Repository path: workshop/translations/futabatei-honyaku-hyojun/R04-v1/translation.md · rendered 2026-09-09

Futabatei Shimei, My Standard for Translation (1906) — complete

Contamination: suspected. Futabatei is a canonical Meiji author and this essay is standard reading in Japanese translation studies, so English discussion of it — and quite possibly a full English rendering — plausibly sits in the lead's training data. Unlike the Schleiermacher and Yan Fu cases, the lead has no specific remembered English wording for any sentence here, and no phrase in this essay has the stock-formula status of 信達雅 or of "moves the reader toward him". The declaration is suspected rather than none because absence of a remembered rendering is not evidence of independence.

My Standard for Translation

How ought a translation to be made? The standard will differ from person to person, so of course nothing can be laid down for all cases. I shall therefore describe the method I have followed up to now.

European writing, if you simply read it, is nothing in particular; but taste it properly and it has a cadence of its own, and read aloud its rise and fall comes out well ordered. It is, in a word, musical. So it is quite interesting merely to sit and listen to someone else reading it. The sense of a passage is in fact taken in better by reading silently — and yet places that my own shaky knowledge cannot fully make out still come across as interesting when read aloud. This is certainly one distinguishing quality of European writing.

Japanese writing, however, has no such movement. On the whole it dribbles along. There is no harm in reading it silently, but read aloud it is exceedingly monotonous. Not only is its rise and fall indistinct — no way of reading it aloud was ever established in the first place, so it is unsuited to being read aloud at all.

Still, if one is going to translate a foreign text at all, one must carry over its movement as well. That was the first of the standards of form that I set myself in translating.

So when I went into how the commas and the periods fall, what caught my eye at once was the piling up of phrases that say the same thing. To take one instance: the sort of thing one meets so often in Macaulay, in spite of and the like. In point of meaning, two or three or four would be enough, and yet for the sake of the cadence one more gets added. But the meaning has already been said out in full, and one cannot very well write in something that means something different, so there is no help for it: a redundant and superfluous thing gets said.

The same happens at the level of the word. The Japanese yatara-mushō — "every-which-way, no-reason", where both halves say the one thing — is an instance; and so are expressions like "he rebuked him strongly and severely", or "talked him round gently, so as to put no corner on it", which are of the same kind as what one meets constantly in European writing. In all of these, over and above making the sense of the sentence clear, an adverb goes in for the sake of the cadence because one wants it in, and where two adjectives are quite enough, a third is added to make three. Nor are the commas placed on grounds of sense alone: they are very often placed for the sake of the movement.

Hence, in translating a foreign text, if you attend to the meaning only and put your weight there, you are in danger of breaking the original. You must — so I believed — swallow the original's cadence whole, and carry that over. So I did not throw away a single comma or period at will: if the original had three commas and one period, then the translation too was to have one period and three commas, and in that way I tried to carry the original's movement over. In my early days at it especially, I made the number of words the same as the original's as well, and did not let the shape be broken, taking it for my sole object to carry over the original's cadence — and I laboured terribly over the form. But in practice it would not go as I wanted it to; there were some things I simply could not bring into line with my standard. I did once resign myself to the thought that I had not the arm to translate by my own standard — but that was never what I really wanted, and long afterwards I still held to this policy as far as form was concerned.

And what came of it? Look at my translations and — dear me, they are hard going: crabbed and hard on the tongue, jerky, and however you take them the workmanship is poor. The world's opinion of them was poor to match. There were people now and then who praised them, but on the whole the voices were against. And yet not one of them took in the feeling I had swallowed — that I had laboured and botched it — and pointed out here is where it fails; nor did any of them see here is how far it succeeded. So praise gave me no pleasure, since it had nothing to do with my standard, and abuse was so wide of the mark that there was nothing to be learned from it. I was, as it were, wrestling a bout by myself: quite detached, in fact, from praise and blame alike, with my eye on a single point and taking pains over that.

