1. My five periods

     My first attempt at translating Shakespeare was in 1882, when I was still a student and translated Julius Caesar. Since that time, my approach to Shakespeare translation has passed through at least five stages. The Caesar translation was produced in classical joruri format under the title Jiyu no tachi nagori no kireaji (A Tale of Caesar: The Lingering Sharpness of the Sword of Freedom), the style being liberally mixed with the syllabic rhythms of the classical theatre. It was slovenly in the extreme, although this initial period in my development was not untypical of translators of the day. Next, I sought a platform for addressing young people through my lectures on literature, published in the journal Waseda Bungaku in 1895 and 1896. I tried to make my translations as literal as possible, adding scholarly notes and writing in prose, but during this period of revival in Japanese literature I became fastidious in my use of vocabulary and Japanese grammar, and without hardly realising it I tended towards an ornate style. My vocabulary was cramped, losing the warmth and style of the original poetry, and even as plain writing tended to sound rather peculiar. This was the second period in my development as a translator of Shakespeare.

     The next stage was marked by experiment in live production of Shakespeare, when in 1908 and 1909 the Bungei Kyokai staged my translation of the first two acts of Hamlet. Since Shakespeare’s play was intended first and foremost for stage production, it seemed expedient to do likewise, and I learnt much from the experience. My goal in studying Shakespeare had been the reform of Japanese drama, but when I came to direct Hamlet a powerful awareness of what Shakespeare meant gradually dawned on me; it was with only a rather vague idea of what lay ahead that I had started to translate the play. It is different now. I could not have hoped for the rise of shinpa and the free and natural rhetoric of the shingeki, for the language of my Hamlet was inevitably touched by the kabuki and seven-five syllabic meter of traditional Japanese poetics.

     At around this time, I translated Macbeth in the excessively literary style of my second period. The late Lafcadio Hearn once told me that ‘Nowadays one would expect to translate Shakespeare in colloquial style’, hinting that I should abandon my literary style, but being preoccupied with staging Shakespeare, I did not take Hearn’s advice kindly. Even after I had started to translate the Complete Works, working on such plays as Hamlet, Romeo and Juliet and Othello, I could not abandon the literary approach. These translations represent the mixed style of my fourth period. This is to say that by following the original in its blend of formal poetry, poetic prose and colloquial language, I hoped to convey its warmth and rhythm. This method relied on guesswork, but could produce some interesting results. A smooth coordination of literary and colloquial language was necessary, and while I adopted a literary style that was comparatively modern, for the colloquial I used language that was as old as possible: in other words, language from kabuki and the Edo period, which was criticised – no doubt rightly – for smelling too much of kabuki, and I too was dissatisfied with it.

     When I translated King Lear, I toned down the literary element, bringing the style a notch closer to contemporary usage. Translating Julius Caesar I again tried to make it more colloquial. The secondary purpose of all these translations was performance by the Bungei Kyokai, and I tried to retain something of the special vocabulary of the original, the balance of high and low, the candour, the urgency of rhythm, all necessary for effective stage performance. When we staged Julius Caesar, I realized the truth of Hearn’s advice, and changed my approach to translation accordingly. This is how I came to enter the fifth period of my translating career: translating in the contemporary language, and I continued to develop even after that.

     My translations of The Merchant of Venice, Antony and Cleopatra, A Midsummer Night’s Dream, The Tempest and Macbeth are all early translations from my fifth period. Shakespeare’s tragic style can be extremely rhetorical, and in such places I have sometimes applied a classical literary style so as to prevent the language from sounding grammatically awkward. Up to then I had tended to use classical Japanese more on the assumption that that was how I thought Shakespeare should sound in Japanese, but if you translate in the old style, with all its distinct associations, then it becomes difficult to understand, and sounds too Japanese, ancient in fact, which is not the case if you translate in the contemporary. I feel that those famous old words which can be difficult even for English people to understand, expressed in complex, sometimes fearfully concise archaisms, richly metaphorical and bound up in old grammar, move me mysteriously and speak directly to my heart, and I believe that in the mouth of a Portia or Cleopatra, a Stephano or Bottom, Macbeth or Lady Macbeth, are to be heard in the Japan of today. When all is said and done, it is the warmth of the language that matters, its mysterious vitality and relevance. The natural and contemporary feel of contemporary colloquial language recalls the unchanging naturalness of Shakespeare’s works.

     Yet in the case of Shakespeare, even contemporary translations require a little qualification. For a contemporary language comprising six or seven parts the styles of inner and suburban Tokyo is weak in vocabulary, and is not up to the task. One is troubled by downtown slang such as oike (‘pond’) and oashi (‘sen coin’), which are derived from Edo usage and the language of the wet nurse but are completely redundant in Shakespeare. It is meaningless for Shakespeare to resound with the strains of Edo, Meiji, Taisho or Showa Japan, or the Yamanote suburbs of Tokyo (1), or the Ginza nightlife. Even in contemporary translation, it is hard enough to speak the language of even one section of one’s audience.

     I first translated Shakespeare in a free and capricious style before my mind became set on stage performance and my style shifted toward Japanese usage until finally ‘warmth’ (jomi) and ‘rhythm’ (choshi) became the two poles of my translations, and I went for a version of the contemporary a little removed from the colloquial. I believed for a long time that the most faithful translation was a literal one, taking it for granted that to confuse one style with another, whatever the linguistic differences, would be imprudent. Literal translations, where the language fluctuates because of the importance of rhythm in verse plays, may tend to verbosity, and warmth will supersede rhythm. Well-polished lyrical verse is something special, but with the translation of lines written for the more crucial parameters of the stage one must use one’s imagination. There is absolutely no point in attempting to translate metaphors and idioms in the style of the original. Metaphors like ‘an iron will’ and ‘a face as fair as a flower’ and the synecdoches one often finds in dialects, like oshaku (‘ladle’) and omakomo (‘straw mat’), can often harm the overall warmth and rhythm. For example, ‘Good morning, sir’ means something quite different in English to how we translate it in Japanese, Ohayo. Some people have translated it literally as Kocho, kimi yo (‘good’ + ‘morning’ + ‘you’), which sounds rather strange.

