I recently announced my latest ebook “Tales of the Disturbed”, and in this article I’d like to talk a little more about the process involved in producing the book.
To be honest, I don’t remember where I first discovered the author Yoshio Toyoshima (豊島与志雄, Japanese wikipedia page). I do know that it took a few weeks of searching on the Internet. One of the things I had done was go through all the ranking of the most popular stories on Aozora Bunko (here is the list for last month), but as you might expect the vast majority of those works were already translated into English. I also did a bunch of searches using keywords like 幻覚的 (which roughly translates to ‘surreal’). There are surreal elements in some of his works, so perhaps this is what led me to him.
Toyoshima has a large number of stories available on Aozora Bunko (over 300) so it took several more weeks for me to sift through them. I picked a few with interesting titles and read them, stopping partway through if I wasn’t enjoying myself, and I also found a few Japanese audio narrations on Youtube. In the end, I think I only choose one of the stories with an audio narration (“Drop Dead!”). I made a list of various stories I enjoyed and why, along with their character count, and used that as a reference when making my final picks. I was fortunate that a book had just been published in Japanese earlier this year with reprints of a bunch of Toyoshima’s stories, and this also gave me some hints.
During the selection process, I noticed that two of the stories I liked had some things in common, so instead of bundling a random group of stories I decided to focus on the theme of mental disturbance and pick a few more stories that fit in that category. After going through a few title ideas I finally decided on “Tales of the Disturbed”, and was glad that this exact title doesn’t seem to have been used before. I also chose a Japanese title: “壊れし者たち”, though it didn’t make it to the final version of the cover.
I was fortunate that most of the stories I liked had not been translated into English before. The only exception was “Ghosts of the Metropolis”, which is part of this compilation by Kurodahan. It made me happy that a publisher had selected one of his stories to publish, however. I didn’t read this story, though, because I wanted to develop my own style for my translations.
The production of “Tales of the Disturbed” was similar to some of my earlier books, which I have described in detail here and here. To summarize, I used Google Docs for the initial translation and most of the editing, Apple’s Pages software for the final E-book construction, and then exported to EPUB to import into Amazon. I used Canva for cover design, and made a few covers from which I selected the one I liked the best (you can see my 2nd favorite cover at the end of this article). I spent some time looking through pictures from the Meiji and Taisho eras in order to find one with a feeling of classic Japan, but I couldn’t find many good photos from those eras, except for one or two that were really expensive (I remember one I liked was $500 on a certain stock photo site.) The final cover I chose was an abstract picture of a spiral ladder made by Swiss photographer Oleg Magni, from the free stock photo site Pexels. You can see more of his work here.
One way this project differed from my other works was the difficulty of the source text. Not only were the stories older than my typical translations (“The Allure” was first published in 1914, over 100 years ago), but there was a significant amount of kanji characters and words not commonly used in modern Japanese. Fortunately, I had several native speakers I could ask to help verify the meaning in a few places.
Besides the age of the source texts, the stories I chose had longer than average sentence and paragraph length, especially “Nightmare” which began with a massive multi-page paragraph. While this reminded me of other classic literature I had read––and I did enjoy the challenge of translating such dense, expressive text––it meant the translations took a lot more time.
In previous projects, after the initial translation I usually did occasional spot checks against the source text, but for stories in “Tales of the Disturbed” I referred a lot more to the source texts, making significant changes even the day I sent the file to Amazon.
In addition to the linguistic complexity, these stories also tended to have a higher conceptual complexity than other works I have translated. “The Kerria Flower,” for example, uses symbolism as a key element of the story. It begins with the following cryptic sentence:
After some thought (and reading through the story once or twice), I ended up rendering this literally as:
In the center of the lake was an eye.
While I feel it’s best for a translator to understand both the literal meaning as well as the implicit meaning of the source text, I don’t think the translator should give hints to the reader about the latter any more than the original author did. (The exception would be if there is some cultural or linguistic difference that makes a literal rendering convey less meaning than the original text).
As another example, “The Allure” is a wild metaphysical ride that I don’t think was intended to be an easy read––and it happens to be my favorite story of this collection. The abstract nature of the story gave me a little more leeway in translation, though there were some linguistic challenges along the way. For example, words like 心 and 魂 are used freely, and while these have rough translations in English, depending on the context the nuance can differ. Especially 心, which can mean “mind” or “heart”. Perhaps the trickiest word in this piece was 生命, which I struggled for hours to find the best word choice for. (In most cases I ended up with the straightforward “life”, though in one case I had to tweak to “lifeform”).
