The Kojiki, translated by Basil Hall Chamberlain, [1919], at sacred-texts.com

II.

METHODS OF TRANSLATION.

To the translator the question above mooted is not one of great importance. The text itself must form the basis of his version, and not any one's,—not even Motowori's—private and particular reading of it. For this reason none of the Honorifics, which Motowori inserts as prefixes to nouns and terminations to verbs have been taken any notice of, but the original has been followed, character by character, with as great fidelity as was attainable. The author too has his Honorifics; but he does not use them so plentifully or so regularly as it pleases Motowori to represent him as having intended to do. On the other hand, Motowori's occasional emendations of the text may generally be accepted. They rarely extend to more than single words; and the errors in the earlier editions may frequently be shown to have arisen from careless copying of characters originally written, not in the square, but in the cursive form. The translator has separately considered each case where various readings

p. xvii

occur, and has mentioned them in the Notes when they seemed of sufficient importance. In some few cases he has preferred a reading not approved by Motowori, but he always mentions Motowori's reading in a Foot-note.

The main body of the text contains but little to perplex any one who has made a special study of the early Japanese writings, and it has already been noticed that there is an admirable exegetical literature at the student's command. With the Songs embedded in the prose text the case is different, as some of them are among the most difficult things in the language, and the commentators frequently arrive at most discordant interpretations of the obscurer passages. In the present version particulars concerning each Song have, except in a very few cases where comment appeared superfluous, been given [13] in a Foot-note, the general sense being usually first indicated, the meaning of particular expressions then explained, and various opinions mentioned when they seemed worthy of notice. Besides one or two terms of Japanese grammar, the only technical knowledge with which the readers of the Notes are necessarily credited is that of the use by the Japanese poets of what have been styled Pillow-Words, Pivots, and Prefaces; and those Pillow-Words which are founded on a jeu-de-mots or are of doubtful signification form, with the one exception mentioned below, the only case where anything contained in the original is omitted from the English version. 8 After some consideration, it has been deemed

p. xviii

advisable to print in an Appendix the Japanese text of all the Songs, transliterated into Roman. Students will thus find it easier to form their own opinion on the interpretation of doubtful passages. The importance likewise of these Songs, as the most ancient specimens of Altaic speech, makes it right to give them as much publicity as possible.

The text of the "Records" is, like many other Japanese texts, completely devoid of breaks corresponding to the chapters and paragraphs into which European works are divided. With the occasional exception of a pause after a catalogue of gods or princes, and of notes inserted in smaller type and generally containing genealogies or indicating the pronunciation of certain words, the whole story, prose and verse, runs on from beginning to end with no interruptions other than those marked by the conclusion of Vol. I and by the death of each emperor in Vols. II and III. Faithfulness however scarcely seems to demand more than this statement; for a similarly continuous printing of the English version would attain no end but that of making a very dry piece of reading more arduous still. Moreover there are certain traditional names by which the various episodes of the so-called "Divine Age" are known to the native scholars, and according to which the text of Vol. I may naturally be divided. The reigns of the emperors form a similar foundation for the analysis of Vols. II and III, which contain the account of the "Human Age." It has been thought that it would be well to mark such natural [14] divisions by the use of numbered Sections with marginal headings. The titles proposed by Motowori in the Prolegomena to his Commentary have been adopted with

p. xix

scarcely any alteration in the case of Vol. I. In Vols. II and III, where his sections mostly embrace the whole reign of an emperor, and the title given by him to each Section consists only of the name of the palace where each emperor is said to have resided, there is less advantage in following him; for those Sections are often inordinately long, and their titles occasionally misleading and always inconvenient for purposes of reference, as the Japanese emperors are commonly known, not by the names of their places of residence, but by their "canonical names." Motowori, as an ardent nationalist, of course rejected these "canonical names," because they were first applied to the Japanese emperors at a comparatively late date in imitation of Chinese usage. But to a foreigner this need be no sufficient reason for discarding them. The Sections in the translation of Vols. II and III have therefore been obtained by breaking up the longer reigns into appropriate portions; and in such Sections, as also in the Foot-notes, the emperors are always mentioned by their "canonical names." 9 The Vol. mentioned in brackets on every right-hand page is that of Motowori's Commentary which treats of the Section contained in that page.