     The particular strength of Shakespeare’s writing is that through its special qualities and rhythms the reader experiences directly the feeling and personality of the characters. What the translator needs to give greatest attention to is to convey the lofty flavour of the words, the distinctions between high and low, the exchange of intelligent minds, the reality of good and evil, the sense of urgency. One has to translate for the stage (and without any additional explanation) the folly and solemnity, the negative capability and the chivalry, the politeness and the candour, the passionate and the cool, the long and the short, the tender and the frank, city and country; a verse translation will be more sensitive to these differences than a literal one. Shakespeare’s best known sayings become almost proverbial in the mouths of actors, which is to say his proverbs become living speech. The same thing may be said in an abstract way, a concrete way, or metaphorically. In translation this may become repetitive, verbose, or even redundant in part. If you translate in a universal colloquial style that elides classical, contemporary, refined and popular, then you will surely not fail. The wordplay of Shakespeare’s comedies may be broadly contained within colloquial language, since homophones in Japanese are more common than in English. Sometimes one can even produce a smoother translation than the original.

2. Preparatory advice

     For a volume entitled Sao gisaku shu (Shakespeare Translations), I translated twenty of the plays from Hamlet to Cymbeline, and these translations are representative of the approach I take to translating Shakespeare. Then, in the winter of 1925, Waseda University Press decided to publish my translation of the Complete Works. What follows constitutes the main part of my memories of how I translated those first twenty works; they explain my approach to translating Shakespeare, and so may be of some use to future translators. Of course, I have included only those points which seem to me most memorable.

     (1)   The characteristic of Shakespearean drama is the striking wealth of its vocabulary, which is multifarious and autonomous in nature. As I have often noted, the number of words he uses amounts to 150,000, including everything from classical archaisms to the popular language of the Elizabethan age, slang, dialect, foreign words, and neologisms. His language corresponds to every type of class and character of man. The same content may vary in style of expression according to person, time, circumstances, and temperament. The way that Shakespeare’s characters speak depends on their class, character, occupation, upbringing, locality, gender, and age. He can be frighteningly funny and frighteningly direct. He has a courteous side to him as well, but also an artlessness. He can be both articulate and inarticulate. Gentle yet intrusive, noble yet intimate, valiant, magnanimous, flippant, sincere and natural. Shakespeare’s pen delves into the heart of all things, and yet he is also skilled at mixing these styles, at weaving them together poetically, musically and mellifluously, above all in a way that is pleasing to be heard. Strangely enough, when some clumsy character is made to speak awkwardly in prose or slang, the sounds that result are not unpleasant but serve as a kind of musical accompaniment. Shakespeare when he is being rough and vulgar is never discordant. People who appreciate share (word and sound play) in Japanese can never tire of Shakespeare’s wordplay, which flows smoothly and rhythmically, and is particularly suited in fact to stage performance.

     (2)   The next feature of Shakespearean drama, and this is not uncommon among classical playwrights, is the care with which he arranges dialogue for the stage. He will give the first half of a line of blank verse to one character and the second half to that person’s respondent, or else he will break the rhythm with an interjection, or will share out a line among three or four characters, taking care not to break the pace for even a moment; these are all typical features of his mature works. If the translator does not allow space for breathing, then the actor will stumble over his breathing. In this and other respects, one must be quite faithful to the original, such is the extent of his karma. The relations between subject and predicate, main clause and subordinate clause, should be left just as they are, keeping the pauses of the original. English conjunctions such as ‘for’, ‘since’ and ‘because’ precede clauses, but if you try that in Japanese, where the sentence order is opposite, the language sounds stilted. If you reverse the normal sentence order and say Hayaku motte kite kure, mizu wo! (‘Bring me some water quickly!’), then it feels rather strained and disjointed, but if you say it the other way round, putting hayaku first and mizu after, then it sounds much more relaxed.

     (3)   The third feature of Shakespeare’s plays are those complicated soliloquies. They are more than simply isolated speeches but serve a rhetorical function within the rest of the play. To apply Macaulay’s distinction, the same content may be expressed abstractly or in an immediate and concrete way, to bring out the jargon, or emphasize the metaphors (2). The same thought or feeling is subject to diverse interpretations. When Japanese scholars first translated Shakespeare, they were surprised to discover this diverse style occurring so frequently. It seemed that Shakespeare created texts rich with feeling, stridently saying the same thing again and again. That is how it seemed, although in the end it was a technique born from the necessity of drama. Lines which are written to be heard rather than read need to be repeated, so that skilled reciters and experienced actors will be able to agree on their meaning without being able to explain them. Composers such as Wagner work out their logic through the methodical repetition of lyrics. If you translate the same thing without varying your expression, then you may naturally feel you are wasting something and harming the original. This is a particularly important point when translating Shakespeare with his rich vocabulary and universal rhetoric.

     (4)   Phrases that come at the end of a scene are aphoristic or proverbial. They foreground popular sayings, and it goes without saying that their translation is not as straightforward as might be imagined. Things can get a bit wild with all those long speeches and subtle witticisms. Various kinds of metaphorical language may be used to simplify rather than ornament the language. You do not always have to translate metaphors using yo ni or gotoku (‘like’, ‘such as’).

     (5)   Another point to beware of when translating Shakespeare is the treatment of pronouns and adjectives. You can omit seven out of ten pronouns in Japanese, and should always be economical with the use of kanji. For example, the word ‘true’ can be translated using two characters – shinjitsu – but there’s a problem with ‘faithful’. Unless you keep a mind to it, your translation may get too verbose, so translate ‘true’ as chu and ‘faithful’ as shin. If you translate Shakespeare strictly into the everyday language peculiar to the colloquial, you will be unable to render adjectives in any memorable way. You will either compress their meaning, or there is the danger you will translate into tedious long speeches. Another problem is what to do with polite forms. When a Shakespeare character says ‘Sir’ or ‘Your Majesty’ or ‘Your Highness’ or ‘My Lord’, there is this troublesome distinction to be made between ‘you’ and ‘thou’, and just as in kanbun (classical Chinese literature) you cannot make any serious changes to the word endings. Even with kings and princesses, it is usually not impolite to call out ‘O King!’, and of course you can adopt this appellation when reporting news. The correct Japanese would be O yo, yurushite kure (‘Forgive me, your majesty’) and one can also say Ima hime ga dekakete iru (‘The princess has gone out’) when things are not looking so good. Adjectives such as ‘dear’, ‘fair’, ‘sweet’, ‘good’, ‘gentle’ and ‘gracious’ are polite words, rhythmically bisyllabic, that may be adopted, but since they sound rather strange in a literal translation, it is better to adopt some kind of standard.

