Monthly Archives: October 2016

Japanese literature review: Kenji Miyazawa’s “Matasaburo of the Wind”

I’ll have to just come out and admit it: of all of the books I have read there is a small group I never made it to the end of, including both Japanese and English works. One of these is Kenji Miyazawa’s (宮沢賢治) “Night on the Galactic Railroad” (銀河鉄道の夜) which was a deeply moving story when I watched the anime adaptation, but several years (~7-8) ago when I tried out the novel a combination of older Japanese (it was published in 1924) along with a somewhat uncommon dialect (秋田弁) made it very hard going. I think I made it 20 or 30 pages, but eventually decided to give up.

Some time back I had filled my Kindle with a bunch of short stories and novels from Aozora Bunko (free classic Japanese works), and on a recent trip when I had a few spare hours I decided to see what stories I had and stumbled across Miyazawa’s short story “Matasaburo of the Wind” (風の又三郎), written in 1933, which is about a mysterious student who comes to a small school in the countryside. Like “Night on the Galactic Railroad”, a combination of things made it somewhat challenging, but as it was much shorter I tried to read it through the end. After a while I finally finished it, and was glad I didn’t give up this time.

Fortunately, I found a great site that translated the lines with heavy dialect into standard Japanese, which allowed me to follow what was going on. The grammar itself isn’t that difficult (although the sentences can be quite long), but you’ll find some words you thought you knew like “まるで” used in ways you may not be have expected. Sometimes it’s easy to figure what the author is trying to say, but sometimes it takes a bit of googling.

The story itself is pretty straightforward, and on the surface doesn’t amount to much more than school children experiencing various things together as they learn and play (with a few exceptions), which I guess you could call a “coming of age” story. But I don’t yet fully understand what makes this story a classic, if in fact it is considered as such (at least it was valued highly enough for a movie to be made out of it). There are several factors in play here: the fact that the same author also wrote many other famous stories (like “Night on the Galactic Railroad” that I discussed above), and also that “Matasaburo of the Wind” is actually connected to several of the author’s own stories, including “風野又三郎” which is confusingly named almost exactly the same, with (nearly) the same pronunciation as 風の又三郎, but with only the second character differing.

I think I will take another re-read or two, plus some research into this author and Japanese literature of that period for me to fully comprehend it’s depth and significance. Reading the other “Matasaburo” story (風野又三郎) would probably give me a very different perspective.

One thing that I found interesting is some subtle connections to Haruki Murakami’s writing style. Whether direct, or indirect through some other author(s), it seems Mr. Murakami was very likely influenced by Mr. Miyazawa (the details I’ll omit for now since it gives away an important part of the story). I’ll just say that it has to do with mixing just the right amount of real and fantasy elements.

This experience gave me a little more courage to go back and attempt “Night on the Galactic Railroad” someday, and it was a good sampler of Japanese literature from the 1920s. I think now it may be a little easier for me to break into other works of the same era.

If you are someone with several years of Japanese experience and haven’t yet experienced classic Japanese works, I think this as good place as any to start. You can read it for free (legally!) on Aozora Bunko here.

”Sluggish Symbol, Inane Illusion” (緩慢な表象と虚ろな幻想): PDF of full novel

Recently, one of my readers (swhp), offered to make a PDF file of a Japanese novel I had translated in full: ”Sluggish Symbol, Inane Illusion”

You can download the PDF at the below link.


In case I plan to come back and edit anything, please just link to this page instead of linking the PDF directly or uploading it to other sites.

It really made me happy that someone enjoyed this novel enough to want to make a PDF of it for their personal collection.

By the way, this is not the authoritative translation I discussed about making in these translators notes. Rather, it just a PDF version of the story as posted in several articles on this blog. But I still may create that someday.

In case you haven’t read this story yet, this is a good opportunity to experience it in an easy-to-read form. You can see the synopsis below for reference:


In the distant, but not too distant future, the profession of book author has disappeared. Every member of society is legally obligated to publish one book during their lifetime, but a second work is prohibited. Every book for sale in a bookstore is someone’s one and only published work. 

This is the story of a boy who keeps his words close to his heart and a girl who is in love with the art of storytelling.

