Translation as editing and adaptation: An example from “End of the Microcosmos” by Liu Cixin

I recently finished translating one of Liu Cixin’s short stories, 《微观尽头》, or “End of the Microcosmos.” It's a very short story, only 2,400 words, about what kind of surprising conclusions we might be able to draw from Large Hadron Collider experiments. Among those witnessing the story’s central experiment are the chief engineer at the research center, the two top theoretical physicists in the world, and an old shepherd.

Why an old shepherd? Because the research center was built on his land in Xinjiang, and the scientists wanted someone who doesn't understand physics at all to witness the event.

Here's where I'll pause to give some background and talk about the larger translation issues I run into. People may think that translation is a straightforward task of converting one language to another like converting file types, when really, translation involves a lot more editing and judgment than that. In particular, translators have to account for the sensibilities of their audience in the target language.

So that leads me to streamline and edit the prose of my translation as I go along. Some of the sensibilities are matters of taste, like anglophones finding prose amateur if it uses a lot of adverbs, while heavy adverb use is the norm for Chinese prose. For example, rather than converting 慢慢地走過來 directly into “walked over slowly,” I can choose to say “trudged over,” which would go over better with anglophones even if it's not as “strict” of a translation.

But then there are the bigger questions of tailoring narratives to fit audiences. Ken Liu introduces the problem well in his recent interview with The New York Times. When does a change to suit audiences become perceived as unfaithful to the original? Chinese fandom has had mixed reactions in particular to the edits made to the second volume of the Three Body series, The Dark Forest, in the name of reducing what anglophone audiences would read as sexism.

There's a delicate balance to be struck there, and I never want to go so far as to censor an author’s work. But I think there are times when such differences in ideology, often reflected in the terminology authors use, can distract readers from the primary intent of a story. Translated work brought into an anglophone world already faces hurdles to acceptance. Why add to that when I can smooth things over in a minimally invasive way?

Which brings me back to “End of the Microcosmos.” I want to bring in additional context here. The old shepherd in the story is Kazakh, one of China’s ethnic minorities, and also a Muslim religious minority. Although I have Manchu heritage, functionally, I'm part of the Han majority ethnic group, as almost all Chinese sci-fi authors are. There are systems of privilege and oppression there, too: The Chinese government is currently using horrific tactics to suppress and forcibly assimilate its Muslim minority populations. So I wanted to be very aware of how I translate the Kazakh character, so that I would, hopefully, not perpetuate further oppression through an insensitive translation.

I couldn't change the fact that the Kazakh shepherd character adhered to similarly negative tropes as the ones indigenous people face in the United States: the character who is ignorant but in touch with the land, positioned against more “cultured” protagonists to provide the mystic sagacity of the Other. Whether or not people even agree with my reading isn't certain. But ultimately, that was the author’s choice, and that choice reflects his Han privilege and perspectives. I would find changing the nature of the character to be too invasive for the story, in that it would change the entire dynamic between all the characters and would affect how the story plays out.

But there were small details with more leeway that I could focus on and improve. The shepherd’s religious beliefs come through strongly in his perception of the universe, as contrasted with the scientists’ perceptions. But, at the same time, the original Chinese manuscript has the character using “Allah” as an exclamation like others would use “God,” such as in “Oh my God!” 

That rang extremely false to me. With some research, I confirmed that feeling and decided to change those instances into more appropriate phrases. Sometimes, it was as simple as saying “What in the world is going on?!” instead of “My God, what’s happening?!” But other times, I felt the exclamation needed to be retained to best show the character’s reaction. I chose to replace those exclamations with the more appropriate terms “mashallah” and “subhanallah.”

The exclamations are an example of where I'd smooth things over because that incongruity would simply be a distraction to the narrative, when in the end, the exclamations are a minor detail that hardly impacts the story. Choosing more appropriate translations is, to me, a way to adapt the manuscript to its target audience, especially so that it can read more comfortably to Muslim anglophones.

I'm not writing all this to pat myself on the back, but rather to show what kind of decisions go into my translations, and to show that translating a story to be sensitive to its audience often takes extra effort and research. For these reasons, I don't think machine translation can ever replace human translation. This example is only a sliver of the kind of cultural nuance translators have to grapple with, and the kind of cultural literacy I expect translators to have, both in the original context and the target context.

Appearances: Dive into Worldbuilding & World Fantasy Convention

My recent and upcoming appearances.