     Next the pronouns. There is no correct translation for ‘my father’, ‘my brother’, ‘your father’ or ‘your sister’, which are all prefixed with pronouns. The first two would be chichi ga and ani ga; the others would be goshinpu or otosan, and oimotosan or omeigo. This is to say that in Shakespeare translation one must first consider vocabulary and usage, and for this reason contemporary Japanese is simply inadequate. To rely on the dialects of the Yamanote, Tokyo suburbs or Ginza nightlife is terribly limiting; one must create a hybrid style that combines local dialect with slang words from Tokyo and its environs, such as subarashii (‘wonderful’), suteki (‘nice’) and hitokko (‘single child’). It should be a colloquial style that makes sense not only to the translator himself but also to the people who know him. Just as Shakespeare uses archaisms, foreign words, slang and dialect as the situation demands, the translator should do likewise: the refined and vulgar language of the ancients, the language of Confucianism and the Chinese classics, of the Tohoku and Kyushu, the language of Akinari, Bakin, Saikaku and Chikamatsu (3), these should all be exploited in an approximation of the refined and vulgar tongue. If one does not, then one cannot hope to capture even one ten-thousandth of Shakespeare’s original spirit.

     You obviously can’t use the standard Japanese used in primary and middle schools. Despite the arguments for simplifying the language by reducing the number of Sino-Japanese characters, you would soon realize their literary impracticability if you tried to translate just one of Shakespeare’s plays. To put something into contemporary Japanese is not the same as using that impoverished language which is daily spoken by a comparatively large number of people today or the standardized Japanese authorized by the officials of the Ministry of Education and educationists for use in textbooks. If contemporary Japanese grammar is to be standardized simply for the purpose of communication, it can be no match for the classical. A language limited to colloquial usage should be used for translating Shakespeare into Japanese. Unless one does so, one will be unable to capture the remarkable creativity of Shakespeare’s language.

3. Stage directions

     With regard to stage directions, there are a number of conditions of which the reader should be especially wary. As I have often noted, the division of scenes and even acts is often unclear in Shakespeare’s plays, and of course detailed stage directions entirely absent. In recent years, as a result of advances in theatrical standards, capital letters, parentheses, italics and other punctuation marks have been applied to the Quarto and Folio texts, and even stage directions have been identified with greater certainty. Yet since the subject is still under review, any claims that the stage directions of today, as they have been passed down with some additions and deletions since the days of Nicholas Rowe, represent Shakespeare’s original intention remain in doubt. The addition of detailed directions to those already in use (such as may be of general use to the reader) will probably be severely criticized by specialists for their audacity. Yet for readers new to Shakespeare, the question of whether stage directions are to be used at all, and their level of detail, is a serious matter. For scholars dependent on existing editions, the matter of to whom speeches are addressed, how speeches should be spoken and with what degree of emotion, are frequently ambiguous and often impossible to answer. When I tried to translate Macbeth and Hamlet for the first time, there were numerous points which puzzled me. The various commentaries available based their opinions on reviews of live productions of Shakespeare and on actors’ autobiographies. Presuming that these notes could be translated, I planned to devise my own commentary, and included detailed and previously unused stage directions: what Dr Johnson might have called ‘a necessary evil’. If I am to be accused of anything, then the fault lies mainly with these stage directions, but for scholars new to the field they are indispensable. What I regret is that before completing these translations, I did not have access to the then unpublished Nonesuch edition, whose stage directions are invaluable.

     If, while I was doing my translations, I had been able to obtain a copy of Harrison’s New Reader’s Shakespeare (4), then I would have saved myself quite a lot of trouble, but I had almost completed it when I saw a copy in Maruzen bookshop. Harrison’s directions are comparatively detailed compared to previous editions; they are not exactly minute, but they do seem to differ in both approach and purpose from mine. One suspects that they are mainly derived from recent stage performances. In any case, my stage directions have nothing whatsoever to do with Harrison’s.

4. Colloquial and Contemporary Language

     Colloquial language is not the same as contemporary language. The contemporary language as it is spoken is the language of one’s inferiors, and its generic boundaries are predictably narrower than the colloquial. Contemporary Japanese is confined to the large urban areas, for example to the everyday conversation of lower and middle class Tokyoites, popular novels, newspapers and magazines, and textbooks prescribed by the Ministry of Education. In the new Showa era, proletarian language has been widely accepted, slang is popular, even in mainstream society; other words have disappeared, double consonants are prevalent, the value of words has declined, as has correct usage quite markedly. It is impossible to translate great foreign literature using only contemporary Japanese. A writer like Shakespeare, with his wealth of vocabulary and lucid style unmatched in English literature before or since, can simply not be translated into the Tokyo dialect of today, even allowing for slang. (There are two or three brave souls who have tried, but to look at their translations, Ophelia and Juliet sound like bar girls and students, and Portia, Lady Macbeth and Gertrude like the proprietress of a tea shop or inn. They speak a vulgar language associated with the mistresses of company men.)

     Yet even this contemporary language may become less narrow-minded and richer in vocabulary. This is to say that just as lower and middle class speech adopt trend words, so too will local dialect become contaminated. Therefore, if one is to add to what is broadly defined as a contemporary style a language that is prescribed by the rules of the colloquial (whether in current use or not), but a Japanese that is influenced by contemporary and classical alike, by vulgar and refined language, by internal and external influences, then Japanese vocabulary will no longer be impoverished. Of course, one can include words which have entered modern Japanese through translation since the Taisho era, while dialect words and slang have already received general recognition. Japanese grammar has changed considerably since medieval times. Translators can use words whose historical changes have become widely accepted. In translation I sometimes intersperse Japanese words with foreign ones. Translating in this way, the Japanese becomes more flexible and pliant than I had expected. The vocabulary may not be listed in dictionaries like Gensen or Kokugo jiten or Daigenkai, but it is rich and broad.

