The seeds for this translation of one of the classic works in Japanese literature were sown over forty years ago. One of the most compelling accounts of poetry I heard as an undergraduate student was a reading by Christopher Logue from what was then his recently published translation of Book 16 of Homer’s Iliad. Logue’s version sounded with an enormous, visceral power. It leapt, twisted, invented, clashed, modernised, contorted, visualised – in ways that made all other translations of the poem seem staid and bland. Since then, it has become clear that this dramatic and verbal energy was no isolated tour de force by a young poet. Forty four years later, Logue’s radical approach to translating the poem (All Day Permanent Red, 2003, a rendering of the first battle scenes in Books 5-8) retains all of its earlier power to wrench and dislocate the original into a contemporary poetic idiom. There is, however, one aspect of the translation that, then as now, is likely at the very least to bemuse, if not actually to shock. Logue cannot read ancient Greek, not a word of it. He has created his Iliad by consulting already existing renderings, developing a sense of what the original is saying, and then taking off to create his own version. In the words of one reviewer, on the surface at least ‘it’s like learning of a deaf man who prepared himself to conduct Stravinsky by watching Fantasia’.

However quixotic or foolish Logue’s task may seem, though, the unquestionable power and richness of the result raises a fundamental question: to what extent, if any, is it possible to translate from a language of which one has little or no knowledge? Is it simply impossible? Or will such a version have to rely upon so many extraneous aids (numerous other translations, massive resort to commentaries and dictionaries, constant oversight by native speakers of the original language, and so forth) as to drown any individual voice in what will be essentially the translation of a collective? If these supports are not available, will such a text inevitably have to be loose paraphrase or imitation or re-composition because the complex connotations of the original cannot be understood? Or may there be some means by which all these barriers can be surmounted, and the original text presented in a close, faithful and resonant way?

This avowedly experimental translation of Matsuo Bashō’s Oku no Hosomichi raises all of the questions mentioned above; and while it may not answer all of them, it attempts at least to scrutinise, test and explore them. The personal journey may be worth describing briefly. At the very beginning of drafting the translation, I knew not a word of Japanese. I had for years been interested in haiku – that infinitely concentrated moment of perception condensed into 17 syllables of verse – and also in travel writing. And Bashō’s name had long been known, as one of the greatest exponents of both haiku and travelogue. But of the language in which he had written, I knew nothing.

Such ignorance might seem problematic enough in a translation from a European language with a similar script and basic structure. But from a language with a demonstrably different script and structure, the ignorance might seem insurmountable. Even a cursory reading in a Japanese grammar is enough to highlight quite radical differences between Japanese and English. Nouns in Japanese, to take a single example, have no gender, or case, or distinction between singular and plural. The Japanese for ‘dog’ or ‘a dog’ or ‘two dogs’ or ‘many dogs’ is the same word (‘inu’, in Romanised Japanese). Verbs, similarly, remain the same whether the person is first, second or third, singular or plural, masculine, feminine or neuter. There are no terms corresponding to the definite and indefinite articles: the tree and a tree are the same word, ‘ki’. Together with three different writing scripts (Kanji, Hiragana and Katakana), which can be written either vertically or horizontally, and a different sequencing of subject, object and verb, these features might seem to make for a total impenetrability – a language rooted in paradox and ambiguity, and understandable only after years of immersion.

Given this context, there are three major ways in which the challenge of translating Oku no Hosomichi has been taken up, and each is worth developing in a little detail:

Other translations         

There are currently no fewer than eight different translations into English of the whole of Oku no Hosomichi, together with several versions of parts of it. Placing these versions alongside each other at every step of the way allows two contrasting features to emerge: the lowest common denominators (whether part of speech or syntactic ordering) that all the versions share, but also the differences in tone and register between them. Consider, for instance, the celebrated opening ‘sentence’ to the travelogue:

The passing days and months are eternal travellers in time. The years that come and go are travellers too (Britton).

Moon & sun are passing figures of countless generations, and years coming or going wanderers too (Corman).

The moon and sun are eternal travellers. Even the years wander on (Hamill).

The months and days are the travellers of eternity. The years that come and go are also voyagers (Keene).

The sun and the moon are eternal voyagers; the years that come and go are travelers too (McCullough).

The months and days are the wayfarers of the centuries and as yet another year comes round, it, too, turns traveller (Miner).

The months and days are wayfarers of a hundred generations, and the years that come and go are also travelers (Sato).

Days and months are travellers of eternity. So are the years that pass by (Yuasa).
 