I spoke with Juliette Wade today on her show Dive into Worldbuilding, where I discussed language, assimilation, and diaspora, and how they relate to my mermaid hunting story “As Dark As Hunger,” forthcoming November 2019 from Black Static. You can view a recording of the interview on YouTube.

I’ll also be attending World Fantasy Convention this week in Los Angeles. I’ll be around most afternoons, but I most likely will not be staying in the evenings, except for the mass autographing session. My full schedule is as follows:

California Screaming: Modern Golden State Horror Stories and Writers
Thursday, October 31 from 2:00 PM to 2:55 PM (55 minutes)

Join horror writers as they discuss modern stories set in California. What makes California a unique setting for horror, both psychological and supernatural, and what can stories set there tell us about the nature of fear?

Matt Maxwell (moderator), S. Qiouyi Lu, Laurie Tom, Kary English

Autographing with Jacob Weisman at Fairwood Press table
Friday, November 1 from 2:00 PM to 4:00 PM (2 hours)

Mass Autographing
Friday, November 1 from 8:00 PM to 10:00 PM (2 hours)

Reading: S. Qiouyi Lu
Saturday, November 2 from 1:30 PM to 1:55 PM (25 minutes)

I will be reading my story “Where a Heart Would Fit Perfectly” (4,200 words).

The Pros and Cons of Sensitivity Readers
Sunday, November 3 from 10:00 AM to 10:55 AM (55 minutes)

Sensitivity readers can help authors avoid mistakes portraying other cultures in their works. So do fantasy authors need sensitivity readers? If so, when? How does one go about finding and hiring them? What are the downsides to having them?

Rhys Ford (moderator), Laurel Anne Hill, S. Qiouyi Lu, Christine Taylor-Butler, Tad Williams

Feel free to say hi if you see me at the convention! I’m always happy to meet other genre fans and creatives.

Naming characters

Methods for creating bilingual character names.

Most of my characters are Chinese-American like me. They usually have a Western name, a Chinese name, and a Chinese surname. I’ve always loved naming characters, and this structure gives me opportunities to explore both branches of my heritage. (I use modified or other methods for non-Chinese characters and characters in secondary worlds.)

Historical name records

I no longer remember where I heard this, but I once came across the idea that names from your grandparents’ era cycle into popularity again in a few generations. Contemporary names from the last decade, meanwhile, can often sound too current. So one way I like to find first names for my characters is to go through the Social Security Administration’s historical name data until I find a name that I like. You’ll be surprised by how many names you think are modern that have actually been around for a long time. (See also: The Tiffany Problem)

Of course, this method has drawbacks: It’s based on data from the US, so names are overwhelmingly Western in origin, and the data doesn’t take into consideration differences in cultural backgrounds. But if you’d like an unremarkable Western name, I find this method to be more fruitful than combing baby name sites.

The Hundred Family Surnames (百家姓)

There are hundreds of Chinese surnames, but the most widely used, as well as some historical examples, are included in the Hundred Family Surnames, available in full on Wikipedia with links to individual surname entries. I like letting my eyes wander until I find something that sounds good or is otherwise significant for the character I’m developing.

I also use the character’s surname as a starting point for understanding their personal history as both a Chinese-American and as a member of the Chinese diaspora. For example, if I choose to transliterate a surname using Jyutping to reflect a Cantonese pronunciation, I’ve made a decision about the character’s heritage. I’ve also made a cultural and historical statement: In work derived from our real-world history, my romanization choice is a reflection of the immigration patterns that the Chinese diaspora has followed, as well as the different sociolinguistic groups that emigrated. So my choice there isn’t neutral; it’s a deliberate act of placing my character within a larger setting and context, which inform the character’s upbringing and perspectives.


I take particular pleasure in choosing a character’s Chinese given name. Unlike English names, the meanings of Chinese names are transparent to Chinese readers and listeners. Names are selected not only for how they sound, but also their semantic connotations. At the same time, an anglophone audience unfamiliar with Chinese sees the name “Xiaoxi” as equally opaque as an English name like “Stephanie.” So I find Chinese names a wonderful place to hide Easter eggs and play with the significance of a character’s name.

The downside is that, to come up with natural-sounding Chinese names, you need some understanding of Chinese. Working with a native speaker is probably your best option, especially to check whether a name seems feminine, masculine, or gender-neutral. Without a native speaker, you can still draw a lot of inspiration from dictionaries. I like Pleco (iOS/Android) in particular for its detailed entries and its interface, which encourages word and etymology discovery. I search terms that might be relevant, or start from a sound that I think is a good match, and go from there.