     This style may not be suitable to the translation of other foreign literature but I believe it is most appropriate for Shakespeare. As in the source, where the aristocrats use archaisms, classicisms, foreign and translated words, and the lower classes use everyday language, slang and dialect words, and invent words of their own, everything happens in its own way. From the point of view of his freedom of expression, Shakespeare is like the ancient goddess Marishiten (5). His style is like that female buddha in its solemnity and subtlety. Its divine power has a mysterious sacred virtue that may take many forms. Neither sun, moon nor heaven can see it, nor could any man ever do so. Sometimes it has three faces, sometimes six or eight pairs of hands; it can fight with many weapons at once; it can hunt the boar, joust with sword, pike, bar and mallet, unleash the bow from its quiver, start a fire, flail a rope. Against this monster, we Japanese are armed only with our classical language of old. It would be the height of madness to suppose that our impoverished contemporary language can do battle with this monster, since to do so is to deny his classical appeal and to cater to mediocrity. Where the original is difficult to understand but mellifluous, where there are several words that resist even close reading, you will sometimes have to resort to awkward kango (Japanese words of Chinese origin) and invented words. The most important thing is not to popularize Shakespeare. One should first seek to translate with a feel for character and what it will look and sound like on stage. Like the recitations of Takemoto gidayu in joruri (6), there are many lines which do not make sense on hearing but do communicate mood. If one is to insinuate the feelings of a scene and truly tell the story, there will be those difficult words and phrases that serve to make the audience laugh and cry. Wherever and whenever, dramatic speech is always like that.

5. Notes on the revised edition

(1) Pronunciation

     The pronunciation of English in Shakespeare’s day was very different from either modern or recent times, and so there is a special way of speaking Shakespeare’s language on stage. I tried to honour this tendency in both my early translations and the succeeding twenty volumes, but in the Gisakushu (that contain my first twenty translations) there are mistakes in the pronunciation, and by applying Irving’s rules on Shakespearean pronunciation (7), I have tried as much as possible to rectify these errors. Even today we cannot always be sure of the pronunciation, although I have received some assistance from a professor at the Imperial University, and since Japanese pronunciation is seldom strictly consistent with the original, I have revised as best I can, knowing that a completely accurate transcription is impossible. To take a couple of examples, Lucio, Lucilius and Lucius become Ryushiyo, Ryushiriyasu and Ryushiyasu, Lucentio and Luciana, Ruchenshio and Rushiyana; it’s a problem! The maid Maria in Love’s Labours Lost is Mariya, but the housekeeper Maria in Twelfth Night is Maraiya. Caius in Julius Caesar is always Keiyasu, but Dr Caius in The Merry Wives of Windsor is pronounced Kizu. What is one to do with all Shakespeare’s French words? Anthony and Antony are pronounced the same despite the different spelling, as are Calpurnia and Calphurnia. Asia in Shakespeare is Esha or Eshiya, and Roma and Rumu both seem to be possible for Rome. There’s no end to these examples. Most nouns were pronounced very differently from today. For example, ‘sleep’ was pronounced ‘slip’, ‘sheep’ ‘ship’, and ‘on’ ‘one’.

     Shakespeare’s spelling is extremely complicated. It has long been known that Shakespeare’s name could be spelt in twenty different ways, possibly even forty; English spelling is much more troublesome than our way of writing with kana, and sometimes one has one’s suspicions. For example, the name of the greatest actor of Shakespeare’s day is usually spelt Burbage, but Burbadge is also possible, and I have seen Bourbage in an old book. The adjective ‘sultry’ is spelt ‘soultry’ in the Folio edition, which sounds quite different.

     The example of ‘sultry’ reminds me of the confusion I have had with words ending with -ri, -ni and -ii. For readers familiar with the pronunciation of the English names, the pronunciation of Henry, Gurney, Ketley and Wolsey as Henri, Gani, Ketori and Uruji, makes no difference, but for people with no knowledge of English the orthography would seem to be rather misleading. It is strange when ji sounds like the Japanese suffix -ji meaning ‘temple’, and tori means ‘bird’. So I changed them to Henrii, Ganii and Urujii. In words like aisu kuri (‘ice cream’) and hankachi (‘handkerchief’), we Japanese have the bad habit of removing the ending of the loan word.

     In the case of foreign place names and also Biblical names, I have sometimes followed Shakespeare’s own practice and adopted the pronunciation familiar to Japanese people. I have rendered the names of characters as accurately as possible but in the translations themselves and in the titles I have tried the old familiar pronunciations. Thus Juriyasu Shiza becomes Juriasu Shiza, Rōmiō to Dyūrietto becomes Romio to Jurietto.

(2) ‘You’ and ‘thou’

     Abbott, in his Shakespeare Dictionary, gives numerous examples of Shakespeare’s usage of these two pronouns. I cannot improve on his explanation of ‘thou’ as the form used for superiors, equals, inferiors, and for non-human creatures, such as dogs, although the rules are quite complicated. One can give a number of examples of its use toward inferiors, but those instances (for example, in colloquial translation) are rather perplexing. The use of the vocative in Japanese has decreased over time, and we are just about left with omae, kisama, anta and anata. At the time of the Meiji Restoration, sono moto, sono ho and kiko could still be heard, but no more. To give a couple of examples, Claudius in Hamlet addresses Laertes with ‘you’ and ‘thou’ in the same speech. Falstaff first calls Hotspur’s corpse ‘sirrah’ (kora, kisama) and then ‘you’. Both anta and anata seem to lack something as translations of this particular ‘you’. I cannot bring to mind any more good examples, and so must leave that to the next opportunity; this is just one of those points you have to be aware of when translating Shakespeare. Contemporary translations often ignore the distinction between ‘you’ and ‘thou’. We often use kimi or kisama wrongly in place of anta or anata.

