New Worlds Theory Post: Fantastical Language

(This post is part of my Patreon-supported New Worlds series.)

I couldn’t fit everything I wanted to say about language in the July theory essay, so for this fifth Friday, we’re going to backtrack to that topic!

Last time we talked about representing accents and dialects on the page, through variant spellings and structures of English. But there’s also the possibility of conlanging: constructing words or a whole grammar for an invented language in your story.

I’m not going to attempt to discuss the mechanics of conlanging in depth here, because it’s a subject that can fill whole books on its own (and has). If this is something you want to pursue, I recommend The Language Construction Kit by Mark Rosenfelder, and I welcome other titles in the comments! Instead we’re going to look at whether you even need to conlang, and how to deploy it when you do.

This subject has gotten more attention lately due to the spate of SF and fantasy films and TV shows, where linguists have been hired to invent Na’vi and Dothraki and a bunch of other languages. But there’s a key difference between that medium and fiction on the page, which is that the former runs in two channels at once — audio and video — while the former runs in only one. The viewer can receive both the sound of the conlang and its meaning at the same time; the reader cannot. Therefore, a wodge of invented language on the screen creates a powerful and effective feeling of dealing with a different culture, but a wodge of invented language on the page creates a reason to skim forward to the next comprehensible bit.

Certainly some authors have managed to deploy their conlangs so effectively that you gain the knowledge you need in the course of reading. I’m told, though I haven’t read it myself, that Richard Adams does this in Watership Down, such that by the end of the novel he can deliver an untranslated line in Lapine and the reader understands it immediately. Very few of us will pull off that trick, though, and so my advice is to be extremely sparing in where you deploy your made-up language.

After all, when do you really need it? If your viewpoint character speaks the language, they’ll understand its meaning; rendering it in the conlang is just obfuscating in a fashion that potentially breaks point of view. If your viewpoint character doesn’t speak the language, how plausible is it that they can pick out the phonemes well enough to spell them properly on the page? If you began talking to me in Hindi or Quechua, I could only transcribe it if you went very slowly, and even then, I would make mistakes. Of course the viewpoint character isn’t actually transcribing, but still: the point is that most of us, upon hearing a language we don’t speak, register it as a smear of incomprehensible sounds. Ergo, especially if you’re writing in first person or close limited third, there’s a good argument for rendering it as “she said something in Sindarin” or “they shouted at me in a language I couldn’t understand.”

Nota bene: Even though the same viewpoint logic applies, I do see greater justification in putting foreign-language dialogue on the page when the language in question is a real one. Regardless of whether the pov character speaks Spanish or not, the reader may, and so giving them the actual words lets them enjoy the pleasure of seeing behind that curtain. But with a conlang, you’ve got a very tall hill to climb before you can assume anybody will have put in the effort to learn your made-up tongue. Until then — i.e. 99% of the time — it’s gibberish to the reader.

Do I mean that you should never bother making up a language? No, certainly not; in fact, in December (the last theory essay for Year Five) I’ll dive into some of my tips for it. But what most authors need is not complete sentences; it’s individual words, given a bit more nuance and structure than the usual “mash some letters together” approach.

Before we get to that, however, I want to swing sideways into the question of how this stuff gets presented on the page — because there’s been a shift in this over time.

Traditionally, style guides say that foreign-language words should be italicized. Lately, though, publishers have been discarding this approach, at least some of the time. The underlying reasons are more complex than I can get into (nor am I the right authority to discuss them in depth to begin with), but briefly, one of the questions raised is: “foreign to whom?” If I talk about a character eating a taco or a burrito, we don’t generally italicize those words, because they’ve been adopted into standard English. If the food they’re eating is a tres leches cake, or was made by the main character’s abuela . . . ? For a Spanish-speaker, that’s no more foreign than me talking about my Nana baking cookies. There’s been a recent trend away from italicizing foreign words in part because deciding what counts as “foreign” tends to privilege the standards of monolingual white Americans. Rather than making constant judgment calls on that front, it’s easier to leave off the typesetting change, which does nothing more than wave a flag the reader may not need.

Turning back to science fiction and fantasy, with all their made-up words . . . again, foreign to whom? These concepts and objects may be new to the reader, but often they’re a familiar part of the world for the viewpoint character. So why italicize them? Maybe we shouldn’t. (Several of my editors recently have encouraged me to de-italicize.) As of me writing this essay, it hasn’t yet reached the point of being a new standard in publishing, but I suspect we’re on our way. The flag doesn’t accomplish much except to tell the reader “don’t worry if you don’t recognize this word” — and does the reader really need that? If the author doesn’t do the work of explaining what the word means, there are problems that italics can’t solve.

Since we’re on the topic of representing speculative concepts in text, I’ll also note the more long-standing swing away from sprinkling capitals on the text like salt. It used to be that ye olde epick phantasie was full of Swords of Destiny taken from the Tomb of the Ancients so the hero could slay the Dragon of Eternity and begin the Age of Peace, but I haven’t seen as much of that lately. These days we’re more likely to say the nobility of a certain realm are starborn rather than Starborn, trusting the word itself to be flavorful without need for a neon sign on the first letter.

But there’s one place where the capitals play a very interesting role, and that’s in talking about ethnic or species groups. In some corners of racial discourse there’s been a movement to capitalize Black, and it calls up questions about how we refer to different populations in SF/F. Sometimes other species are Orcs or Elves, while humans are just lowercase humans (though for Tolkien they were Men: problematic in a different way). Very commonly, aliens are Vulcans or Bajorans, named for their home planet — with an assumption that each planet has only one sentient species — while humans are rarely called Terrans. Of course that carries political implications, since lumping all members of a species together under their planetary name might not be something they appreciate . . . but then, the same is true with non-fictional groups, where talking about Asians obscures the radical differences of culture between, say, Korea, Pakistan, and Russia.

