The earth lay white under the night sky

“The earth lay white under the night sky”.

This is the Seidensticker translation of

夜の底が白くなった。
Yoru no soko ga shiroku natta.
Bottom-of-night [subj] white became.

This is the second sentence of Kawabata’s Yukiguni. What insights about translation can we glean from this example?

My suggested translation is

Whiteness tinged the depths of the night.

I will discuss the rationale for this below. But first, let’s look at Seidensticker’s translation.

Omissions, additions

Let us start off by examining the Seidensticker translation. Immediately we notice that he has introduced four new elements nowhere to be found in the original: earth, laying, under, and sky, and omitted floor. The only things recognizable from the original are “night” and “white”.

I’m trying to imagine the gears whirring in Seidensticker’s head as he tried to make this sentence fit together. “Let’s see, we’ve got a 底/soko, which is the bottom of something, and we’re talking about a landscape so hey it must be the ground, let’s call it the ‘earth’, which sounds good, but then I have to do something with the 夜/yoru, and you can’t say ‘night of the earth’ but since there obviously must be a sky up there I’ll call it a ‘night sky"‘ and the earth can go ‘under’ it. White is just white, so that’s OK, and I know it’s talking about the snow (maybe I should put that right in the translation to make it real clear?), and since snow is always on top of the earth, I can say it’s ‘laying’ on the earth, which has a kind of nice ring to it, we all like the word ‘lay’. I can’t figure out the nuance of natta, so I’ll just skip that.”

Hey, we’ve all been there.

“Earth” and “laying” and “sky” sound very literary and lyrical, eminently Nobel-worthy, except that Kawabata never wrote them. Now, opinions vary widely on the degree to which translators can or should take liberties in the interest of adapting a work to the exigencies of foreign languages and foreign readers. But is there a viable theory of translation, no matter what the type of content, where the translators is allowed to, encouraged to, or even applauded for, just making stuff up out of thin air?

Believe me, I understand the temptation and the motivation. We want the English text to come out sounding “good”. It needs to work as English. If we need to introduce some new element to make that happen, is that such a problem? If we need to remove something from the original that just doesn’t “fit in”, is that such a problem?

Actually, I think it is a problem, with very few exceptions. I believe that as a general rule the elements in the original can in fact be translated and recombined in ways that create beautiful, readable target language text if only one makes the effort and takes the time to do so. Let me explain why, starting with the last part of the sentence about turning white.

白くなった, tinged with whiteness

In my suggested translation I have rendered 白くなった as “whiteness tinged” and moved it to the beginning of the sentence. What’s up with that?

In terms of the “white”, could the 白い (shiroi, white) in the original Japanese mean something other than simply “white”? After all, Japanese uses this word to refer to old people’s gray hair. Could it actually mean “whitish”, or “having a white tinge”? Does the construction 白くなった necessarily mean that something turned completely white, or could it be becoming somewhat or partially white? For instance, if we say grandpa’s hair 白くなった, it does not necessarily mean every single hair on his head turned white; it could be that his hair acquired a whitish tinge or was whiter than before.

Let’s also be very clear: there’s absolutely nothing about the original that indicates that the “whiteness” involved was necessarily that of snow laying on the ground. It could have been the whiteness of snow still falling! It could have been the reflection of the snow lighting up the entire sky! Or it could have been a combination of all of those things! All these possibilities, which Kawabata was depending on the reader to imagine and choose from, have been ruthlessly eliminated by Seidensticker’s insistence that it could only have been the white snow laying on the earth.

And if you agree that 夜の底 should be “depths of the night”, as I propose below, the depths of the night can hardly be or turn entirely white.

In terms of the position of whiteness in the sentence, I would make the case that whiteness is the central aspect of this sentence. It is mainly about whiteness, not the depths of the night (or earth). This view is supported by the use of the particle が in the sentence. From the standpoint of English syntax, that makes the whiteness a good candidate to lead off the sentence.

As translators, we well know (but all too often forget) the power and relevance of not slavishly following Japanese word order, or part-of-speech categories, like 白い being an “adjective” in Japanese so it must be an adjective in our English translation.

Finally, let us delve into なった/natta/became a bit more deeply. Once he was committed to the “earth laying under the night sky”, Seidensticker had dug himself into a hole when it came to dealing with this なった. He literally had no place left to stick “became” or “had become”. He couldn’t say "the earth laying under the night sky had become white”. He had to back off to a simple past tense, “lay white”. But if Kawabata’s intention had really been that it simply “was” white or “lay” white, then the Japanese would have been 白くなっていた (natte-ita, “had become”), or 白かった/shirokatta (“was white”). No, Kawabata was definitely trying to convey something specific with this natta. He was trying to convey a feeling of change compared to the darkness inside the tunnel or prior to entering the tunnel. now that we were in the snow country. This should be conveyed with a verb that implies changing or becoming—like “tinge”, for example. I do not think this violates any rule of substituting my judgment for the author’s or introducing new elements. I would argue that “whiteness tinged” is almost a direct translation ion of 白くなった/shiroku natta.