(3) Lower class speech

     Many contemporary translations have tended to ignore the language of the lower classes, dishing out phrases such as ko shite okure and o shite okure (i.e. ‘please do it’) irrespective of class. Ko shiro and o shiro belong to the Tokyo lower and middle classes, and have been widely adopted as slang by upper-class women. Kings, princes, dukes, counts and marquesses, men and children, all sound the same in these translations. Especially when young Japanese use polite language, they forget those mysterious qualities of gentleness and pliancy which cannot be found in foreign languages. I’m thinking of usages like the polite prefix o- and the suffix -mase, as in asobase, which soften the verb. Phrases such as nande o-okori asobasu no desu ka? (‘Why are you angry?’) and doshite o-waraware ni naru no desu ka? (‘Why do you laugh?’), o-kimono and o-obi, are strung together with all kinds of colloquial verbiage. If one is translating plays from a feudal age, it is inappropriate for noblemen to speak with the rough tongue of the people. Between asobase and the o prefix, one finds a suteki and a subarashii, a chippoke (‘tiny’) and nannan nuki, an icchatta and a nani nani shiteru wa yo (‘doing something’), a true linguistic cocktail. This is women’s language, of course, and class discrimination will be especially high among women. As I’ve said before, the language of those who wait on the nobility, such as Juliet’s nurse, and Mistress Quickly in King Henry IV and The Merry Wives should be inferior. Japanese contains numerous pronouns with which to refer to oneself and other people. It’s extremely complicated. It has long been assumed that the pronouns that refer to oneself all begin with wa, whereas those referring to others start with ka or na, but in translating classical texts there are several strange and interesting variations to be observed. This might make greater sense if I were to give a few examples, but I am in the middle of revising my early translations, the weather is far too hot, and so I apologize for leaving this argument incomplete.

(4) Selecting vocabulary

     In a nutshell, I believe that the vocabulary of old Japan is poor, and that both Japanese vocabulary and grammar are in need of reform. One cannot forgive lightly those reckless individuals who freely confuse ancient words like namida gumashii (‘crying’), hitozuma (‘wife’) and shijima (‘quietness’) with the language of the café and local dialect. Such people take excessive liberties with old Japanese. If we go back a hundred, even three hundred years, we can find plenty of suitable words, and so I find it quite unacceptable to invent new words that are impure, unpalatable, unintelligible and corrupt. We should experiment with words that are less corrupt. In his biography of Shelley, Symonds notes that the poet draws on the whole breadth of the English language (8). If one were to make such exhaustive use of the Japanese language – from the Kojiki to Nihon Shoki, from Norito to Manyo (9), the literatures of the Heian, Kamakura and Muromachi eras right through to the three centuries of the Edo era, and the few decades of Meiji – that would be quite a trawl! There are many words whose orthography has changed, but also words like keibetsu (‘contempt’), sekihai (‘narrow defeat’) and renko (‘lead away’) that have remained unchanged, and these must be adequate for the time being. It’s not just the words. ‘Which’ clauses such as nani nani wo nashi eta (‘did something’), nani nani no hoka no nanimono demo arienai (‘cannot do anything else’) and nan to nan no nani nani de attaro koto yo (‘would be a bit of this and that’) have – to my slight shame – become standard usage in the Japanese of today. They resemble – though only superficially – phrases such as samui desu (‘it’s cold’), karai desu (‘it’s spicy’) and iku desu (‘I’m going’) found in the styles of Tayama Katai, Otsuji Shiro and Inoue Masao (10), which makes them all the more intolerable. It is only a short step from Shikari, heika (‘Yes, your majesty’), Iya, heika (‘No, your majesty’) and Kocho, kimi yo, to a corrupt language. That is how mistakes creep into translation: through blind imitation. It is not just incorrect Japanese but it also makes for incorrect translation.

(5) Free translation

     Literal translation is the most faithful style of translation, but as a result is lacking in warmth and rhythm: often the blend of meter and rhythm is unavoidably awkward and the style rather strange and comical. When you translate haiku and tanka into foreign languages, a literal use of the original form may be thought expedient, but translating foreign literature into Japanese is something else. If the translation is too direct – apart, that is, from bilingual translation – then it will be unintelligible without notes. In dramatic poetry, where the emphasis is on feeling, warmth and individual character, then one will try to recreate the speech of a person, scene or dramatic moment. For example, a direct translation of ‘Yes, your majesty’ would be Sayo desu, heika, whereas the correct Japanese is Sayo de gozaimasu. Polite verb forms such as gozaimasu and asobase carry the same semantic weight as ‘highness’, ‘my lord’ and ‘sir’. It is the same with Ophelia saying ‘Aye, my lord’ and ‘There, my lord’ to Hamlet, but when we think of the petite and reticent Ophelia, Japanese equivalents such as Yoroshu gozarimasu and So ouketori kudasaremase seem quite wrong. How about Hai, dozo or Sa, dozo? ‘My lord’ does not mean the same as dozo, but does express a similar level of politeness. Another example is when Polonius approaches Hamlet in his books, and asks what he is looking at, to which Hamlet replies ‘Words, words, words.’ A direct translation is Kotoba, kotoba, kotoba, but that lacks the necessary connotation. Monku, monku, monku sounds better, and is easier to understand when spoken on stage.

     Translating similes, metaphors, metonyms and synecdoches as they are sounds most unnatural, and often interferes with the sense of the original. The metaphors of Chinese poetry, English poetry and above all anything classical do not necessarily represent the writer’s original ideas. There are many hackneyed examples like ‘iron will’, ‘a face as fair as a flower’ and ‘a back bent like a willow’. It would be ridiculous to become slaves to such expressions. The language of drama is intended primarily for the ear; it is unreasonable to apply the same criteria as the translation of lyric poetry, which is read by the eye, and of novels.

     Translating songs is another thing. Instead of treating each phrase as it comes, it is better to aim for a unity of harmony and rhythm. An old folk song should sound like an old folk song, a children’s song like a children’s song, a dirge like a dirge, a song of congratulation like a song of congratulation, a jig like a jig, and ideally jazz like jazz. It would be strange to translate the song in Ophelia’s mad scene in high flown language or in a mixture of the ornate and vulgar.

     The purpose of dramatic translation is to convey the style (iki) of the original in the words spoken by the actors, which means that sometimes one has to discard verbosity, cliché and painful metaphors. A certain foreigner tried to translate Mokuami’s famous plays Kirare otomi and Benten kozo, but came unstuck with lines like mage mo Shimada ni yui ga hama and irokoi mo Satsuna toge no gakeppuchi (11). Isn’t it enough just to capture the basic style of the original?