The Notes translated from the original are indented, and are printed small when they are in small type in

p. xx

the Japanese text. Those only which give directions for pronouncing certain characters phonetically have been omitted, as they have no significance when the original tongue and method of writing are exchanged for foreign vehicles of thought and expression. The songs have likewise been indented for the sake of clearness, and each one printed as a separate paragraph. The occasionally unavoidable insertion in the translation of important words not occurring in the Japanese text has been indicated by [15] printing such words within square brackets. The translator's Notes, which figure at the bottom of each page, do not aim at anything more than the exegesis of the actual text. To illustrate its subject-matter from other sources, as Motowori does, and to enlarge on all the subjects connected with Japanese antiquity which are sometimes merely alluded to in a single phrase, would require several more volumes the size of this one, many years of labour on the part of the investigator, and an unusually large stock of patience on the reader's part. The Notes terminate with the death of the Emperor Kenzō, after which the text ceases to offer any interest, except as a comment on the genealogies given in the "Chronicles of Japan.'

Without forgetting the fact that so-called equivalent terms in two languages rarely quite cover each other, and that it may therefore be necessary in some cases to render one Japanese word by two or three different English words according to the context, the translator has striven to keep such diversity within the narrowest limits, as it tends to give a false impression of the original, implying that it possesses a versatility of thought which is indeed characteristic of Modern Europe, but not

p. xxi

at all of Early Japan. With reference to this point a certain class of words must be mentioned, as the English translation is unavoidably defective in their case, owing to the fact of our language not possessing sufficiently close synonyms for them. They are chiefly the names of titles, and are the following:—

It must be understood that no special significance is to be attached to the use of such words as "Duke," "Suzerain" etc. They are merely, so to speak, labels by which titles that are distinct in the original are sought to be kept distinct in the translation. Many of them also are used as that species of hereditary titular designation which the translator has ventured to call the "gentile name." 10 Where possible, indeed, the etymological

p. xxii

meaning of the Japanese word has been preserved. Thus omi seems to be rightly derived by Motowori from ohomi, "great body"; and "grandee" is therefore the nearest English equivalent. Similarly murazhi, "chief," is a corruption of two words signifying "master of a tribe." On the other hand, both the etymology and the precise import of the title of wake are extremely doubtful. Hiko and hime again, if they really come from hi ko, "sun-child" and hi me, "sun-female" (or "fire child" and "fire female"), have wandered so far from their origin as, even in Archaic times, to have been nothing more than Honorific appellations, corresponding in a loose fashion to the English words "prince and princess," or "lord and lady,"—in some cases perhaps meaning scarcely more than "youth and maiden."

The four words kami, ma, miko and mikoto alone call for special notice; and ma may be disposed of first. It is of uncertain origin, but identified by the native philologists with the perpetually recurring honorific mi, rendered "august." As, when written ideographically, it is always represented by the Chinese character , the translator renders it in English by "true"; but it must be understood that this word has no force beyond that of an Honorific.

Mikoto, rendered "Augustness," is properly a [17] compound, mi koto, "august thing." It is used as a title, somewhat after the fashion of our words "Majesty" and "Highness," being suffixed to the names of exalted human personages, and also of gods and goddesses. For the sake of clearness in the English translation this title is prefixed and used with the possessive pronoun, thus: Yamato-Take-no-Mikoto, His Augustness Yamato-Take.

p. xxiii

With regard to the title read miko by the native commentators, it is represented in two ways in the Chinese text. When a young prince is denoted by it, we find the characters , "august child," reminding us of the Spanish title of infante. But in other cases it is written with the single character "King," and it may be questioned whether the reading of it as miko is not arbitrary. Many indications lead us to suppose that in Early Japan something similar to the feudal system, which again obtained during the Middle Ages, was in force; and if so, then some of these "kings" may have been kings indeed after a fashion; and to degrade their title, as do the modern commentators, to that of "prince" is an anachronism. In any case the safest plan, if we would not help to obscure this interesting political question, is to adhere to the proper signification of the character in the text, and that character is , "King." 11