Japanese learning site review:

Recently I found about the site, which at first seems strangely named, but makes more sense when you see it is a combination of the words “Reader” and “Japanese”.

The main theme of this site is to sell E-books which are bilingual English/Japanese. Generally I don’t recommend studying parallel text too much unless you have some good fundamentals under your belt, however having said that these books are designed expressly for the purpose of those studying Japanese and contain useful annotations on word usage, etc, so if used correctly they can be a great tool to learners of different levels.

There is a short sample E-book which you can easily download on the site, and once you find a book you like (they are organized by difficulty levels) then you can purchase it for Kindle on Amazon. The prices seem pretty reasonable for what you get ($3.00 for one of the products I looked at).

If you have questions about any of the Japanese in any of the E-books, the site even has a cool service that lets you ask unlimited questions about a single short text for a one-time fee.

While the bilingual E-books are cool and generally a rare resource, I actually like the blog of this site even better which talks about Japanese grammar and other topics. There is some useful articles like this one which talks about using って to introduce a new topic. I wish I could make more time to write about more Japanese grammar myself, but I’ve been more focused on translation side projects lately.

It looks like this entire site is made by a single guy who is (at least) trilingual and has pretty impressive credentials in terms of having been involved with Japanese since kindergarten, lived in Japan for several years, and having worked as a professional translator. It seems he has an amazing devotion to Japan’s culture and language, and I think nearly anyone of any Japanese level can learn something from him.

Japanese drama review: Mr. Nietzsche in the Convenience Store (ニーチェ先生~コンビニに)

Having become a member of the streaming service Crunchyroll a few months ago, I’ve sampled a few anime series, but for over half of them I ended up bored and never made it to the end. One exception was “Time of Eve” which I reviewed the other day.

Besides anime, Crunchyroll also has a few live-action dramas, though the selection isn’t that great so it’s even harder to find something good. But I was fortunate to stumble upon a totally hilarious comedy series recently which I just finished watching today, “Mr. Nietzsche in the Convenience Store”  (ニーチェ先生~コンビニに).

This show’s format is similar to an American sitcom, where the scenes are shot on only a few locations, and the focus is more on comedic dialog than a complex plot. As you might have guessed from the title, the setting is a convenience store, and the characters are mostly employees. While the actors don’t necessarily exemplify any award-winning acting, the characters are each pretty unique and they interact well together. For example, there is an ultra serious, morbid guy who wants to become a monk, a lowlife who thinks of nothing but get-rich-quick schemes, and a self-absorbed, childish manager.

The comedy is generally dark, extreme, and outright insane, but I enjoyed a large part of it.

You would think that the language of a bunch of people working at a convenience store at night would be pretty down to earth and informative for those learning Japanese, and while I won’t say it’s completely useless for that purpose, the intellectual/eclectic/weird nature of the jokes means there is a good deal of Japanese you might not hear too often in daily life, whether it’s buddhist sutras or made-up words (the latter is most commonly heard in the manager’s speech). But if you are looking for a challenge feel free to watch this series with the subtitles off and see how much you can figure out.

I can’t say whether the comedic style will be your thing, but I highly recommend watching one episode to see if you like it. You’ll enjoy it even more if you nibble on a pack of chocolate squid while watching. (That’s a reference to one of my favorite jokes in the show)


Japanese novel short excerpt translation: “Passaggio” (パッサジオ) by Hitonari Tsuji (辻仁成)

I recently reviewed Hitonari Tsuji’s Passaggio, and in this post I will be translating a small portion of it.

I usually like to translate the first chapter (or at least the first few pages) of novels that I enjoy, but in this case I found a certain passage that was written so beautifully I decided to translate just that portion instead. Since I don’t plan to translate the rest of the novel anyway, I think this passage will be more impactful than the first few pages, and will help you see the descriptive abilities of this author without giving away any of the story.

[taken from pages 30-31]

English translation

The sun had already began to dip down behind the woods, but the sky was still bright and stood out in stark contrast against the gloomy, ever-darkening trees. In the gentle evening sky, where a blue glow lingered that was neither navy nor indigo, a dazzling gradient boasted a range of colors from noon-sky hues to night-sky shades, all while an impatient Venus shined boldly as the center piece of the spectacle.