The reason was that my reverence for literature was strong. When Turgenev, for example, wrote his works, the state of mind he was in was an intensely holy thing; therefore, I believed, translating him must be holy in the same way, and not a character, not a phrase, must be treated lightly.

But the form of a piece of writing differs, of itself, according to the writer's 詩想 — his poetic conception. Turgenev has Turgenev's style; Tolstoy has Tolstoy's. So it is with everyone who has made a house of his own. This is the same in Japan and in China: style stands in an inseparable relation to the man's poetic conception, and the movement of the writing differs from one man to the next. It follows that in translating, one cannot take some one style and fit it to everybody. Turgenev is Turgenev, Gorky is Gorky; you must take in each man's poetic conception separately, and — to put it strictly — walking, standing, sitting or lying down, you must hold body and mind exactly as the original author's, and carry his poetic conception over faithfully. This is truly the fundamental and necessary condition of translation.

Take Turgenev now for an instance of it. His poetic conception has not the aspect of autumn or of winter; it has the aspect of spring — and not early spring either, nor mid-spring, but the aspect of late spring, just at the point where the cherry has blazed into full riot and is beginning, a little, to scatter; there is something in it of walking a long narrow path with cherries planted down both sides, on an evening when a beautiful spring moon, blurred and blurring, is shining in a hazy middle sky far off. To put it shortly: within the voluptuous beauty, somewhere, a lonely place — that is Turgenev's poetic conception. And since, as follows of itself, that air runs right through his novels, in translating him you must not let go of that feeling; you must write always as that man, or it will keep slipping out of true with the movement of the writing. At this point it will not do to fuss vainly over commas and periods and the other matters of shape. First swallow the poetic conception, which is fundamental; and only then translate, without breaking the poetic form.

In fact, when I translated Turgenev I meant to keep his poetic conception always before me and to assimilate myself, truly, to it — and somehow I did not manage it. Say I did not manage it; the standard was there all the same. Only, turning it over in my mind in various ways, I came to think that although one may not flatly declare it impossible to translate by the standard I had set up, it is at the least a difficult way of going about it for me. Why? Because, first of all, I cannot write Japanese well. It is nearly a case of Russian prose being clearer to me than Japanese prose: that is, I have the power to taste the original, but the power to reproduce it does not go along with it.

So I looked about for some other method of translating, and Zhukovsky's manner struck me as interesting. Zhukovsky is a Russian poet, but he has made his name rather as a translator. He translated a great deal of Byron, and did it oddly well. To be sure, the Russia of that day was in a social condition that bred little Byrons in quantity, and Zhukovsky in particular rang true among the iron — so it may be that he came to coincide with Byron's poetic conception without seeking to, and that this is why he succeeded so greatly. At all events his translations are splendid Russian.

But set them beside Byron's originals and the manner of saying is very different. He makes an oblique opening into a level one and a level into an oblique; where the original rhymes he leaves it unrhymed; or he puts in adjectives and adverbs that are not in the original, cutting and shaping as he pleases. That is, for the most part he demolishes the original entirely, makes a verse form to suit himself, and translates the meaning only.

And what happens when you read the two side by side? My English has always been a little thick on my tongue, so I cannot talk large about it — but at all events the translation gives the savour and the poetic conception better than the original poem does. What ten readings of the original will not yield, the translation yields at one, with all its various beauties; and when you consider the impression of it, it is Byronic through and through. Which is to say, in short: rather than tasting Byron in unsteady English, one gets far more for far less labour by reading Zhukovsky's version.