(6) Economy of meaning and rhythm above verbosity

     Shakespeare is admired for the unprecedented wealth of his vocabulary. In the present century, when a loud debate on what is and is not canonical has arisen, it is difficult to tell whether this wealth of vocabulary is really Shakespeare’s. If Shakespeare’s works were written by seven or more other people, or if they were the joint work of three or four, or else corrected by seven or eight, then Shakespeare’s actual contribution would be only 50% or less. My view on the matter is not wholly serious. When it comes to the wealth of Shakespeare’s vocabulary, particularly with regard to translation, what one should note are the gradations of rhythm and meaning hidden in Shakespeare’s plays, the nuances, universality and compactness of those delicate transpositions. Readers will be impressed by the vibrancy of those distinctions in rhythm and movement between each kind of character – king, prince and nobleman, religious and fools of every class – to every kind of merchant, artisan, servant, man, woman, old and young. Male and female of every class and character are powerfully expressed. With a writer as well known as Shakespeare, there were at least some thirty or forty court intellectuals who could have helped him. There is an insubstantial theory that one of these was the female novelist Mary, Countess of Pembroke: the Murasaki Shikibu of Elizabethan times.

     Since even women’s expressions carry a universal weight in the way they are conceived and as women write them, distinctions between high and low, intelligent and stupid, old and young will be written very differently by people depending on whether they are provincials, foreigners, madmen or fools. Indeed the character of people is not only a matter of the style of language they use. Degrees of feeling are also expressed through subtle linguistic nuances. The familiarity and respect which other people show us is registered through their language. The language of every character is vital, and if properly respected, even stage directions are unnecessary; psychological modulations are easily grasped, as if you are looking at them face to face.

     That is only if you can read the original well enough. Reference to other Japanese translations is not helpful in this regard. It has been said that a certain German translation of Shakespeare is better than the original but I have no doubt that it is completely different. It is rather like the box in the tale of Urashima Taro. Once Urashima opens it his spirit vanishes in a puff of smoke. 

     If in this sense the true value of Shakespeare is difficult to convey even in languages with common roots, such as French or German, then the amateur efforts at translating Shakespeare in the early Meiji era should have been impossible. That they should have sought to translate Shakespeare with our cryptographic writing system, our mass culture, our far from standardized spoken language, with the various obscurities surrounding Japanese names, with foreign loan words. I wish to request reformers of our language that they take into serious consideration the complex demands of Shakespeare translation when they discuss issues such as limiting the use of Sino-Japanese characters, standardizing the language, reforming punctuation and so on, and that afterwards they publish their findings.

     It is worth giving some thought not only to Hamlet but also to how the works of any famous foreign playwright might be performed in Japan. If we are to stage in Japanese plays held to be exceptional in their original language, we must at first consider a suitable style of translation, whether this be the modern style of the Hongoza and Shintomiza theatres (which do not make the distinctions of traditional kabuki) or else the hybrid literary style popularized by Morita Shiken (1). If, for example, we wish to capture something of the spirit of the past in translating plays from feudal times, then we will want to use language peculiar to that age, and thus the style of the Kojiki to raise the spectre of the age of the gods (2). 

     Secondly, we need to consider how we interpret a play when it is brought to the stage. Should we heed the opinions of foreign scholars, or else take our cue from traditional theatrical practice, or indeed the techniques of our well-known contemporary actors? And so long as we do not mistake the author’s intention, then would it not be better to provide our own new and uniquely Japanese interpretations?

     For example, there are numerous differences, both straightforward and more complex, between Chushingura as it was originally written by Takeda Izumo and as it is performed nowadays (3). In the characterization of Yuranosuke various discrepancies have emerged between the original text and the personal preferences of actors, which is why there is no single version of the play shared by the kabuki and puppet theatres, by actors of past and present, including those of today like Danjuro, Sojuro, Danzo and Udanji (4). Famous foreign dramas are psychologically more complex than our own, and their characters are more difficult to interpret, so that a character as unique as Hamlet produces all the greater a variety of interpretations. Just as the opinions of scholars and theatre critics are not necessarily accurate and to the point, nor are the most popular actors necessarily faithful to a text.

     The sensible way to proceed is first and foremost to imitate the best of the foreign actors as they perform a role like Hamlet with their refined technique and reference to the latest critical opinion, but if in the end our understanding is insufficient, or else if we do understand in part but find it hard to gain the cooperation of our actors, or rather if we do try to capture faithfully the hidden meaning of a foreign play and convey that to our own culture, and – setting fidelity as our standard – concentrate exclusively on reproducing the illusion of foreign acting styles, what in the end will we have achieved? If a famous British actor like Beerbohm Tree were to act Yuranosuke in the style of Danjuro, he would surely fail by a long shot. Would it not be better if, rather like Madame Butterfly and The Darling of the Gods (5), he were to make his own interpretation of the role of Yuranosuke, although with some reference to the original kabuki meaning of course? For Japanese people the role is a comic one, and so if Takeda Izumo’s play were to create the same impression among foreigners as it does among us Japanese then it might be accounted some kind of success. Probably if Beerbohm Tree were to perform in the style of Danjuro, then foreigners would find it different although not strange, and in fact be rather drawn to the performance. The issue of whether or not it is appropriate serves, at any rate, to illustrate my first question.

     To return then to the question of how Japanese actors should perform Shakespeare’s most controversial play, Hamlet, in Japan, I should now like to consider what approaches they have taken and the various styles of Shakespeare translation. I do not recall the exact details, but when I translated part of The Merchant of Venice for experimental production by the Bungei Kyokai, although I had followed the original text, I was criticized for my conspicuous use of old Japanese language. I do not know why this should have been so, but it was probably to do with its basically old-fashioned, pre-Meiji style. Perhaps my critics felt that I should have adopted the Hongoza style or the hot-blooded style of the Iiza theatre for the more intimate moments (6). If I had done so, then it would have been quite different from Shakespearean drama.