Just then, a wind passed through the trees and the sound of music appeared out of nowhere. It seemed to be coming from the other side of the now pitch-dark forest. The music sounded neither classical nor ambient, with a melody line falling and rising again and again, like the energetic dance of a bumblebee. It possessed a certain randomness to it, as if someone was fumbling at a keyboard instrument with their eyes closed. And yet, it would be wrong to say that the result was not a well-formed piece of music. There was a subtle internal structure, as if just enough randomness was added intentionally based on a well-defined calculation. The source of the sound seemed to be a synthesizer tracing the notes of a musical scale, triggering sampled sounds of water dripping.

Original Japanese text



Translation notes

Originally I was thinking of using this translation as a good case study to show the steps I go through during translation, but as I went through this I realized that a great deal of it goes on in my head before I actually put words to paper. So this time I’ve just given the translation with these notes, but in the future I may try to give a little more detail and show various versions up to the final.

I actually did the first rough draft of translation on a plane, so I looked up the meanings to some of the more advanced words first before I took off, even if I thought I probably knew their meanings already.

Because of the nature of this passage I went for a fairly non-literal translation, going more for a natural and expressive result than literal correctness. For example, rather than saying (somewhat) literally “remaining brightness that was neither blue, nor navy blue, nor indigo”, I said “blue glow lingered that was neither navy nor indigo”. This is partly because each of the colors in Japanese was written with a single word and in English I didn’t like the sound of just “navy” without classifying it as a shade of blue first, and also because the flow of two colors sounded better than three to me. I also changed the word “balance” in the second to last sentence to “subtle internal structure” since I felt it expressed what the author was trying to say better. Finally, I connected the last two Japanese sentences into one in order to improve the flow, and avoid translating the two words “よう” and “らしい” back to back which have similar meaning.

I also used some varied vocabulary to avoid repeating the same word several times, like in the first sentence I used “woods” and “trees” for the word 森 (mori) which is used twice.

I love how the author works to engage your visual sense, and then goes on to engage your audial sense, using the sense of touch (via the wind) to connect things together.

Japanese novel review: “Passaggio” (パッサジオ) by Hitonari Tsuji (辻仁成)

Of the last five books I’ve read, two were in English and three in Japanese. The English ones were both pretty good (one biography, one fantasy), but since they had little to do with Japanese I haven’t posted reviews for them here. The first two Japanese books were pretty mediocre, so I decided against writing reviews of them as well.

The last of these Japanese books, Passaggio by Hitonari Tsuji (published by Bunshun), was not just good enough to make me write a review about it, but was a very redeeming read in many ways.

I came across this novel in a Powell’s bookstore in Portland, Oregon which bills itself as the largest independent bookstore. The Japanese section of this massive, labyrinthine place generally has a good selection of Japanese used books. It doesn’t beat Kinokuniya or Bookoff, but from what I’ve seen it has more Japanese books than any bookstore in the US that doesn’t specialize in Japanese media.

After reading through the synopses on the backs of 15-20 books, Passaggio was one of the few that really caught my attention. Its cover was quite strange and somehow eerie (though I think the awkward image could have been produced better). At under 200 pages, its length was also attractive since I wasn’t looking for a several month commitment.

As for the story itself, while I won’t be giving any spoilers I will translate a portion of the book’s synopsis (the latter half I omitted seems like it gives away too much):


A professional singer who has lost his voice right before a concert follows a mysterious, alluring doctor, and ends up stumbling upon her grandfather’s immortality research lab hidden deep in the mountains.


One thing that I enjoy in a novel is a sense of mystery that carries you through to its conclusion, and this novel fulfills that nicely. There are some surreal parts where the boundary between a believable story and an unbelievable one is nearly crossed, and these were reminiscient of some of Haruki Murakami books. These supernatural elements were not quite enough to categorize the book as “fantasy” (though it is clearly fiction) and not enough to lump it in with everyday popular fiction.

Its plot is put together very well, with few unnecessary scenes or chapters. The ending was also excellent, predictable in some parts but resulting in an interconnected, complete whole.