So I thought to myself: translation will not succeed unless it is done this way. In my own way of going about it, the upshot of clinging to the shape is that the strength of the pen is bound by the shape, so that the thing is hard to read and cramped. Better to do as Zhukovsky did — set the shape entirely aside and bring out the poetic conception the original holds. So I thought; and yet I am a coward, and could not, for all that, bring myself to do it. Why not? Because to go at it in Zhukovsky's manner you must have sufficient strength in the pen — you must be able, even having demolished the original poem, to give its poetic conception a new poetic form — and it seemed to me that this strength was not in me. Judging by the method of translating I had followed all along, it may not succeed, but the shape is held fast to the original, so one does not go disastrously wrong. Whereas if you go at it Zhukovsky's way, then if it succeeds the thing is one blaze of splendour, but if once it fails there is nothing so wretched in the world — so it cannot be borne without a good deal of faith in one's own arm. I was, after all, not so bold as that; it looked to me altogether too risky, and I could not carry it through. And so I went on with the old method of translating, and ……

But all that belongs to the time when I went at translation with a serious head on my shoulders. Of late — no, it does not bear talking about.

(Seikō, January 1906.)

Translator's log

Written at translation time, 2026-07-25, and frozen before any evaluation was designed and before any source page or synthesis was written. Decisions and alternatives, not quality claims.

The four terms the essay is built on

1. 詩想 → "poetic conception", kept in characters on first appearance. This is the essay's load-bearing term and English has nothing that fits it. It is not style (Futabatei distinguishes it explicitly from 文体), not voice (it is a property of the writer, not of an effect on the reader), not vision (too visual), not sensibility (too passive), not inspiration (too momentary — his is a standing disposition from which the style follows). "Conception" was chosen for the sense of something conceived and carried; "poetic" imports a genre association the Japanese 詩 does carry, and I accepted that because Futabatei applies the word to novels without apology. The whole of the essay's second half stands or falls on this word, so the risk is concentrated in one place. Live alternative I nearly took: leaving it shisō untranslated throughout, which the project's strategy set would call copy-opaque, and which I rejected because it would have made the argument unreadable for exactly the readers this artifact is for.
2. 音調 → "cadence"; 文調 → "movement (of the writing)". These two are near-synonyms in Futabatei's usage and I split them across two English words on purpose, because he keeps them apart: 音調 belongs to sound and to reading aloud, 文調 to the writing as a whole and to the individual writer. English has one obvious word, "rhythm", which I refused because it would have collapsed a distinction the essay uses. The cost: "movement" is vaguer than 文調 and has to be propped by "of the writing" on each of its first appearances.
3. 筆力 → "strength in the pen" / "the strength of the pen". Literally brush-strength. "Ability" and "skill" were live and are what an English writer would say; both were rejected because they abstract away from the hand, and Futabatei's argument is precisely about a physical capacity he can feel he lacks. Kept literal, at the cost of a slightly quaint English. Related: 手腕 (ude, "arm") appears twice and I kept "arm" both times — "faith in one's own arm" — for the same reason and with the same cost.
4. 詩形 → "poetic form"; 文体 → "style". Unproblematic, recorded for completeness of the term set.