Of course, there are many ways to name characters, including naming them after historical figures, celebrities, etc. However, I find that Chinese people don’t tend to name people after others, particularly relatives—the idea of my brother sharing my grandpa’s Chinese name feels incredibly weird. So I prefer to start from scratch and have fun with meaning.

Feel free to share this issue, which is open to the public. If you enjoy my newsletter, I have both free and paid subscriptions available. Thank you!

Translating verse: A couplet from Chiung Yao’s “Princess Pearl”

Stylistic approaches and considerations for translating poetry.

First, some housekeeping: I’m switching newsletter providers from Mailchimp/Tinyletter to Substack. While most of my newsletters will be free, I’ll have at least one paid post per month for subscribers. Subscriptions start at $5/mo or $50/year. Substack also has more community features than Mailchimp and has more robust archiving features than Mailchimp, so feel free to share links to my newsletter!

The content of my newsletters will also be changing. Rather than sending irregular updates about my publications and appearances, I’ll be focusing more on craft, process, and methods. Posts may range from brief notes to deep dives. Please feel free to send me any suggestions for topics you’d like me to discuss.

And now, for our main topic:

Translating verse: A couplet from Chiung Yao’s “Princess Pearl”

All translation is both an act of interpretation and an act of judgment. The translator has the power to adapt and edit texts for their target audience, and there is no such thing as a “perfect” translation that will convey every semantic nuance between languages.

Still, I find prose more straightforward to translate than poetry, as there are usually limited numbers of interpretations for each sentence, and there is a flow to the story that prose adheres to. Poetry, on the other hand, not only has vast room for stylistic liberties, but also for semantic interpretation. For a truly wild extrapolation of this, read Douglas Hofstadter’s Le Ton beau de Marot: In Praise of the Music of Language (1997), which features several translations of the same French poem as well as commentary on the act of translation.

I want to illustrate just one example of translating verse in this issue. The following couplet appears in the first chapter of Princess Pearl (sometimes called My Fair Princess) by Chiung Yao (《還珠格格》作:瓊瑤):


A conservative approach that captures the diction of the Qianlong period (1850–1861) would be something like the following:

How distant the mountains, how distant the rivers;
how far the scenery on this long journey!
Longing for the past, longing for the present;
longing until so too disappears the spirit.

But, although the translation hints at the repetition in the original Chinese, it doesn’t quite capture the playful rhythm, which becomes more obvious when transliterating the Chinese verse into pinyin (presuming, as is probably likely, that the author was writing for a Mandarin-speaking audience):

shān yě tiáotiáo, shuǐ yě tiáotiáo, shānshuǐ tiáotiáo lù yáoyáo! 
pàn guò zuó xiāo, yòu pàn jīnzhāo, pàn lái pàn qù hún yě xiāo. 

In addition to the repetition of terms like 迢迢 tiáotiáo, the -iao rhyme (spelled “yao” when the syllable begins with a vowel) occurs throughout, creating a lilting pattern fitting the verse, which is meant to be sung.

Here’s another take on translating the verse, one in which I try to preserve the rhyme and assonance:

Faraway mountains, faraway rivers;
faraway lands on a path far away!
Longing for yesterday, longing for today;
longing to and fro for my spirit’s belonging.

It’s a futile question to ask which translation is more “accurate.” Both of them are, strictly speaking, accurate, in that they convey the semantic content of the original Chinese. Both translations have their pros and cons. The first conveys the majesty of the original while losing some playfulness, while the second captures the spirit of the original but may sacrifice some clarity with its wordplay.

The job of the translator is to decide between interpretations based on their experience, understanding of the text, relationship with the target audience, and their own discretion. Ultimately, I would say that a translator’s voice is the sum total of all these kinds of decisions in a piece. While I wouldn’t argue for a translator to impose their voice over the original author’s, I think it’s overly simplistic to act as if the translator doesn’t leave a trace on the translated manuscript. Even machine translation has its own philosophy toward semantics and style encoded into its development.

If you enjoyed this newsletter, please feel free to share it and to subscribe. I’ll be delivering future issues on similar topics in writing, translation, and editing. Thanks so much for reading, and I hope to see you back in the future!

Loading more posts…