Passaggio touches upon some important universal themes, including science, immortality, and music. It’s definitely a book for adults, with a few scenes of violence and sexual content not appropriate for children. But more importantly, it contains a philosophical depth that may have you thinking long after you’ve finished the last page.

I didn’t realize it at the time I bought this book, but Hitonari Tsuji is actually a famous Japanese author who has written over 40 novels, several of which were made into movies. I have read one of his other books before (アンチノイズ), but it was before I started this blog so I never wrote a review of it.

I discovered the author is actually a rock musician himself, and after doing some research about his life it almost seems like this book has a biographical aspect to it. While this novel is technically fiction, I feel that Hitonari really wrote from the heart about some of the themes interwoven into this book.

As someone who is interested in writing and storytelling, I found this book quite educational. Not only is the plot composition done skillfully, but the author makes good use of metaphors throughout, and his overall descriptive ability is superb. If I were to make a nitpick I’d say some of the paragraphs run on a little too long and could have been edited down a bit. But given this book was done in the early part of his career (it was published in 1998, about 9 years after he began writing), I’m sure his style is even more refined in his later works. His is still actively writing as far as I know, and has published a novel last year.

The Japanese in this book is advanced for a few reasons, one of which is that the author’s dense style tends to fill up pages with little space to breathe. The realistic treatment of music, science, and medicine requires there to be a few terms you may not be familiar with (ex: 声帯, vocal cords), although you don’t have to be a scientist or a doctor to follow what is going on. Furigana (Kanji reading hints) are also quite sparse. On the positive side, the dialog is easy to follow and lacks any regional dialects, and the book’s relatively short length makes it possible to complete even if you have to struggle.

After thinking about it, I’m on the verge of calling this book a classic because of it’s unique treatment of universal themes, told through a tight, well-connected series of events. Except for the fact that some of the technology referenced in the book (floppy disks, etc.) is over a decade old, I felt this story has aged very well in over 15 years. To me, a “classic” is one of those books I was forced to read in high school that was a challenge to finish, but touched on important, universal topics. If nothing else, I feel Passaggio falls squarely into this category, and I can imagine myself highlighting key phrases like I did back in those days and mulling on the various themes of the book.

You can find this book on Amazon Japan. You might be able to find it on E-book sites, though I haven’t checked any of those.

Unfortunately, this book has not been translated into English. In fact, I wasn’t able to find English versions of any of Tsuji’s books, even those who are more popular than this one (though there seems to be translations of some of his works into a few European languages). Why there are no English translations is a complete mystery to me.


The Art of Translation: My process for translating from Japanese to English

Starting late last year, I’ve translated as a paid side job at Gengo for a few months, as well as translated parts of fictional stories as a hobby, roughly 25-30 chapters worth. By no means would I call myself a “professional” translator, but I feel I have had enough experience to improve my translation skills significantly.

For several weeks now, I’ve been thinking it would be good to capture some details of how I do my translations, partially as a note to my future self, whether I eventually go on to do translation someday full-time, or end up finding another hobby to consume my free time. Though I’ve known there was a big group of fans translating Japanese Manga to English, I’ve recently discovered there is also a surprisingly large amount of people (on sites such as doing translations of light novels and other textual media. So I was also thinking that maybe my experiences would be useful to those who are considering doing translation themselves, either as a hobby or as a career.

In the rest of this post I will be describing in detail my general process for doing fiction translations. Non-fiction does share many similarities, but there are also some differences, so I will be primarily focusing on fiction here. My workflow has evolved a bit over time, and what I will cover here will pull heavily from the process I used for this recent translation. I’ll also be including a few things that I recommend but may have not actually tried myself. My process is designed such that I pretty much do everything myself, since for the most part I haven’t had people do any extensive editing of my work. If working with a team then there would definitely be some big changes to the workflow.

Also, this process is designed around obtaining the best possible quality result, without worrying about deadlines. If you are doing a paid translation with the clock ticking, then there would be some significant modifications. That is even more relevant if you were doing this for your full time job, since faster output generally means more money in your pocket.