Decisions in the prose

5. The essay is spoken Japanese on the page — 言文一致 prose by the man who largely invented it — with だ/である sentence-endings, いや and さて and で as discourse openers, and interjections mid-argument (いや実に読みづらい; 夫れ never; どうも). I pitched the English as spoken-essay English rather than academic English: "dear me, they are hard going", "no, it does not bear talking about", "and what came of it?". The alternative was a neutral scholarly register, which would have been easier to write and would have hidden the fact that the founding statement of modern Japanese colloquial prose is a talk.
6. …… at the end of the penultimate paragraph is Futabatei's, not mine, and I kept it. The essay's argument breaks off unresolved and then adds one dismissive coda; the trailing dots are the seam.
7. 独り角力を取っていた → "wrestling a bout by myself". Hitori-zumō is a live idiom (a fight nobody else is in) and also a real thing — a Shinto rite of wrestling an invisible spirit. English "shadow-boxing" was live and would have been more idiomatic; rejected because it means practising, whereas hitori-zumō means exerting yourself in earnest with no opponent, which is Futabatei's point about a decade of unnoticed labour. "Wrestling a bout by myself" is a calque with the sport kept.
8. 佶倔聱牙 (kikkutsu-gōga, a four-character compound from Han Yu, "crooked and grating on the teeth") → "crabbed and hard on the tongue". The source has a gaiji note here (※［＃「敖／耳」…］) because the second character is outside the standard set; that is a fact about Aozora's encoding, not about the text, and I did not represent it.
9. やたらむしょう → "yatara-mushō — 'every-which-way, no-reason', where both halves say the one thing". This is the only place I glossed inside the running text rather than in the log. Futabatei offers it as a specimen of Japanese pleonasm, so a reader who cannot see that it is pleonastic loses the example entirely; substituting an English pleonasm ("null and void", "each and every") would have silently replaced his evidence with mine. This is scaffold in the project's strategy list, chosen over substitute.
10. 「強く厳しく彼を責めた」 and 「優しく角立たぬように説得した」 → "he rebuked him strongly and severely" and "talked him round gently, so as to put no corner on it". The second is itself a Japanese idiom (角を立てる, to raise a corner = to cause friction) offered as an example of padding, so I calqued the corner rather than smoothing it to "without giving offence". Same reasoning as 9: these are exhibits.
11. 仄起 / 平起 → "an oblique opening" / "a level one". These are Chinese regulated-verse tone-category terms, and Futabatei is applying classical Chinese prosody to English and Russian metre. I translated them literally rather than converting to English metrical vocabulary ("iambic"/"trochaic"), which is what a modern translator would probably do. Converting would make the sentence intelligible and would delete the most interesting thing in it — that a Meiji translator reaches for Chinese prosody to describe Byron. This is the same decision-shape as refusing to supply "natural selection" for 天擇 in T-yanfu-yili-yan-R04-v1, made independently in the same session, and I note the coincidence rather than defending it.
12. 鉄中錚々 ("clanging among the iron", from the Hou Han shu — the best of a mediocre lot) → "rang true among the iron". Calqued, and it is thin in English; the alternative was "stood out among them", which loses the metal.
13. Turgenev's cherry-blossom passage is the one stretch of the essay that is not argument but demonstration, and it is a single long Japanese sentence with 〜も〜も〜も accumulation. I kept it as one long English sentence broken by dashes and semicolons. おぼろおぼろとした — reduplicated 朧 — became "blurred and blurring", an English reduplication that is not idiomatic; the alternative "softly hazed" was smoother and lost the doubling. This is the same mimetic-reduplication problem A-shaw-spider-thread records for するすると and くるくる, and I made the opposite choice from Shaw's: he converted mimetics to manner-verbs, I kept a doubling English does not want.
14. 行住座臥 ("walking, standing, sitting, lying") is a Buddhist four-character formula for the whole of waking life. Kept as the four postures rather than rendered "at every moment", because the four postures are what makes the demand physical, which is what the sentence needs.
15. Honorific and self-referential register: Futabatei's 自分 throughout is a plain, slightly gruff first person, not 私. English has one "I". Third instance in this session of a source-side pronominal register that English cannot carry (the others: Yan Fu's 不佞, Schleiermacher's man/wir shifts). Recorded, not solved.

What I could not do

16. The essay's central claim is about a property of Japanese prose that my English cannot exhibit. Futabatei says Japanese writing has no cadence and is monotonous read aloud, and that European writing does have one. Whatever the truth of that, the English translation of the claim is written in the very prose he says possesses the quality — so the reader of my version encounters the assertion in a medium that silently contradicts its second half and gives no access to its first. There is nothing to be done about this; it is worth recording as a case where translation cannot preserve the reader's ability to evaluate the claim being translated. internal-judgment-only.
17. The 言文一致 achievement is invisible in English. The single most remarkable thing about this text as an object — that it is written in ordinary spoken Japanese at a date when essays were not — has no English correlate at all, because English essays had been written in something near the spoken language for two centuries. Every device I used to signal "this is talk" reads in English as informality, i.e. as a choice of register within a range, whereas in 1906 Japanese it was the making of the range. The project's naturalness sense assumes a target register can be picked off a shelf; here the source's register was being built as it was used, and the translation cannot show that.