Shakespeare’s rhetoric does not stop with whatever words that Shakespeare is thought to have written but communicates itself to Benedick’s ‘furthest inch of Asia’ (Much Ado About Nothing, 2.1.244-5), and so to modern Japan. This is a time and place Shakespeare that does not know, but translators can seek to know Shakespeare’s rhetoric so to recreate their response in translation; the translator can never ‘be’ Shakespeare, but can end up doing something similar to what Shakespeare was doing. Shakespeare’s plays were first translated into Japanese in the late 19th century, and while 20th century translators such as Tsubouchi Shoyo and Fukuda Tsuneari were relatively certain of what Shakespeare could and should mean in their culture, their reluctance to admit to anything like an ideal style of Shakespeare translation is reflective of the assumptions that have guided translation theory as a whole since the 1960s. 

     One typical assumption is, as Anthony Pym comments, that ‘no text fully determines a translation of that text, if only because translations rely on observations and interpretations’ (Pym, 26-7). Shakespeare translations are to some extent determined by the rhetoric (which is to Shakespeare’s credit), but their basis in observation and interpretation inevitably gets in the way of attaining equivalence with the source texts; it is the effort of translators to attain the illusory goal of equivalence that may also account for their creativity. Ernst-August Gutt (2000) goes even further in suggesting that while translations cannot be reduced to fixed meanings originating in the source text, translators can transfer to readers a way of making sense of the source material that is derived from their original and decisive efforts to interpret it. 

     Shakespeare translators, whether intentionally or not, teach their readers how to read Shakespeare. Shakespeare’s rhetoric, as it categorizes and compares data, also teaches readers how to make sense of ‘Shakespeare’, and as an active, communicative medium influences translators in the choices they make. Shakespeare translators can therefore imitate the rhetorical processes of Shakespeare, but this is directional equivalence that is described according to the changing norms of the target culture and audience rather than by reference to any absolute ideal. Shakespeare translations, no less so in Japan than elsewhere, are notoriously bound by the historical norms against they which they are produced. 

     Among a broad range of stylistic and historical approaches, the study of prosody in Shakespeare translation, their use of rhythm, rhyme and also internal rhyme and word play, is essential to appreciating the efforts of translators to attain equivalence. Prosody is fundamental to Shakespeare’s rhetoric, and any sense of the differences between Shakespeare’s English and a language like modern Japanese must begin with the sounds or phonologies of the contrasting languages that govern their respective prosodies. Prosody connects translators with the poetic norms of their target language, and enables them to make the interpretive connections across the texts of translations that in turn fulfill rhetoric’s task of helping readers and audiences to make sense. As one example, this article looks briefly at the use of accentual rhythm in modern Japanese Shakespeare translations.

     Shakespeare’s metrical prosody relies on the regular distribution of stressed syllables across the phrase and line, most typically the iambic pentameter, whereas Japanese is not a stressed language and relies instead on changes in pitch from low to high to low pitch and so on: what is called pitch accent. The precise features of Japanese pitch accent are exceedingly complex, not least because of the dialectal differences between standard Tokyo Japanese (hyojungo) and the accents heard in the western Kansai region and other parts of Japan. Nevertheless, a number of basic points can be made. The first is that, although Japanese translators often use dialect for the speech of comic and lower class characters such as the mechanicals in A Midsummer Night’s Dream, the language of Shakespeare translation in Japan is a literary version of Tokyo Japanese, and with allowances made for regional actors, the accentual system heard in Shakespeare productions is therefore that of Tokyo Japanese as well.

     The accents on individual words can be checked in accent dictionaries (e.g. NHK Hoso Bunka Kenkyusho 2016), according to which a high proportion of nouns (including proper nouns) are accented, but verbs less frequently and particles almost never. These accents can be transcribed, as in the examples below, where I have indicated a slight rise in pitch in the first mora of some words with an oblique stroke and the accented morae of words are numbered 1, 2, 3 and so on. The accents noted in my transcription are only dictionary accents, since accentuation is generally influenced by the overall structure of the phrase and sentence. In Tokyo Japanese, the pitch rises on the second mora of the phrase, is intensified on the first accented mora, and then falls away over the rest of the phrase, so that, as J.F. Backhouse notes, ‘within a single phrase, the basic accentual contours of individual words are smoothed into a single contour’ (Backhouse, 35). More than in everyday conversation, these structures are likely to be emphasized and even exaggerated in performance in order to project lines among a live audience.