The shared denominators here can be easily identified:

• moon/sun (months/days)

• travellers/passing figures/wayfarers/wanderers

• eternal (eternity)/countless generations/of the centuries/of a hundred generations

• year/years/another year

• come and go/wander on/pass by

• too/also/so/even
   

Yet these common features are orchestrated very differently. The choice between moon or month, and sun or day, is answered by five translators in one way, and three in another. The definite article is in one case applied to both nouns; in five other cases, to only one; in two cases, to neither. Levels of diction vary: the generic ‘traveller’ is occasionally repeated, or juxtaposed against the more antique resonances of ‘wayfarer’ and ‘voyager’. In terms of rhythm, too, there are manifest differences: from the curt, rather banal ‘even the years wander on’, through the slightly convoluted, over-explicatory ‘and as yet another year comes round, it, too, turns traveller’, to the persuasive iambic stresses of ‘the years that come and go are travellers too’.

Such analysis of both the common and the individual features in each of the translations soon makes apparent the strengths and weaknesses of each rendering. Tired diction here, inappropriate register there; natural, unforced cadence here, resonant phrasing there. And as word is compared with word, phrase with phrase, an almost intuitive sense develops, not only of what Bashō’s original says, but of how it can best be translated into English. The mental notes made can be illustrated by reference to Bashō’s title, Oku no Hosomichi:

 

phrase only occurs at one point in the narrative (‘kano ezu ni makasete tadoriyukaba, oku no hosomichi no yamagiwa ni tofu no suge ari’)

oku = general name for the northern provinces; can also mean ‘interior(s)’ or ‘inner recess(es)’
no = links two nouns: at, in, of, on
hosomichi = thin/narrow + road/path/way


Narrow Road to a Far Province (Britton): no article before ‘Narrow Road’ – evocative, or dulling? ‘far’ is good. ‘Province’ – accurate but lacking resonance?

Back Roads to Far Towns (Corman): too overtly urban and modern. Why highlight ‘towns’? ‘Back roads’ suggests a detour from existing ‘main roads’, which is surely not what Bashō meant. Monosyllabic rhythm?

Narrow Road to the Interior (Hamill and Sato): no article again before ‘Narrow Road’ – I’m torn between finding this productively suggestive and rather bland. ‘Interior’ is good, intimating both geographical and psychological conditions.

The Narrow Road to Oku (Keene): ‘to’ is better than ‘of’ in evoking sense of travel towards. But leaving ‘Oku’ untranslated will surely produce blankness, rather than telling ambiguity, to an English-speaking reader. What, who, where is ‘Oku’?

The Narrow Road of the Interior (McCullough): seems slightly prescriptive in resonance. ‘Of’ implies that the ‘Interior’ has already been reached, rather than travelled towards.

The Narrow Road to the Deep North (Yuasa): evocative, suggesting both the difficulty and the penetration of the journey. Perhaps ‘Deep North’ is a little free, but it well conveys the sense of a far-off land, reached only after difficulty.



Would it be worthwhile taking Yuasa’s hint and risking a very free adaptation of ‘Oku’ – something that evokes an emotional landscape, as well as a geographical one? Would ‘The Narrow Road to a Far-off Land’ do? On second thoughts, ‘Far-off Land’ could bring to mind a misleading fairy-tale dimension (‘Somewhere over the rainbow…’). Probably better, after all, to keep to ‘the Deep North’.
 

This kind of thought process, brought to bear on each word, phrase and section as the travelogue unfolds, results in a continual flow of judgment, both conscious and instinctive, about the most effective word, syntactic pattern, rhythm, and tone of voice.


Word-for-word translations

In addition to the support provided by the eight translations above, there is a further resource: Makoto Ueda’s word-for-word versions of many of the haiku that punctuate Bashō’s travelogue. A single example will show how valuable even a literal translation of each word can be:
 

shizukasa ׀ ya ׀ iwa ׀ ni ׀ shimiiru ׀ semi ׀ no ׀ koe
 
stillness ׀ ! ׀rock ׀ to ׀permeate ׀cicada׀ ’s ׀ voice

Not only does this literal version indicate the order of the words and images in the original, but it also gives important signals about their relationships. The word ‘ya’, for instance, performs the function of a kireji, or ‘cutting word’. In Nobuyuki Yuasa’s words, ‘when a kireji is used in the middle of a poem it cuts the stream of thought for a brief moment, thereby indicating that the poem consists of two thoughts half independent of each other.’ ‘Ya’ also expresses a sense of wonder or excitement, and the closest English equivalent would probably be an exclamation mark. The opening five syllables, then, must evoke the wonder and profundity of the stillness, and conclude with the slightest of pauses before the poem resumes. When it does, another important signal is given: not simply the outer contrast between ‘rock’ and ‘voice’, but also the fact that ‘rock’ and ‘voice’ is the ordering, not ‘voice’ and ‘rock’. In other words, the ‘permeation’ of the rock is a preceding process before the suspended climax of perception, the cicada’s voice, is heard.