As for requirements to be a good translator, I think you really need to be fluent or near-fluent in both languages. In order to be able to produce a natural translation, it’s best if the target language (the language you are translating into) is either your native language, one you’ve spoken in daily life for several years, or you just happen to be a linguistic wizard. Despite my interest in translation I haven’t even really considered translating the other way, from English to Japanese, because I know how awkward it would turn out. However, if you are working with a native speaker of that language who can edit for you, then you still can end up with a good result.

In order to be a great translator, I think translating into your native tongue is not alone sufficient. You also need to have great writing skills and editing skills in your language. I haven’t researched this deeply, but I feel that many professional translators have some experience writing and sometimes even publishing their own stories. Having said that, if you don’t have that much experience in writing, I think you can slowly learn this as you go, give that you have sufficient experience reading in your native language.

Even if you are not near fluent in the source language (the language you are translating from), if you have a strong interest in translation I think it’s fine to go ahead and try out translating something. I guarantee you’ll learn a great deal even if you never show the result to anyone. But if you do decide to show it somewhere online, I recommend finding another translator to double check your work, or at least make it clear at the beginning of your post what your linguistic level with the source language is.

But enough introduction–let’s get to the process.

Step 1: Read through the material (carefully)

This may sound obvious, but the first thing I usually do is carefully read through the entire work I am going to translate without thinking too much about translation itself. Even if I just plan to do a single chapter of a novel, I generally like to read the entire book first.

This also may seem obvious, but also make sure you understand it completely (at least the part you are translating). Even if you are fluent in the source language, there can be parts you don’t understand, and by this I don’t mean just words and grammar, but also deeper things like what the author is hinting at at a deeper (philosophical, spiritual, etc.) level.

Not understanding the source text completely will virtually assure you don’t end up with a great translation. Sure, you can put it off until during the actual translation process, but once you start thinking in two languages at once you’ll have less brain capacity to do that. So get it squared away now.

The only time I would consider it safe to skip this step (and translate as you read for the first time) is if you are on a tight time schedule, or if the later chapters haven’t been written yet.

Step 2: Background research

As part of step 1, there may be parts that no matter how many times you read them through, you still don’t fully understand what is going on.

One example is in this translation, where the source text had a small section about making charcoal. To be honest, though I had used charcoal before I had little idea of what it actually was, or how it was made. Although this only mattered for a few sentences, I decided to watch a Youtube video or two about the charcoal making process (in Japanese) and it was quite informative.

Another time you would want to do background research is if you are translating a genre you aren’t too familiar with, where you might want to read a book, or at least a short story in that genre. I’ve read my share of fiction books in English, and I’ve never done this yet. But for some genres I might still consider it.

Step 3: Decide translation priority

Before you start translating, I think it’s good to decide where you want to spend the most effort and time.

This may vary depending on the reason you are translating, for whom, and whether it is paid or not, but personally I think the title itself is the most important thing. It’s often the first thing people see, and it heavily influences whether they will start reading the work. After that, I think a summary or a teaser (like what is shown on the back of a novel) is second in importance. After that, of course the introduction, whether it is a prologue or first chapter, is extremely important for the same reasons.

If you are releasing in pieces (like chapters) and you have some control over the timing, another thing you should decide early on is how far ahead you want to translate compared to what you release. Generally, I recommend translating a few chapters ahead, or you can think of this as translating a few chapters before you release one. So for example you could translate chapters 1,2, and 3, and then release 1. The reason is that this may help you catch mistakes in translation early and also help with consistency across chapters. (See the section on pipelining later in this article)

For example, once or twice when I was translating a chapter, I realized that there was a better term than the one I had chosen, and this was a term that was used for the last few chapters. In those cases, I could go back and just edit the posts, but you don’t always have such an opportunity, and even if you do you’ll probably get the majority of people reading it as soon as it comes out.

Step 4: Prepare to translate

I’ve had my share of problems with formatting and having things get messed up with cut-and-paste, so I have developed a certain process to avoid these things so I can focus most of my time on the fun, and most important part: the translation itself.

Since my end result is usually to publish the translation on a blog, this process is tailored to that. If you were producing a Word document, you should change things accordingly.