     Japanese pitch accentuation is a habitual process that is acquired in childhood, and is therefore quite different from the stresses heard in Shakespeare’s blank verse. With blank verse, readers and actors recognize stresses as a shared literary convention, but whether a Japanese translator is instructing the reader to shape the line in a certain way is far more ambiguous. Through my research into Japanese translations of A Midsummer Night’s Dream (Gallimore, 35-55), I have ventured to suggest that the distribution of pitch accents might be correlated with logical semantic patterns, and might in that sense have an interpretive function, but I would ‘stress’ that this is not a literary feature recognized as such within the norms of Japanese poetics and Shakespeare translation. What can be said with more certainty is that pitch accent, along with lexical choice and sentence structure, contributes to the distinctive rhythms and styles of individual translators, but do not provide the kind of metrical template familiar to students and actors of Shakespeare in the source language that is considered essential to interpreting the text (e.g. Pye, 153ff.). As is seen in the following examples, in Japanese translation accent distribution works with lexical choice and sentence structure to generate distinct, if irregular rhythms that indicate how the line is spoken. These rhythms seem to me primarily to give the line its pace and momentum, and while they are not exactly key words and phrases, it is worth examining the words and phrases that are accented.

     Theseus’ opening speech in A Midsummer Night’s Dream (1.1.1-6) is a lyrical opening to a lyrical play that demands to make sense and to be spoken well in both English and Japanese:

This is how Odashima Yushi in the mid-1970s (Odashima, 8) and Matsuoka Kazuko in the late 1990s (Matsuoka 1997, 9) translated the speech. Odashima’s translation contains a total of 156 syllables with 25 accents. Matsuoka’s is slightly shorter with 142 syllables, and also has 25 accents. Phrases with more than one accent such as furui tsuki and wakai atotsugi are likely to lose the second and ensuing accents. The back translations indicate the lexical emphases; particles (e.g. the topic marker wa and subject marker ga) are marked ‘P’.

by the way - beautiful Hippolyta - our P - wedding P - time P

majika ni sematta. Tanoshii hibi wo ato yokka sugoseba

soon P - has pressed - enjoyable days P - after four days - will pass

shingetsu no yoi to naru. Daga nanto modokashii koto ka,

new day P - evening - P become - but - how! - impatient - thing ?

kono furui tsuki no kakete iku no ga. Watashi no nozomi wo nakanaka

this old moon P - wanes P - P - my P - hopes P - rather

kanaesasete wa kurenu, mamahaha ya miboujin ga itsumade mo

grant P - does not give - step-dame or - widow P - forever P

ikinagaraete wakamono ni yuzurubeki zaisan wo kuchisaseru you ni.

live on young people P ought to hand over inheritance P wither like P

Saa, utsukushii Hiporita, watashitachi no konrei no toki mo

well - beautiful Hippolyta - our P - wedding P - time P

chikazuita. Shiawasena hibi ga yokka tateba

approached - happy days P - four days - will pass

shingetsu da. Daga, aa, furui tsuki ga kakeru no ga

new moon - is - but - oh - old moon P - wane P - P

nanto osoku omoeru koto ka! Tsuki wa watashi no yokubou ni matta wo kakeru.

what! - slowly - can think - thing ? - moon P - my P - desires P - waited - P wane

Marude mamahaha ya miboujin ga nagaikishi

just like step-dame or widow P outlive

wakai atotsugi ni yuzuru zaisan wo suriherasu you na mono da.

young heir P hand over inheritance P wear away like thing is

It is noticeable that in both translations the accents fall on key lexical items: in Odashima, ‘now’, ‘beautiful’, ‘Hippolyta’, ‘time’, ‘near’, ‘approached’, ‘enjoyable’, ‘after’, ‘exceed’, ‘new moon’, ‘become’, ‘but’, ‘what’, ‘irritating’, ‘thing’, ‘old’, ‘moon’, ‘run’, ‘satisfy’, ‘mother’, ‘widow’, ‘forever’, ‘live long’, ‘inheritance’, and ‘cause to decay’; in Matsuoka, ‘well’, ‘beautiful’, ‘Hippolyta’, ‘time’, ‘approached’, ‘days’, ‘elapse’, ‘new moon’, ‘but’, ‘oh’, ‘old’, ‘moon’, ‘wane’, ‘how’, ‘be thought’, ‘thing’, ‘moon’, ‘waited’, ‘suspend’, ‘mother’, ‘widow’, ‘live long’, ‘young’, ‘heir’, and ‘inheritance’.