     It does not need to be said that Shakespeare’s plays for the most part are intricately rhythmical in their language, and that about 70% of the lines are written in wonderful unrhymed verse, which is more systematic in structure than the seven-five syllabic meter of Japanese prosody, and should never be translated in the unadulterated prose style of contemporary usage. If translated in prose, much of the feeling of the original will be lost due to the lack of poetic style, and in many places it will be impossible to harmonize the meaning with the literary form. Having said that, since Shakespeare’s particular strength is to blend seven parts verse with three parts prose in his arbitrary manner and to blend his historical and contemporary – what is more, his decidedly ‘plain’ contemporary – in his free and easy manner, a graceful classical style will be unsuited to this blending of the poetic with colloquial prose. Chikamatsu and Shakespeare share this interest in the plain style, which is evident when one compares them with the works of more academic writers who imitate classical styles. At the same time, the rhythms of Shakespeare are considerably loftier than those of Chikamatsu, more diverse and profound, and thus the language of Chikamatsu is insufficient in itself for translating Shakespeare. For an appropriate style we would have to go back as far as the Taiheiki, Seisuiki and Heike monogatari, but opting mainly for the historical styles of the Genroku and Kyoho eras and traditional kigeki comedy, a hybridized version of the syllabic meter of Bunka and Bunsei era literature, and Mokuami’s early plays, would also be a good compromise (7). Well, it would only be an imitation, and the Mokuami and Bakin styles by themselves would not do. Shakespeare’s speciality is to draw a thin line between historical and contemporary, knitting together high and low into a single glorious epoch, which is one key as to why his plays have lasted to this day. He is different in this way from even the best of Ben Jonson.

     The Elizabethan age is basically similar to our own Genroku and Kyoho eras in the way that the latter broke with the Momoyama culture of the previous century, and yet displays some of the qualities of the 10th century Heian era, being on the one hand extremely aristocratic and feudalistic but at the same time democratic in its openness to the wave of individualism known as the Renaissance. Generally speaking, the kings and queens of 15th and 16th century Europe were not as dignified in their displays of power as our own imperial household of today. Corresponding more with the Tokugawa shoguns, the fictional monarchs and princes in Shakespeare’s plays, such as the prince of Denmark and the ancient British kings of King Lear and Macbeth, are of a lower rank than Japanese emperors, somewhat comparable in other words to the lords of our feudal past, to the local tandai commissioners of Kyushu and the kanrei officials of Kanto (8). When we recall how Hamlet’s father sleeps in his orchard unattended by pages and courtiers, we realize the populism of Shakespeare’s plays: that a frankly unrestrained domestic style might go well with the classical thoughts of Shakespeare’s kings and queens. Moreover, where Shakespeare uses only colloquialisms, and as in the extreme case of Hamlet where one would supposed that his royals would not use such colloquial language even for the uninhibited Elizabethans, then a very classical and graceful style of translation is in the end inappropriate. In particular in the case of Hamlet, who changes in his ‘antic disposition’ from one moment to the next, his lines can only be translated by resorting to the kabuki style of Chikamatsu. I believe it is ultimately impossible to capture Hamlet’s characteristic of detachment through the hybrid Shiken style or more generally in Meiji Japanese.

     Shakespeare’s rhetoric does seem somewhat similar in its ordinary rhythmicality to that of kabuki drama. The sound ‘r’ is pronounced with a rolled tongue so that a word like ‘friend’ when emphasized at the end of a line comes out as furienzu in Japanese pronunciation. In Shakespeare, the syllables at the end of each line are stressed. Stresses in words such as ‘myself’ and ‘my lord’ indicate how these words were spoken long ago. The rules for saying words such as konnitta (‘today’) and zenaku (‘good and evil’) are similar to the convention in Noh chant whereby the beat is stressed (9). I believe, therefore, that as times have moved on and rhetorical styles changed considerably through actors such as Henry Irving (10), caution should be exercised with the translation of original texts in the significance we attach to meter and in what we do with the archaisms, and yet, to repeat my previous point, the seven-five syllabic meter of Bakin remains taboo, and the traditional narrative style is too verbose, and is thus infeasible. The limp rhythms of seven-five meter are quite incapable of conveying the energy and crackle of the original text. Shakespeare’s iambic pentameters, for all their formality qualities, can sound quite free and informal, even on occasion almost like prose, and in this respect inevitably remind me of Chikamatsu.

     For these reasons, my translation of Hamlet is of course only a provisional attempt, and since it is intended for stage performance, I have imitated so-called classical forms, in native terms the classicism of Noh drama against the romanticism of Chikamatsu. It seemed only natural, therefore, to adopt first a mixture of Noh chant and other styles, but I ended up rather perplexed as I realized it would not do for my translation to have the sustained tense feeling of Noh chant. That was an aborted effort. My next concern was the little ditties Hamlet sings to himself, since I knew nothing about how they were meant to be sung. I guessed they could not be sung with the kind of feeling to which Japanese people are used from Genroku kabuki, even less in the manner of shintaishi (new-style poetry), and so I improvised with a kind of ballad in kyogen style (11). Ophelia’s songs contain some vulgar lyrics, but their mood and melody were well suited to my improvised style.

     At this level, translation style also relates to the actor’s gestures and delivery, which is the great challenge of translating for Japanese actors. For Japanese actors to portray Westerners on stage, and not just Westerners but English people, and not just English people but those of Elizabethan times, and not just Elizabethans but those of Shakespeare’s plays, and to portray such unique and yet diverse characters as Hamlet, Claudius and Polonius, each with their own authentic story, is without question intellectually infeasible, but whatever Shakespeare scholars might say, if we are to grasp the bull by the horns and assert that human feelings are fundamentally alike, we should be able to find some ways of portraying these roles.

     As I said to begin with, whether we should persist in our obeisance to existing theatrical models or else look for new styles and interpretations under the pretext of ‘Shakespeare by and for the Japanese people’, ‘that is the question’. I have been alone in favouring the second of these positions, and while my view is no doubt reflected in the performances of the Bungei Kyokai, my reasons are rather complicated. If I were to explain what I have been saying with regard to my translating style in even twice the number of words, I suppose that only very few non-experts would be able to understand what I would say, and I would like to apologize here for this failure. This is the view of myself and the Bungei Kyokai, and since others are welcome to disagree with us, if we are going to be criticized, or else I am required to defend my company, then may the above arguments suffice. One point I would add is that the new styles and interpretations I have mentioned are certainly not intended for self-serving purposes, since my sole aspiration can only be to grasp the author’s original intention. Not even famous foreign actors can necessarily be said to be set solely on finding new interpretations. In the case of Hamlet, even the better known interpretations of the play have followed numerous directions since Goethe and Coleridge, and when contemporary productions in Europe and North America add something to the text, and the differences between actors’ interpretations are so extreme that they seem completely opposed to each other, there can be no one with the author’s authority to say which is exactly correct. That is something I would like readers to understand.