Of all the words for which it is hard to find a suitable English equivalent, Kami is the hardest. Indeed there is no English word which renders it with any near approach to exactness. If therefore it is here rendered by the word "deity" ("deity" being preferred to "god" because it includes superior beings of both sexes), it must be clearly understood that the word "deity" is taken in a sense not sanctioned by any English dictionary; for kami, and "deity" or "god," only correspond to each other in a very rough manner. The proper meaning of the word "kami" is "top," or "above"; and it is still constantly so used. For this reason it has the secondary sense of "hair of the head;" and only the hair on the top 

p. xxiv

of the head,—not the hair on the face,—is so designated. Similarly the Government, in popular phraseology, is O Kami, literally "the honorably above"; and down to a few years ago Kami was the name of a certain titular [11] provincial rank. Thus it may be understood how the word was naturally applied to superiors in general, and especially to those more than human superiors whom we call "gods." A Japanese, to whom the origin of the word is patent, and who uses it every day in contexts by no means divine, does not receive from the word Kami the same impression of awe which is produced on the more earnest European mind by the words "deity" and "god," with their very different associations. In using the word "deity," therefore; to translate the Japanese term Kami we must, so to speak, bring it down from the heights to which Western thought has raised it. In fact Kami does not mean much more than "superior." This subject will be noticed again in Section V of the present Introduction; but so far as the word Kami itself is concerned, these remarks may suffice.

To conclude this Section, the translator must advert to his treatment of Proper Names, and he feels that he must plead guilty to a certain amount of inconsistency on this head. Indeed the treatment of Proper Names is always an embarrassment, partly because it is often difficult to determine what is a Proper Name, and partly because in translating a text into a foreign tongue Proper Names, whose meanings are evident in the original and perhaps have a bearing on the story, lose their significance; and the translator has therefore first of all to decide whether the name is really a Proper Name at all or simply a description of the personage or place,

p. xxv

and next whether he will sacrifice the meaning because the word is used as a name; or preserve the original name and thus fail to render the meaning,—a meaning which may be of importance as revealing the channels in which ancient thought flowed. For instance "Oho-kuni-nushi-no-kami," the "Deity Master of the Great Land," is clearly nothing more than a description of the god in question, who had several other names, and the reason of whose adoption of this special one was that the sovereignty of the "Great Land," i.e. of Japan (or rather of Idzumo and the neighbouring provinces in north-western Japan), was ceded to him by another god, whom he deceived and whose daughter he ran away with. 12 Again Toyo-ashi-hara-no-chi-aki-no-naga-i-ho-aki-no-midzu-ho-no-kuni, which signifies "the Luxuriant Reed-Moor, the Land of Fresh Rice-ears,—of a Thousand [19] Autumns,—of Long Five Hundred Autumns" cannot possibly be regarded as more than an honorific description of Japan. Such a catalogue of words could never have been used as a name. On the other hand it is plain that Tema was simply the proper name of a certain mountain, because there is no known word in Archaic Japanese to which it can with certainty be traced. The difficulty is with the intermediate cases,—the cases of those names which are but partly comprehensible or partly applicable to their bearers; and the difficulty is one of which there would seem to be no satisfactory solution possible. The translator may therefore merely state that in Vol. I of these "Records," where an unusual number of the Proper Names have a bearing on the legends related in the text, he has, wherever feasible, translated all those which

p. xxvi

are borne by persons, whether human or divine. In the succeeding Volumes he has not done so, nor has he, except in a very small number of instances, translated the Proper Names of places in any of the three volumes. In order, however, to convey all the needful information both as to sound and as to sense, the Japanese original is always indicated in a Foot-note when the translation has the name in English, and vice versâ, while all doubtful etymologies are discussed.

Footnotes

xvii:8 For a special account of the Pillow-Words, etc., see a paper by the present writer in Vol. V, Pt. I, pp. 79 et seq. of these "Transactions," and for a briefer notice, his "Classical Poetry of the Japanese," pp. 5 and 6.

xix:9 The practice of bestowing a canonical name (okurina ) on an emperor after his decease dates from the latter part of the eighth century of our era when, at the command of the emperor Kuwam-mu, a scholar named Mifune-no-Mahito selected suitable "canonical names" for all the previous sovereigns, from Jim-mu down to Kuwan-mu's immediate predecessor. From that time forward every emperor has received his "canonical name" soon after death, and it is generally by it alone that he is known to history.

xxi:10 See Sect IV. of this Introduction and Sect. XIV, Note 5 of the Translation.

xxiii:11 Conf. Section LVI, Note 7.

xxv:12 See the legend in Sect. XXIII.