A little while ago I was working on the draft of a novel when I came up against a task I’ve never found especially interesting: making up a word to describe a fictional concept. I enjoy higher-level linguistic worldbuilding, but the nitty-gritty of inventing vocabulary is tiresome. The seed of this essay was a simple thought: why am I using a convention I myself don’t like, and is there a different approach I could use instead? I found the process of answering that question for myself fruitful both as a reader and a writer, and it’s that process I want to record here. This essay is meant to be neither exhaustive nor universal. It stems, instead, from my personal engagement—as a bilingual speaker of English and Slovenian—with a core conceit of science fiction and fantasy.
Pseudotranslation is the idea—prevalent in, but not unique to, SF/F—that the author did not write the stories we read, but translated them from some fictional language into English (or whatever other language we happen to be reading in).1 Sometimes, the act of pseudotranslation is called out explicitly: J.R.R. Tolkien and Gene Wolfe framed The Lord of the Rings and The Book of the New Sun as translations from a fictional language. More often, the pseudotranslation is implicit. Any book in which a character speaks or thinks in a fictional language, but that language is rendered as English—in other words, virtually all secondary-world fantasy and a large proportion of science fiction as well—is by definition a pseudotranslation.2
For the most part, the suspension of disbelief demanded of us by pseudotranslation is not taxing. It’s not so hard to accept that English is standing in for one or more fictional languages, and with a bit of linguistic worldbuilding, the author can help us maintain that illusion through, say, the use of linguistically diverse character names and other proper nouns to give us a sense of what the different languages might sound like.
The notable exception, for me, is the use of made-up common nouns. These are ubiquitous in SF/F to the point that they are equated with the genre as a whole, especially by its detractors,3 and I almost feel guilty for taking aim at them! Clearly, many readers of SFF are not bothered by made-up words. Much of the time, neither am I. Occasionally, though, my brain will shift into a more analytical mode; the suspension of disbelief will be broken; and I’ll be left thinking that surely there’s a better way to talk about made-up concepts.
Having said that, my purpose is not to identify a problem (and certainly not to claim I have a solution). Instead, I want to think about the way I (as a reader) interact with made-up words in pseudotranslated works. I want to identify why they sometimes bother me and sometimes not and develop techniques through which I (as a writer) can engage more deliberately with the concept of pseudotranslation.
First let’s define what kind of made-up words I’m talking about. I’m not concerned with words like “ansible” or “orogene,” because these are derived from existing English words (“answerable” and “orogeny,” respectively). Nor am I bothered by newly coined compound words like “stormlight.” My brain happily accepts these as future developments of modern English, or as words in a parallel, fictionalised English. Even though at least one of these examples (stormlight) is a pseudotranslation, my disbelief is suspended. The same holds true for other languages—in Girl, Serpent, Thorn, for example, my brain engages with the text as a pseudotranslation from a fictionalised version of Persian, and thus Persian-derived words like “parik” don’t bother me any more than “ramen” would in a genuine translation from Japanese.
A more interesting example is the plethora of made-up words in Neal Stephenson’s monumental Anathem. This is a book whose foreword informs the reader that it is set on a planet that is not quite Earth, on which not quite rabbits munch on not quite carrots. It’s not a leap to infer that the characters are speaking not quite English—but, Stephenson is quick to reassure us, he has no intention of referring to rabbits as (say) tibbars4 because the minor difference between a tibbar and a rabbit would only distract the reader. A very sensible choice, but a curious comment to make for an author who proceeds to fill his book with dictionary definitions of made-up words, least of all the titular anathem. He gets away with it because his neologisms are, for the most part, astonishingly deft: I can’t think of a better example of made-up words that sound like their meaning. Saunt, sitting as it does at the intersection of “savant” and “saint,” is a natural word to describe the most revered of Anathem’s philosopher-mathematicians; the double meaning of anathem is captured by its straddling of “anthem” and “anathema.” Stephenson explicitly establishes Anathem as a pseudotranslation, then uses this conceit as he sees fit: for concepts with existing English counterparts, such as small herbivorous mammals, existing English words will do. For more culturally specific concepts—for example, the rituals and ceremonies that pervade every page of the book—Stephenson has coined words that convey the sense of the concept. In other words, my brain is interpreting anathem not as a word in a made-up language—even though I am told explicitly that that is what it is!5—but as an English translation of such a word, in the same way that every other word in the novel is an English translation. The fiction holds together because I’m never asked to imagine what that original made-up word might look like.