I find it’s very helpful to be able to see the translated next right below the original, paragraph by paragraph, as it makes it easier to compare.  However to do this you’ll need the original text in digital form. If this is a paid job you should have that, but if you’re doing it as a hobby you may not. If possible, I recommend buying a digital copy and sometimes if you play around with it you can cut and paste little by little (though this only is realistic for a few pages and becomes tedious). If you are doing a light novel that is released online without any license, it’s easy to just cut and paste. Another idea for a work that you only have a physical copy is to scan pages in, so you can display them on your screen instead of having to look back and forth.

(Note: I’ll be talking about things in terms of chapters from here on, but you can choose your own unit of work as you want depending on the project.)

Since I publish on Wordpress, I’ll start a new article at this step and paste in the entire original chapter into the “Text” mode pane. The reason is so that I don’t carry over any weird formatting from the original source, in the form of hidden characters that can display differently with WordPress. One quirk with WordPress is that extra vertical spaces (carriage returns) will get erased when you click “Save”. So before doing that, after I paste the original text I stay in the “Text” mode pane and add the following HTML manually on each line that is empty.

<br class=”blank” />

Now, I can safely click the “Save” button, though I still will work in the “Text” mode until a later step. The reason is the same– there are weird quirks with WordPress, like for example there can be different types of carriage returns that result in differently sized vertical spaces. Keeping to “Text” mode keeps things straightforward, with no mysteries.

Step 5: Make a first cut at translation

Now comes the hardest, and most important part: the initial translation effort.

When I make a first cut at translation, my goal is to go through it from start to finish, line by line, and produce a good rough translation that conveys the meaning of the source text good enough that I could show it to someone if asked to. I always try to use proper grammar, natural wording, and generally do not skip sections or jump around.

If I come to a sentence that I can’t immediately think of a good translation for, I will spend as much time as needed–sometimes as long as a few minutes–thinking of a reasonable translation. Generally I’ll try to err on the side of making an over-literal translation as opposed to a artistic/non-literal translation, since I find it is easy to convert to a non-literal translation later as compared to going in the reverse direction.

I’m not going to spend too much time on the actual act of translation in this post, but this step is definitly more of an art than a science. However, I generally look up words online in a variety of places, including dictionaries and other sites. I’ll use native dictionaries (i.e. Japanese->Japanese) to more fully understand a word or phrase, and sometimes English dictionaries to make sure I properly understand a word I am considering using. I’ll also use thesauruses to look for word suggestions.

Often, even after searching in several places, I won’t be able to find a translation for a term or expression that I am satisfied with. Even if I find a candidate in a dictionary whose meaning is appropriate, it may not fit the tone of the passage or the personality of the character who is speaking. It’s also not unusual for dictionary examples to contain words that are no longer commonly used in modern language. In times like this, I just sit and think about it until it comes to me, sometimes looking through related materials looking for a better term. This is really the hardest part since there is no well-defined process. Just cutting and pasting a dictionary word (or a whole phrase) without completely understanding the nuance involved carries a high risk of creating an awkward result.

Besides individual words and phrases, generally the longer the sentence is, the more time it takes for me to shift around the elements in my head in English until I get something that sounds reasonable. I remember the first few paragraphs of one translation (Hibana) being especially difficult grammatically and it taking a great deal of time to sound even remotely natural. If a sequence of events is being described in time order you will usually want to persist the relative ordering of those phrases. But since Japanese allows modifying a noun using an entire phrase before it (which can be arbitrarily long), often you need to shift these around for sentences like this (example: “すごく疲れてきた僕は。。。”)

One rule I generally try to not break is to preserve paragraph and sentence boundaries from the source material.These both can have a major effect on the style of a passage, and I often find that keeping things grouped in the same way results in a good translation. However, in rare cases when it’s hard to make it sound natural in English, I may break a long sentence up into several smaller sentences, or, even rarer, join multiple sentences into one.

If possible, I recommend trying to complete the first attempt at translation in one sitting, though that may be difficult depending on its length and your available time. The reason is that I find that once I am “in the zone” I become much more productive and am less prone to make consistency errors within a chapter. If you do it across two or more sittings, you might want to review your work each time before continuing the translation.