Even when talking about made-up things, how we talk about them can reflect interestingly on the issues of the real world!

The Patreon logo and the text "This post is brought to you by my imaginative backers at Patreon. To join their ranks, click here!"



About Marie Brennan

Marie Brennan is a former anthropologist and folklorist who shamelessly pillages her academic fields for inspiration. She recently misapplied her professors' hard work to the short novel Driftwood and Turning Darkness Into Light, a sequel to the Hugo Award-nominated Victorian adventure series The Memoirs of Lady Trent. She is the author of several other series, over sixty short stories, and the New Worlds series of worldbuilding guides; as half of M.A. Carrick, she has written The Mask of Mirrors, first in the Rook and Rose trilogy. For more information, visit, Twitter @swan_tower, or her Patreon.


New Worlds Theory Post: Fantastical Language — 8 Comments

  1. Pingback: New Worlds Theory Post: Fantastical Language - Swan Tower

  2. Humans are less commonly Terran, but Terran isn’t that rare in SF I’ve read either. Or variants like Tellurians.

    The Liaden books have Liadens, Terrans, and Yxtrang as human types, along with Clutch Turtle, ssussdriad, and norbears.

    Yes, Watership Down gets you to the point of reading a simple sentence. (‘Silflay hraka, u embleer rah,’ replied Bigwig.) Though it doesn’t rely just on context: key words have footnotes or parentheticals defining them. In my epub, they’re italicized on first use but not after.
    I just noticed there’s an early phrase which is translated, but I can’t parse; I don’t know if Adams had more conlang he didn’t reveal, or just made it up.

    Hoi, hoi u embleer Hrair,
    M’ saion ulé hraka vair.’*

  3. The fun really begins when language, meaning, communication are not restricted to either sound or sound-analogs. Consider a gesture of negation, or posture that changes “I disagree with you” to “Designate your second and choose your second.”

    Now throw in dialects that incorporate differing nonverbal aspects (Kris Smith’s Jani Killian series beginning with Code of Conduct is one well-done example). Oh, the fun this can create in the “real world,” too!

  4. Somewhat tangential to language, but it would be interesting to see a post on types of speech which are and are not available in different cultures.

    For example, the large majority of modern works, even if ostensibly set in Ancient Rome or fantasy-middle-ages, assume that most people talk sarcastically at least some of the time. And yet, if I look at 19th century novels, that doesn’t happen. Jane Austen has wonderfully pointed descriptions of Mrs Bennett (the horrible mother in Pride & Prejudice), but no one in the book is ever sarcastic about her failings in direct conversation. Even the narrator is not sarcastic, just vividly descriptive. Charles Dickens describes the horrors of Victorian England without resorting to sarcasm. As far as I can tell from my (limited) knowledge of nineteenth century literature, the authors simply didn’t recognise sarcasm as a potential form of interpersonal communication. I know even less about, say, historical Chinese literature, but I would guess that sarcasm in dialogue is rare to non-existing And yet, most modern authors don’t even query whether sarcasm is appropriate to their fictional culture.*

    Another speech act might be a direct refusal. Some diplomat once said that the Japanese have ten words for ‘no’ and never use any of them. I once learned some Bahasa Indonesia: their verb ‘to refuse’ literally translates as ‘don’t want to’ because they see directly refusing to do something as unthinkably rude, you just drag your feet and quietly don’t do the thing.

    I’m sure there are more examples: speech acts that could theoretically be expressed in a given language but never are. (Or so rarely that they are shocking: where is the fantasy where a foreigner just says ‘no’ to something and all the locals are as shocked as if he urinated in public and it’s a major scandal and has serious consequences?)

    *Shout-out to an author who gets it right, Rachel Neumeier writes the Tuyo series with two realistically different cultures neither of which are anything like modern Western culture, with suitably different speech patterns and, among other things, no sarcasm.

    • Bonus extra question: what is the equivalent of sarcasm to 19th century English? Ie what is a mode of speech which we could use now without any changes to grammar, but which modern English does not use? Anyone?

    • Don’t forget that what we see in surviving written accounts may bear little or no resemblance to oral communication for reasons other than language. One particularly excrutiating example, closely related to “sarcasm,” is the excesses of reputation-protective libel law. “Sarcasm” and “hyperbole” as defenses to an expensive libel suit were simply not available in early-nineteenth-century England. In parallel, the law of that time punished true statements that brought an identifiable individual into disrepute even more harshly than did the libel laws. One of the (many) reasons that Jean-Marie Arouet (known to us as “Voltaire”) lived in Switzerland is that his writings ran afoul of so much French reputation-based law that he was unable to retain a French lawyer; this worked in the eighteenth and nineteenth centuries because “extradition” wasn’t a Thing yet (and especially not in Switzerland).

      Conversely, sarcasm is rampant in private letters from the period that have survived, especially between Men of Letters who didn’t like each other. The key there was that a private letter was not published to another, so it didn’t fall inside of defamation law (common law or in continental Europe). One can only imagine what the written communications that have not been deemed worthy of study by historians, or that have not survived, reveal. “Contemporaneous published fiction” bears at best a peripheral relationship to “contemporaneous discourse.”

Leave a Reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.