It follows that problems sometimes arise when I am asked to do that. This is the category of made-up words I want to focus on: the “angreal” of Wheel of Time, the “valonqar” of A Song of Ice and Fire. In short, any word that is presented, in-universe, as belonging to a made-up language (ignoring any necessary transliteration into the Latin alphabet), and which moreover is obviously invented. (I’m using this phrase loosely to exclude the types of words discussed in previous paragraphs.) Such words are loose threads through which, in my experience, the internal consistency of linguistic worldbuilding begins to unravel—especially when multiple languages intersect. Why are these words untranslated when every other word is not? A common answer might be “because they are in yet another language, and not the one the characters are speaking.” But this answer only boxes the author in: what happens if a made-up word is, elsewhere in the book, spoken by a character in that other language? Almost always the answer is that nothing changes. English stands in for that language as well, except for the made-up word, creating a sense of foreign-ness where there should be none. The real answer is “because the word stands for a concept that does not exist in English,” but that’s an unsatisfying explanation when I’m looking to immerse myself in a fictional world. Can we do better?
Let’s take a specific example. In Wheel of Time, the various in-universe languages spoken by the characters are mostly descended from a single ancestor: the Old Tongue. We are told that the Old Tongue sounds familiar even to characters who do not speak it—and yet we as readers do not have this experience. This familiarity is lost in pseudotranslation. One response might be to say so what? Many things are lost in translation from real-world languages as well. Such is the bane, or the gift, of the translator. But that would be making a false analogy: despite the name, pseudotranslation is nothing like translation. The difficulties inherent in translation are entirely out of the control of either author or translator. The author of a pseudotranslation, by contrast, is in full control of one side of the equation. Their fictional language is a black box. It can be given any characteristics the author wishes to achieve a desired effect. Robert Jordan, for instance, could have chosen to make his Old Tongue sound more like Old or Middle English—then that frisson of familiarity would have been preserved. Instead, he opted for a language that looks unapologetically fantastical to the anglophone eye, full of apostrophes and unusual letter combinations.6
I’m not claiming one choice is better than the other—only that he had a choice, and where there is a choice, there is a tool. In the remainder of this essay, I want to think about how writers might harness that tool.
Here’s a message I recently sent my mother:
Ja, nočni razgled nad Kobejem je baje ten million dollar night view.
If I were to read it out loud, I would feel a difference between the Slovenian and the English part. My accent changes—I read the Slovenian in my Slovenian accent and the English in my English accent. If I wanted to, I could read it all in my English accent or (with a bit more effort) all in my Slovenian accent, but that’s not what happens by default. My brain makes the switch automatically. I might be tempted to represent it like this:
Ja, nočni razgled nad Kobejem je baje ten million dollar night view.
To most anglophone readers, of course, the difference was already apparent—it’s a shift from a language they don’t speak to one they do! My adding italics isn’t doing much. But on the off chance you also speak Slovenian, perhaps the italics serve to more accurately record the shift that occurs in the spoken sentence. The distinction in fluency experienced by non-Slovenian speakers has been replaced by one of typography.
Now suppose this sentence was not in Slovenian and English at all, but in two different fictional languages. In a pseudotranslated work, we might represent it like this:
Yeah, the night view of Kobe is meant to be a ten million dollar night view.
…which is a limp, repetitive noodle of a sentence. But what if we reapply our italics from earlier?
Yeah, the night view of Kobe is meant to be a ten million dollar night view.