     The week-long production of Hamlet is a colourful addition to the literary world, attracting wide artistic interest. I received an invitation, and went to see it, but due to some slight problem had to arrive and leave early. It is to my regret, therefore, that my eyes and ears were deprived of the spectacle unfolding itself, although the impression I took home with me of what I did see of this lengthy work was certainly a vivid one. A number of points raised themselves which would be difficult to broach directly with either Dr Tsubouchi or his actors, and so during the performance at least I restrained myself. My publication of them now is no more than a pitiful attempt to give substance to my respect for the enthusiasm of Dr Tsubouchi and the efforts of his actors.

     Hamlet is a play written about three hundred years ago in England. It is unrhymed, written in so-called blank verse with five beats to the line. Based on their awareness of these superficial features, one might well expect the minds of modern Japanese audiences, whether appreciative or critical, to be made up about this play before reading it. What I mean to say is that rather than reading Hamlet with a belief bordering on superstition that its concerns are closely bound up with the realities of modern Japan, I prefer to take a more critical stance.

     Perhaps I can make myself clearer by alluding to a few facts. If one were to ask the several thousand people who saw the production whether they had enjoyed it so much that they had lost all thought of themselves and become completely absorbed in the action, then there probably would not be even one who could say that they had. I have no doubt in my mind that there was such a discrepancy of interest between the play and the audiences.

     If I were asked by Dr Tsubouchi to explain this difference, I would want to respond that it is England, three hundred years of history, and a torrent of unfamiliar poetic language that come between ourselves and the play. I would state unflinchingly that a man called Shakespeare was standing up there and ruining all our pleasure. If the gap between Hamlet and a Japanese audience is to be properly closed, we should not need England or three hundred years of history or the poetic language or all those troublesome adjectives. Hamlet by itself is enough.

     Dr Tsubouchi’s translation comes across as a model of fidelity and respect for the original text. It is hard to imagine the pains of translation unless one has experienced them for oneself, and in that respect I have a deep regard for what he has done. Yet it is to my profound disappointment that it is precisely because the Doctor is so faithful to Shakespeare that he ends up being unfaithful to his audience. He uses not a single word or phrase to appeal to Japanese psychology or customs. To the very last, his distorted Japanese follows the original to the word; the discrepancies are painful to behold. The translation has no room for the basic qualities of Shakespearean drama. In daring to translate the play, it is almost as if he has turned his nose up at us Japanese. The translation itself may be satisfactory but to hope that he can satisfy a Japanese audience in the theatre is like offering a sweet tooth French wine in place of sweet Masamune saké. Rather than being such a faithful translator of Shakespeare, the Doctor must choose between giving up the idea of staging his translation, or, if he is to go ahead with the performance, of being unfaithful.

     To compare Shakespeare’s works with a mirror reflecting the works of nature, unconditionally accepting the judgment of Westerners, is to relegate our own tastes and is thus a disgrace and mutual loss. I feel, indeed I would insist, that to put it around that Shakespeare is some kind of authority on reality is a considerable lie. There might be some impartial purpose in the linking of paragraphs to reveal the causes and effects of joy and anger but for such expressions to be cloaked in joy and anger is repulsive, unnatural, outrageous. Such an idea has never been used as a vehicle of mutual understanding by either the Japanese of today or the English people of today or of Shakespeare’s times.

     Once one realizes that in actual fact this unnatural and reckless way of thinking, which has only the slightest human connection with the Japanese people, is none other than Shakespeare’s poetic vision and thus a skilled form of expression that lifts the commonplace and conventional to another world, then this so-called Shakespearean drama, quite apart from stimulating audiences away from the impartial vicissitudes of dramaturgy creates a unique kind of poetic land, and unless one is a native of that land, one is denied the right to savour its pleasures; that is the particular challenge of Shakespearean drama.

     If you are the kind of person who just goes along with the story, as anyone can do, but ignore the poetry, or else are unwilling to make the effort to understand Shakespeare’s poetry, then you will incur nought but frustration and mental conflict. We must be ruthless with ourselves in this regard. Yet Dr Tsubouchi and his actors would probably regard as childish the interest that their audience has in Shakespeare’s capricious powers of poetic expression. This is a point that I should particularly like to emphasize.

     In my own experience, although this is also the opinion of European scholars, Shakespeare creates a poetic world which is out of the ordinary in character. The way to appreciate this world is through years of study gradually to become conscious of the natural state that lies behind it. We are so far removed in time and place that our hearts can never beat as one with Shakespeare, but if we spend enough time looking at the words on the page, then their hidden meaning will reveal itself in all its depth and fullness. Even in the case of a simple haiku, experience teaches us that if we read it in the same cursory fashion as we look at advertisements then neither our minds nor hearts can share in the riches that reward a more thorough reading. When I attended a performance of Shakespeare in England, the speed of the dialogue was faster even than a steam train running through Hakone. Even for English people of today who have received a normal education, it is often the case that Shakespeare’s words appear too poetic and lacking in sense when uttered on stage. If as a result people are unable to enjoy those rhythms that emerge from the particular arrangement of accents in recitation, I feel that most of them will not bear sitting long hours in the theatre. Shakespeare was a poet, and poets steal fire from heaven, but even more than that, we have to become aware of the magical force of these words to appeal to audiences with rhythms that transcend common sense. Shakespeare’s lines, just like Noh and utai, have this peculiar rhythm and timbre that must be grasped decisively if one is to sustain an audience’s interest. Neglect this point, and one ends up losing both the poetry in a phrase like seiran kozue wo fukiharatte (‘brushing the treetops with mountain mist’) and the colloquial force of a phrase like Oi chotto kite kure (‘come here a moment’).

     Shakespeare’s plays are nowadays frequently played in Western countries, and the critics there always complain that actors do not understand the poetry and speak it in a haphazard manner no different from prose, ruining its natural beauty. This criticism is even leveled at real poetry arranged in metrical form. Dr Tsubouchi’s Hamlet, even through its simple intonation, does not reproduce the poetic beauty that Shakespeare achieves in compensation for his distance from reality, so that we cannot be seduced by its elegant mystique; nor does it enable us to be fascinated by the sight of ordinary human beings acting on a stage.