I am considering writing one or more posts showing my translation process for specific passages, and if you are interested please leave a comment to that effect.

Step 6: Editing (with embedded source text)

The rest of the process from here on is basically proofreading and editing. There are two major things to check: that the translation accurately communicates the original meaning, nuance, and tone of the original passage, and that the resultant translation is in completely natural, grammatically correct English.

I always go through several iterations of editing, and my priority gradually shifts from the first of these objectives (accuracy) to the second (naturalness).

Since the previous step (initial translation) also involves an attempt to make an accurate, natural translation, inherently this step is similar. However, the difference is now that much of the hard work has been done, I can dedicate my energy to the remaining portion, which to be honest can still be quite challenging.

One good thing about the editing process is that it shares much in common with editing a story you have written yourself from scratch (not involving any translation). So if you have experience with that type of work, you can carry it over here easily. I have a bit of experience doing editing for a few short stories and novel chapters I’ve written, and I feel it really helped improve the quality of my translations.

You can find many websites and books about editing and improving your writing style, but here are a few things that I pay attention to:

  • Avoiding repeated words that are the same or sound similar (unless you purposefully want this for effect)
  • Punctuation (especially comma usage as I find it has a major effect on the flow of a passage)
  • Consistency (using the same terms throughout the entire work, unless there is a specific reason to vary something)
  • Unnecessary wordiness (ex: long sentences that can be shortened and retain basically the same meaning)
  • Word choice (is there a synonym that fits better with the tone or flow of the sentence, or is less likely to confuse the reader?)
  • Spelling (an obvious, but very important consideration)
  • Word definitions (just because English is my native language doesn’t mean my understanding of all words is perfect, so I am not afraid to double check the real meaning of a word or phrase in a dictionary)

There is a set of considerations to make which will vary depending on your source language. For Japanese, here are a few that I think about:

  • Particles are not often translated word for word, and some are very difficult to translate
  • Need to make clear subjects and topics which are often omitted in Japanese, unless rare cases where the original material is purposefully vague
  • Verbs can also omitted in Japanese, and because of word order differences in English you often have to make the verb explicit in the translation
  • Translating regional dialects (very difficult, maybe I can write another article about this)
  • Japanese’s many first and second person pronouns (ore, boku, kimi, anata, anta, etc.) can be very hard to translate, so don’t force it. For first person pronouns, try to use word choice to adjust the tone of a character’s dialogue instead.
  • Make an effort to translate various politeness styles (i.e. desu/masu-chou) by using things like word choice and usage (or lack of) contractions, etc.
  • Though it depends on the author, I feel incomplete sentences are generally employed more often in Japanese, and in some cases it may not be natural to retain that in English
  • Be aware of the differences between Kanji variants of the same word, which can range from subtle (見る vs 観る) to drastic (攻める vs 責める)

Above all, don’t ever get regimented in translating one word or grammar pattern the same way each time. For example, just because you translated the word nante or the ending -tara a certain way before, that doesn’t mean you want to translate it the same way every time you come across it again. The context is always important. However, if the author is using repetition of some word for emphasis, odds are you will want to use repetition of some word in English to achieve a similar effect.

Generally I go through this process several times, depending on the complexity of the original work and the reason I am doing the translation. I would say the minimum is 2-3 times, but for some difficult translations I have done as long as 5 of more iterations. Ultimately, I don’t leave this step until I am very satisfied with my translation.

One important thing you can adjust which usually doesn’t change the meaning itself, but can have a major effect on the fluidity of a sentence, is word order. Try a few different word orders and see which sounds best.

I feel it is important to leave some time in between each iteration, at minimum 24 hours or so. The reason is to let your mind get refreshed and reset from whatever you were thinking during the last iteration. It’s a funny thing, but it’s not uncommon for me to come across mistakes, unnatural phrasing, or more appropriate ways to translate something on the 3rd, 4th, or even 5th iteration.

One trick I often use to look for mistakes and unnatural-sounding phrases is reading the text aloud, and I highly recommend trying this.

Sometime during this editing step I generally change to working in “Visual” mode of WordPress since I will not be tweaking formatting much any more.

