nin-toku and his Minister Take-Uchi on the subject of a wild-goose laying eggs in Central Japan. These birds are not known to breed even so far South as the island of Yezo.

xlvii:43 See the legend in Sect. LXXIV.

xlviii:44 Mr. Satow suggests that awo ("blue" or "green") means property any colour derived from the awi plant (Polygonum tinctorium.)

l:45 Only the foot-notes of the original are omitted, as not being essential.

li:46 See the story of Prince Karu, which is probably historical, in Sects. CXLI et seq.

lii:47 The custom of using surnames was certainly borrowed from China, although the Japanese have not, like the Koreans, gone so far as to adopt the actual surnames in use in that country. The "gentile names" may have sprung up more naturally, though they too show traces of Chinese influence. Those most frequently met with are Agata-nushi, Ason, Atahe, Kimi, Miyatsuko, Murazhi, Omi, Sukune, and Wake. See above, pp. xv-xvi.

liii:48 See Sect. XXV (the second Song in that Section).

liv:49 See Sect LXXI, Note 12.

liv:50 See Sect. XLII.

lv:51 Representations of these clay images (Tsuchi-nin-giyō) will be found in Table XII of Mr. Henry von Siebold's "Notes on Japanese Archaeology," and in Mr. Satow's paper on "Ancient Sepulchral Mounds in Kaudzuke" published in Vol. VII, Pt. III, pp. 313 et seq. of these "Transactions."

lvi:52 See Sect. LXXXVII.

lvi:53 See Sect. XCVII.

lvii:54 A translation,—especially a literal prose translation,—is not calculated to show off to best advantage the poetry of an alien race. But even subject to this drawback, the present writer would be surprised if it were not granted that poetic fire and grace are displayed in some of the Love-Songs (for instance the third Song in Sect. XXIV and both Songs Yamato-Take's address to his "elder brother the pine-tree," and in his Death-Songs contained in Sect. LXXXIX).

lvii:55 and .

lviii:56 Viz. of the Chinese and (in the modern Mandarin pronunciation wên and pi). Mr. Aston would seem to derive both the Japanese term fude and the Korean put independently from the Chinese . The present writer thinks it more likely that the Japanese fude was borrowed mediately through the Korean put. In any case, as it regularly corresponds with the latter according to the laws of letter-change subsisting between the two languages, it will be observed that the Japanese term would still have to be considered borrowed, even if the derivation of put from had to be abandoned; for we can hardly suppose Korean and Japanese to have independently selected the same root to denote such a thing as a "pen." As to the correctness of the derivation of fumi from , there can be little doubt, and it had long ago struck even the Japanese themselves, who are not prompt to acknowledge such loans. They usually derive fude from fumi-te, "document hand," and thus again we are brought bark to the Chinese as the origin of the Japanese word for "pen."

Next: V. Religious And Political Ideas of the Early Japanese, Beginnings of the Japanese Nation, and Credibility of the National Records