…which is not much better, but we have at least created an awareness in the reader that the second half is in some way different from the first; that the repetition of “night view” is, perhaps, not just clumsy writing. In other words, we have represented a difference in language within English.
This is by no means a novel observation. There’s a long tradition of writers using typographical tricks to represent different languages. (Chevrons (<>) are a favourite.) A novel in which different languages are represented solely through typography would be a curiosity, but not an especially elegant one. We can, however, imagine subtler expansions of this principle. What if, instead of a typographical distinction, the difference between languages is one of style? This brings us closer to an authentic translation: different languages rendered into English read differently, because of differences in their syntax, grammar, vocabulary. A character whose thoughts are stylistically distinct from their speech is an interesting representation of fluency in different fictional languages!
In the context of made-up words, perhaps typography alone can get us a little further. Here’s the bit of dialogue that inspired this essay, in which a character is describing a concept from her first language while speaking a second language:
“There is a word in the Krej language, [], which I would translate as something like partnership. When two people decide to spend their lives together, that is the word we use, but it can be anyone. Lovers. Siblings. Friends. Strangers, if they’re particularly melodramatic.”
The square brackets were a placeholder for that word I never got around to inventing—but the story is told from this character’s perspective. Shouldn’t words in her first language feel at least as familiar as words in her second? To render them as made-up words is to draw a stark distinction in familiarity that, to her, should not exist—that should in fact run in the opposite direction.
Instead, I came up with this version:
“There is a word in the Krej language, life-share, which literally means “together-beingness” but which I would translate as something like partnership.”
Taking inspiration from my reading of Anathem, I’m coining an English term for this concept. I’m also rendering it in italics to set it apart from the rest of the sentence,7 to indicate to the reader that a switch to a different language has occurred, a point I’m reinforcing with the grammatically awkward literal translation. Finally, the blander term “partnership” speaks to the difficulty in conveying cultural concepts across language barriers. (Why does she not simply render the word as “life-share”? Because that is my representation of the meaning of this fictional word, not hers; from her perspective, she is speaking the word in her own language and, unlike me, does not have the time to come up with a more nuanced translation.)
It’s early days. I don’t know if I’ll keep this construction. No doubt some readers would prefer I simply invented a term and moved on, or sidestepped the issue entirely. But this version engaged my interest in a way that making up a word never did. Immediately it suggests further possibilities—what if I represented this word in two different ways depending on who is speaking? A made-up word for some characters, an English version for others? My solution feels to me like a true expression of pseudotranslation: a translation that is not quite that, that is paradoxically in the same language as the words that surround it.
Call it a gimmick or a literary technique. We writers of secondary worlds are stuck with pseudotranslation—so let’s have some fun with it.
1 We might consider this a cousin or a subset of the broader “translation convention,” whereby e.g. anglophone films set in other countries nevertheless feature characters speaking English. I’m going to stick with “pseudotranslation” because it is specific to both written media and fictional languages.
2 A good way to illustrate this point is through the cognitive dissonance of subtitled movies. When Han Solo has his confrontation with Greedo in the cantina, Greedo might be subtitled as speaking “in an alien language,” or ideally “in Rodian”; but should Han be said to be speaking English or Basic? Either way, a disconnect is created.
3 Consider Raymond Chandler’s parody of science fiction from 1953:https://www.theverge.com/2015/8/12/9136311/raymond-chandler-google-letters
4 Much less smeerps.
5 Or am I? Perhaps the glossary has also been translated…
6 Your mileage may vary—in writing this essay, I discovered that other people have exactly the opposite reaction to the Old Tongue! Regardless, the point remains that it does not resemble English in the way it is said to resemble other languages in the series, i.e. we cannot look at a word in the Old Tongue and guess at its meaning.
7 Here it’s important to emphasise the difference between this decision and the convention of rendering (real) non-English words in italics, a practice to which I am opposed. The necessity for italics here arises from the fact that these are two fictional languages that have both been pseudotranslated into English; the use of non-English words in English prose is an entirely different matter.
© 2024 Filip Hajdar Drnovšek Zorko