     It was Odashima in the 1970s who standardized the use of the free verse line in Japanese Shakespeare translation, and indeed both translations adopt this format. In free verse translation, each line is of about 25 syllables in length, which is about the most that an actor can utter in a single breath. In the example I have quoted, Odashima makes striking use of this metrical structure through the enjambments that occur between four out of the five line breaks. These, in turn, support a Shakespearean movement of the rise and fall across the speech that may in turn be supported by the accent distribution, for example Tanoshii hibi wo ato yokka sugoseba / shingetsu no yoi to naru (‘four happy days bring in / Another moon’). Matsuoka is more abrupt and ‘masculine’; her Theseus is the successful general who does not waste his words and is in confident command of his wife-to-be: Shiawasena hibi ga yokka tateba / shingetsu da. Moreover, the strong accents on tanoshii (‘enjoyable’) and modokashii (‘irritating’) in Odashima’s version suggest in this defining opening speech will be about the withholding and satisfaction of pleasure, and the accents in nanto osoku omoeru koto ka (‘methinks how slow’) in Matsuoka’s that hers will focus on the passing of time.

     This method of reading the prosody can be extended across the substantial body of Japanese Shakespeare translation, for example to the following of Macbeth’s soliloquies (1.7.1-7) in Matsuoka’s translation (Matsuoka 1996, 41):

Yatte shimatte, sorede yatto to keri ga tsuku nara,

done - done - then - at last P - be settled - if

sassato yaru ni kagiru. Ansatsu to iu ami wo nage

quickly - do - P - limit - assassination - called - net P - cast

kekka wo taguri yoserareru nara, nakimono ni shite

result P - draw in - be drawn in - if - dead person - just because

seikou ga tsukameru nara. Kono hitogeki ga

success P - grasp - if - this single blow P

subete de ari, subete no owari da to sureba – kono yo dewa,

all - P is - all P - end - is - P if it does - this world - P

toki no nagare ni ukabu kono yo to iu asase dewa,

time P - flow - P - float - this world - called - shoals - P

raise no koto nado dou demo ii.

afterlife P - thing - etc. - whatever is good

In this excerpt, the accented items are ‘be settled’, ‘quickly’, ‘limit’, ‘grasp’, ‘single blow’, ‘all’, ‘is’, ‘all’, ‘world’, ‘time’, ‘flow’, ‘world’, ‘afterlife’, ‘thing’, ‘et cetera’, ‘whatever’, and ‘good’, which in their sequence more or less describe Macbeth’s desire for a crime without conscience. In the source, the feeling of tension and rapid thought are expressed through metaphor (‘bank and shoal of time’), alliteration (‘surcease, success’), and above all the concentration of monosyllabic words for which the play is well known. These techniques can be imitated to some extent, for example the repetition of sounds in yatte shimatta, sorede yatto and sassato yaru, but whereas the iambic rhythm is distributed evenly throughout the source in the blank verse structure, it is remarkable how Matsuoka’s translation leans emphatically into the final two lines with its concentration of ten possible accents. These accents serve, as it were, to summarize the rhetorical gist of the soliloquy, and in order to alleviate the pace can be contrasted with unaccented phrases such as to iu asase dewa.

     My way of reading the translations is intended as a logical extension of the rules of Japanese prosody that I have been able to test by recording a native Japanese speaker recite one of the translations and subjecting the recording to computer analysis, revealing that significant changes in pitch more or less corresponded to my predictions (Gallimore, 161-7). Yet my accentual approach does not conform to conventional Japanese poetics. Conventional Japanese prosody is syllabic rather than accentual in structure, most famously the five-seven-five of haiku poetry. The successive grouping of five and seven syllable units generate powerful rhythmic sequences that support semantic and metaphorical layers of poetic expression, and in the case of traditional kabuki drama of narrative as well.