While these works are (mostly) fiction, the fact that they are set in the real world meant there were certain topics that took additional background research. For example, there was one paragraph in “An Unfortunate Misunderstanding” that referred to the aftermath of World War II, and I spent a few hours just to make sure I was conveying the meaning properly (despite the fact that this part was arguably not central to the story itself).
The story “Drop Dead!” posed another type of challenge, since it was rumored to be about Osamu Dazai, the author of “No Longer Human” (人間失格). I did some research about this connection, but did not adjust the translation in any way to make it sound more or less like it was about Dazai. However, I did put a translator’s note after the story to explain this potential historical connection. For those who were familiar with Dazai, I wanted to see if they could make this connection before I told them about it.
For proofreading, while I did go through each piece a large number of times (15-20), as usual I leveraged a few kind-hearted people to help ferret out mistakes and unnatural parts I missed. It seems like sometimes I’ll overlook certain mistakes repeatedly on a handful of proofing passes, and I need to work on starting with a fresh mind each time (giving time between iteration passes helps, though). Grammarly also helped me catch a handful of issues, albeit less than last time (there seemed to be a lot of false positives this time). Oddly, the spell check that Amazon provides as part of the upload process didn’t seem to catch one of the obvious spelling issues I discovered later. It wasn’t like I was relying on that, but I was surprised that it didn’t give a single spelling mistake for any of the EPUB files I uploaded. As always, I used the practice of reading the text out loud to help catch mistakes and awkward phrasing. (This gave me the idea to do an audio narration for one of the stories, so stay tuned…)
For the ebook construction process, by using a Pages file of a past project I was able to save a lot of time and trial-and-error. The only main frustration I ran into was once when uploading a new EPUB file or cover image to Amazon didn’t reflect in the MOBI file that was exported (though it did reflect using the online previewer). I ended up writing to Amazon support and it took them over 24 hours to provide a response (which didn’t provide much, if any useful information), although by that time the problem had resolved itself.
Unlike my past books where I use formatting with a space between paragraphs, this time I wanted to get the visual presentation as close as possible to a professionally made book, and went with no spaces between paragraphs, plus indenting except for the first paragraph of a set. The drawback of the no-spaces formatting is that the page count is less, and that (combined with the long paragraph length and the fact I didn’t include the source text in parallel format) made it look like this book has less content than my other books, whereas in reality I think it has the highest word count.
There were only a few passages that I felt needed translation notes, so instead of making numbers referencing a separate section (which I did in my first book and discovered it was very time consuming) I simply used brackets embedded within the text.
A week or two before releasing the book I had stopped by a bookstore where I saw a book of short stories translated from Japanese. Below the title for each story was the date for first publishing and the original work’s name in romaji. I thought this was a nice touch and decided to emulate it, except that I used Japanese instead of the romaji. The publishing dates are important for these stories since there are some historically related parts, and the titles in case someone wants to easily find the original texts for comparison purposes.
As with other books, I did some proofing on a variety of devices, including an iPhone, iPad, and Kindle Paperwhite. Fortunately, I didn’t run into any weird platform inconsistency issues (which I had seen before when I attempted drop caps with a past book). Previewing on a variety of devices not only allowed me to check the formatting, but I noticed that looking at the text in a different medium (as opposed to Pages or Google Docs) helped me catch mistakes that I had otherwise missed.
A final significant difference with this project was that when I started in October, I gave myself a pretty hard deadline of the end of the year. I choose my initial content so that I could make that deadline easily, and then ended up adding two more stories when I realized I had the time. At some point, my deadline firmed up as Dec 21 in order to line up with a few things, and the last few days took a lot of editing to make that goal (fortunately I had a few days off work that really helped).
While it can be argued that spending another week (or more) on the book could have improved the quality a little, I think setting goals like this is an important part of my training as a translator. This sort of thing helps train not just understanding how long it takes to translate a work of given complexity and length, but also being able to choose battles in terms of what areas to give more focus and which less in order to make the deadline. One thing I did have to sacrifice for time is going back and forth with proofreaders on certain topics, and having them go through 2nd and 3rd revisions after I mad changes based on their feedback. So if there were any mistakes that slipped into the final version, I can only blame myself (:
(In retrospect, it probably would have made more sense to wait until early January to release the book, and I’ll keep that in mind for future releases)
If you would like to see more details about the book, or read a sample, please see the Amazon page here. I have set the price to $0.99 for the time being. While I only get around $0.33 on each purchase, I can tell you that to me every sale counts!