Step 7: Editing (without the embedded source text)

All editing up to this point was with the original source text embedded every other paragraph. While this is great for checking for accuracy, it can be distracting when you are just trying to consider the fluidity of your translation, and ultimately you will not be delivering your final translation in this form anyway.

That’s why it’s important to remove all the original text and do at least one more full editing iteration (or more) before you consider yourself done. I’ve tried before to delete paragraph by paragraph as I do an iteration, but found that distracting, so I just usually go through it once deleting all the original text without really trying to read anything. This is one of the steps where it is clear that a specialized translation tool would save time.

One of the most important parts of the editing steps, determining whether something sounds natural, cannot be broken down into any set of rules or conditions. It can only be done using a linguistic intuition that is built up over time, from reading hundreds if not thousands of works in that language. Since I have been a fairly avid reader since I was young I feel I have developed a reasonable sense of this, however I plan to keep reading more books to further polish it. Whenever I read a book I now try to analyze the style as I go, thinking about what I like about it and what I don’t. If I’m lucky, I will be able to leverage more and more of what I read in my future translations.

However, one trick that may help is to search for phrases you are considering online. For example, let’s say I am considering using the expression “she batted her eyelashes”. I can search for it on Google (using quotes) and see over 50k hits. (Make sure you advance a few pages since sometimes Google cheats). However, using a phrase like “I slowly jumped” yields only ~50. While the latter is not grammatically incorrect, you could argue it is a little unnatural, or at least uncommon. Ultimately, you’ll still have to use your word-sense instincts to make your final decision of how to translate something.


When you start to comfortable with your own translation process, you can start using what I call pipelining to achieve more efficiency. This means you are working on different portions of several chapters on any given day. This will save time especially if you are trying to leave some time between successive iterations of editing of a given chapter. While you won’t want to edit a chapter twice in one day, you can edit two chapters, each in different stages of the process. Here is an example of this flow:

  • Day 1: Read Ch 1
  • Day 2: Initial translation Ch1, Read Ch 2
  • Day 3: First edit Ch1, Initial translation Ch2, Read Ch 3
  • Day 4: Second of Ch1, First edit of Ch 2, Initial translation Ch 3, Read Ch 4

Final thoughts

Since there is a certain amount of artistic license required to do fiction translations, there is a subjective element about what makes the “best” translation of something, and this can differ from person to person. Surely, if you had 10 professional translators translate a single page of a fictional work, you’ll get some wide variance between the results, though some of this will depend on the content of the source text.

Having said that, to a reader who is a native speaker or at least fluent in the language a book is written in, there is a huge difference between text with badly written, awkward grammar, and text that is clean and easy-to-read.

After all has been said and done, the most important thing is not that every single phrase was translated 100% perfectly, but rather that the overall impression received by the reader is favorable, and he or she continues to read to the end, and of course enjoys the story. Of course any philosophical or otherwise messages the author tried to convey should also come through in the translation, though that can be subjective as well. I’m not advocating for slipshod translations, since accuracy is very important, but rather that at some point you should take a step back and think about the work as a whole. This is one area which can be very different from non-fictional translations, where accuracy is even more important, and in some cases it may be OK to sacrifice some naturalness for the sake of accuracy. For example, even a well-written product manual would be hard to “enjoy” as you read through it, so the translator’s priorities may be shifted around a little.

Once I remember going back to read an old translation I had did several months ago, for which I had forgotten much of the original text in Japanese. Reading it with a fresh mind, I remember enjoying it in a new way, and I realized that if something is translated sufficiently well, I might actually enjoy it more in English than in the original Japanese. This is because no matter how long I study Japanese, it is hard to acquire the same sensibilities as I have gained in my native language from a lifetime of reading in that language. This will certainly vary from person to person, since some people may consider themselves to be equally fluent in both the source and target languages, but I use this as an indicator of what makes a good translation. In other words, if I were to forget something is a translation and all the details about the original text, can I really enjoy that translation on it’s own, as if it was originally written in the target language.

I hope you enjoyed this essay on my translation process. Feel free to let me know if you have any comments or questions.