These features can emerge only from a word-for-word rendering of the original. But once recognised, they become part of the larger thought processes described earlier. How best to convey the sense of total stillness in five syllables, and then a momentary hiatus? How to anchor the rather generalised terms ‘permeate’ and ‘voice’ in a sensory immediacy? Is there one cicada, or are there many – and which is the more effective? The answers provided by the eight major translations conclude with Ueda’s version and then my own:
 

In this hush profound

Into the very rocks it seeps –

The cicada sound.

 (Britton)

quiet
into rock absorbing

cicada sounds

(Corman)

Lonely stillness -

a single cicada's cry

sinking into stone

(Hamill)

How still it is here -

Stinging into the stones,

The locust's trill

(Keene)

Ah, tranquillity!

Penetrating the very rock,

a cicada’s voice.

(McCullough)

In seclusion, silence.

Shrilling into the mountain boulder,

The cicada’s rasp.

(Miner)

Quietness: seeping into the rocks, the cicada’s voice

(Sato)

In the utter silence
Of a temple,
A cicada’s voice alone
Penetrates the rocks.
 

(Yuasa)

the stillness –
seeping into the rocks
cicadas’ screech

(Ueda)

the utter silence …

cutting through the very stone

a cicada’s rasp

(Chilcott)

The strengths and limitations of all of these versions will be quickly discernible. But it is worth noting that, unlike many translators, I have adhered to the basic 5-7-5 syllabic count of the original for all the haiku in this translation, and have chosen to avoid capitalisation and most punctuation marks. Such typographical signals can often seem intrusive, directing response rather than allowing the suggestiveness and ambiguity of the original free rein. Beginning a haiku with a capital letter and ending it with a full stop suggests the perception is contained solely within the words. But in truth, Bashō’s haiku begin before the first syllable is uttered, just as they sound long after the seventeenth syllable has been heard.


Native speakers and resources

Whatever support other translations can give, however, there are inevitably moments when some crux arises that can be resolved only by appeal to a native Japanese speaker. Sometimes, the crux has to do with connotation and resonance (‘is a closer in meaning to b or to c? or is it to both b and c with a touch of d?’). Sometimes, it has to do with cultural circumstances or positions that are very different from those in the western world (for instance, are the many holy men that Bashō meets in his journey best described as ‘monks’, ‘priests’, ‘high priests’, ‘abbots’, ‘bishops’, ‘archbishops’, or indeed none of the above?). But when such questions have arisen in this translation, I have been able to avail myself of the native and bi-lingual knowledge of Dr Mark Jewel, of the University of Waseda, and of Masami Sato, of Hanazono University. To both, I offer my sincerest thanks for the generosity of their help and advice. I am grateful, also, to Kendon Stubbs, co-director of the Japanese Text Initiative at the University of Virginia; the text of Oku no hosomichi presented here is used by permission of the JTI, Electronic Text Center (http://etext.lib.virginia.edu/japanese). My thanks are due, too, to Peter Goodman and Stone Bridge Press for their kind permission to reproduce the map of Bashō’s journey contained in their Bashō’s Narrow Road: Spring & Autumn Passages, trans. Hiroaki Sato, 1996.


Conclusion

Whether this translation of Bashō’s Oku no Hosimichi has proved or disproved the possibility of translating from a language of which one has little or no knowledge, is for individual readers to determine. The best judges, presumably, will be those readers who are totally bi-lingual, as sensitive to every register and nuance of Japanese as they are of English. For myself, I began drafting the translation entirely sceptical, believing it would prove impossible. And yet it has emerged and is here. At the beginning of this introduction, I quoted the words of one reviewer about Christopher Logue’s version of the Iliad: ‘it’s like learning of a deaf man who prepared himself to conduct Stravinsky by watching Fantasia’. My final position, I hope, may be of a partially hearing man who prepared himself to conduct Stravinsky by discovering, at least, how to read a musical score.