     Japanese Shakespeare translation is treated as a genre of modern Japanese drama, which is usually written in modern colloquial Japanese, and for that reason avoids traditional syllabic meter (shichigocho) unless its use is justified in context. As one example, Kuwayama Tomonari of Kyoto University has recently made an experimental translation of A Midsummer Night’s Dream that is entirely in shichigocho, and has been produced by a Tokyo-based company; Kuwayama’s rationale is that the meter is easily recognized by Japanese audiences and conveys the play’s poetic qualities. Syllabic meter is not necessarily archaic but it is associated with certain literary and dramatic traditions that are now marginal to mainstream Japanese culture, so that its use is likely to be more ornamental than regular. The Shakespeare translator Oyama Toshikazu remarked that ‘even in prose, there is a tendency to follow the [seven-five] rhythm pattern, especially when one’s feelings are excited’ (Oyama, 32), and debatable though the extent of its continued usage may be, Oyama’s point does explain why we can find numerous occasional examples of shichigocho in the translations, for example Nakano Yoshio’s translation of the opening chorus of Romeo and Juliet (1948) and Kawai Shoichiro’s translation of the lovers’ sonnet from the same play (2005). These are both cases of heightened emotion demanding a more formal register than ordinary prose: of strong, potentially subversive feelings being contained within a powerful cultural norm such as shichigocho.

     The difficulties of shichigocho translation are obviously those of matching lexical choices with the syllabic structure and of justifying its purpose to a modern readership. Yet it is an example of familiarization, or Japanization, that stands at the opposite extreme from the occasional attempts in modern Japan to conform Japanese poetics to Western, primarily English prosody. In the late 19th century, the writer Yamada Bimyo, having appropriated English grammatical structures in his successful campaign to amalgamate the colloquial and literary styles of Japanese (the genbun icchi movement), faced rather tougher opposition to his serious proposal that Japanese poets adopt regular stressed meters such as Shakespeare’s iambic pentameter, and thus vary the pitch of individual words and phrases as the flow of the line demanded it with complete disregard for the rules of Japanese pitch accent. As Kawamoto Koji explains,

     Bimyo was attempting to establish new metrical rules, whereas (as I have been arguing) pitch accent may already be functioning covertly in translation to shape interpretation, and when combined with other features such as syllabic meter may even have a poetic function. One wonders why Bimyo bothered, and presumably the reason why he did so was because he believed in those early days of language reform that unnatural pitch variations might gradually be tolerated and accepted by native speakers. Moreover, Bimyo’s proposal touches on a tension between the mechanical and meaningful dimensions of prosody that is fundamental to any discussion of Japanese Shakespeare translation. Accentual prosody is ‘mechanical’ in the sense that accentuation is meaningless in itself, and yet is used ‘meaningfully’ both to communicate meaning and enhance the flow of the text. Syllabic meter is also meaningless in itself, but bears considerable cultural capital not to mention generating rhythms that have a rhetorical purpose.

     Accentual prosody is therefore one of a number of means at the translator’s disposal of imitating Shakespeare’s rhetorical technique and its interpretive function of clarifying and enhancing meaning, in other words of making Shakespeare both readable and speakable in Japanese. Japanese translators such as Matsuoka systematically compensate for the lack of equivalence between Shakespeare’s English and their target language (which is only a more extreme case of the lack of equivalence between any two languages); apart from accentual prosody, other conventional devices for doing so are metaphor, repetition, sentence structure, lineation and enjambment, the use of mimetic and affective expressions (e.g. sassato), and idiom. This article has barely touched on the normative context, except to suggest that while a stressed, iambic line is unacceptable (and far less acceptable than one in syllabic meter), a freely rhythmical style combining pitch accent with other poetic devices is likely to comply with cultural norms, or at least be measured against individual preferences in matters of poetic and dramatic style. These preferences exist as a spectrum within a cultural field with its own inner dynamics and boundaries, which is described by Sameh Hanna in his study of Shakespeare translation in Egypt as

The prosody of Japanese Shakespeare translation may be one example of an ongoing – and often productive – debate between those who seek to create capital out of Shakespeare in modern Japan (in the commercial theatre, for example) and those who insist on the autonomous standards of their interpretive community (the reading of prosody by Japanese academics, for example). Neither group can claim absolute understanding of Shakespeare’s intentions, and each group is bound by its interpretive viewpoint.