夜の底, the depths of the night

Seidensticker is reported to have called 夜の底/yoru no soko (lit. “bottom of the night”) a “striking figure of speech”. Let’s give him the benefit of the doubt and assume that he knew that this phrase existed in Japanese as opposed to having been invented by Kawabata. What it actually means is, as various references explain, the darkness of night, the depths of night, or the dark of the night.

This expression is also seen in Akutagawa’s Rashomon, in the sentence 急な梯子を夜の底へかけ下りた (I scampered down the steep ladder into the recesses of the night). Another sample usage is 夜の底に姿をける (disappear into the depths of the night).

But in his translation Seidensticker, as he himself openly admits, wantonly omits this phrase, retaining only the “night” part. His stated rationale was that he did not want to put “white” and “night” next to each other lest the inadvertent rhyme distract the reader. But there were plenty of ways to avoid that without discarding the phrase entirely.

With the benefit of hindsight, Seidensticker later said he wished he had translated it as “the floor of the night lay white”. He thus agreed that bringing in “earth” and “sky” was gratuitous. But why, after going to all the trouble of rethinking this translation, would he continue to insist that what he himself called the “striking phrase” in the original about the depths of the night just be discarded, and go back to the mindless translation “floor”?

Let us not forget that Kawabata was a founder of the literary school known as the 新感覚派 (Shinkankaku-ha, new perception/impression/sensation), which, according to Riichi Yokomitsu, another of its founders, viewed perception as “direct, intuitive sensation of a subjectivity that peels away the naturalized exterior aspects and leaps into the thing itself." Hmm. The translation of “floor” could not be more distant from this this vision.

Bottom line: there is no compelling reason not to translate 夜の底 (yoru no soko) as what it actually means, which is “the depths of the night”. Perhaps this would be clearer if we looked at 底 (soko) more closely.

The depths of understanding

Like so many Japanese words, the character 底/soko has an extremely broad range of meaning, extending far beyond “floor”.

It does in fact mean the bottom of some container, like a cup. It can also refer to the bottom of a body of water like a river or lake (海底), although interestingly it it can mean either the actual bottom, or the lowest layer of water (the “depths”), or the lowest layer of anything for that matter.

“Depths” is actually one of its basic meanings. It does not have to be something physically in a downward direction. We can say 心の底, kokoro no soko, “the depths of my soul”, or 腹の底, kara no soko, “your gut” (or “bowels”), which is where cries of anguish might come from, or belly laughs (腹の底から笑う, “laugh from the bottom of your belly”), or depths of your being. It could be used in an expression like 戦略の底 (shin-ryaku no soko), meaning the heart of your strategy. You can say 底が知れぬ (soko ga shirenu, “unfathomable depths”), or 底が浅い (soko ga asai, lacking depth). (It can also mean foundation, or base (including the mathematical use of the word), or core, or essence.)

Of course, it’s also the sole of your foot. :-)

So no, Ed, it’s not just “floor”.

Rhythm and cadence

Let us not neglect to mention a prominent aspect of Kawabata’s original Japanese: its choppy cadence of short sentences, especially this sentence and the next (信号所に汽車が止まった, shingsho ni kisha ga tomatta, the train pulled up at a signal stop). The Japanese has four main “beats”: YORU no SOKO ga SHIROKU NATTA. The distinctive rhythm is memorable. Seidensticker’s translation completely misses this aspect of the original. That is why I prefer something like “WHITENESS TINGED the DEPTHS of the NIGHT”, with the same four beats.

Perhaps we’re overthinking this, but is it possible that the prevalence of the “O” sound in the original is meaningful, or something we should worry about when translating a novel like this? We have yOru no sOkO ga shirOku natta. We already have “i” sounds in “night” and “white”; is there some way to express the soko or natta part using “i” as well? There are no obvious candidates, so we will throw in the towel here.

More on 底

Although the origins of the character 底 will not really help us understand its broad semantics or use in 夜の底, kanji aficionados might be interested to note that this character is a so-called 会意兼形声文字, kaii ken keisei moji, meaning a semantic/phonetic compound. There’s the roof of a house, and under that a bowing or stooping person (the 氐) standing on or perhaps picking something up off the ground, this component also providing a phonetic value, while being semantically associated with the the meaning “low”, as in the common character 低い/hikui.

When studying 底, don’t forget all the Chinese compounds that use 底, like our old friends 根底 (kontei, foundation/cause, “deep roots”), 奥艇 (okutei, remote depths), 胸底 (kyōtei, bosom), 基底 (kitei, basis), or 徹底 (tettei, thoroughly, literally “penetrating to the bottom”).