New Worlds Theory Post: Inventing Trees

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

It’s the fifth Friday of the month, and that means it’s time for a theory post!

This time I’d like to talk about one of the questions that constantly plagues me, which I shorthand as “when should I invent a tree?”

By that I mean, when should I describe a real tree (an oak, a pine, a gingko) versus making one up for my fictional world. Trees are of course not the only thing this applies to; writers also invent flowers, birds, weapons, musical instruments, types of alcohol, and so forth. For simplicity’s sake, I’m going to just say “trees,” with the understanding that it’s metonymy for the whole range of things you might or might not choose to make up.

I generally see three three common reasons why writers invent trees for their worlds. The first, and in my opinion the worst, is that the writer doesn’t actually know much about nature, and finds it easier to just make up names than to do the research. This has the virtue of protecting you against getting it wrong: if you call your tree whitebark, nobody can tell you that birches don’t grow in that kind of environment, or don’t get that big, or any other detail you might have screwed up. Or your invented tree has no analogue at all; it’s just a rannet, which could mean anything you want it to.

I consider this bad for a number of reasons, of which laziness is actually the least important. (Theoretically we shouldn’t be lazy, but let’s face it: inventing a whole world is a lot of work. Of course we’re going to cut corners where we can get away with it.) No, the real problem is that it often leaves the underlying problem untouched, which is that the writer knows bugger-all about nature, and therefore won’t really be integrating it with the story. If they sub in whitebark for birch, it’s probably because they need one feature — its paleness, or the fact that you can write on its bark — and the rest fall by the wayside. It’s even worse with our imaginary rannet tree, because what does that look like? It’s, uh, a tree. With leaves. If you’re lucky, the writer notes its “rough trunk” in passing. The characters wind up traveling through a landscape that might as well be a kid’s crayon drawing, where all the trees are generic brown sticks with green blobs on top. With the bonus issue that if the reader can identify a clear analogue, some of them will be annoyed that you didn’t just say “birch” when you meant birch. (This is known in science fiction as “calling a rabbit a smeerp.”)

The second reason you might invent trees is somewhat better, though it brings its own problems along. You may do it simply because you need the environment to feel alien rather than familiar, populated by strange flora and fauna. If you’re writing science fiction set on another planet, this is a necessary step to take; unless the planet has been terraformed with Earth species, it’s going to have its own kinds of tree (assuming it has trees at all). A secondary world fantasy isn’t required to do the same thing, but it can, for that added touch of distance.

Here’s the catch, though. If you’re going to make up trees, then my feeling is that you have to commit. Don’t give me rannet on one page and oaks on the next, unless there’s a good in-story reason for it (somebody write us a portal fantasy about invasive species!). Give me a properly estranging environment, or give me a familiar one, not half-and-half. And if you do this . . . then refer to my point above. For your rannets to feel real, I need to see in my mind’s eye what they look like, how tall they grow, how their leaves move in the wind. You can’t rely on your reader to fill in from their own personal experience, because they’ve never seen a rannet. And if you don’t provide those details, we’re back to the kid’s crayon drawing.

The third reason is the one I like best, but also in some ways it’s also the clunkiest: you might mix rannets with oaks in your secondary world because you need a tree with some feature that just doesn’t exist in nature. Maybe rannet leaves are purple, and you have a really powerful need for purple leaves in your story. Or maybe they have medicinal properties, like they’re natural antibiotics or hallucinogens. Or they can be dried, ground up, and used as contact explosives. (Don’t walk through a rannet grove in autumn . . .) Nothing in the real world will fit the bill, so you need to invent something.

This is clunky because it’s pretty obvious what you’re doing. I generally prefer to scour the internet looking for something real first, and resort to invention only if I have to; are there really no hallucinogens (tree-based or otherwise) which might plausibly grow in that environment? Does it have to be purple leaves, or could I use a flower instead? Etc. It’s more work, but it also means I get to leverage the recognition factor to make my description more vivid. Only if what I’m doing really just doesn’t work with real-world materials do I resort to making something up.

Mind you, I have a double standard here, which is that I’ll more readily accept invented things the more obviously supernatural they are. Make up a bird called a hopperwill just because you want something whose call sounds like “hop, hop, hop,” and I’ll give you the side-eye. Make up a bird called a gauri whose wings are made from slats of bone (as I did for “Calling Into Silence”) and I’ll nod and think that’s a cool bit of worldbuilding. (How does the gauri fly? It’s a supernatural creature, is how.)

So there’s no clear-cut line for when you should make up a tree (or a flower or a musical instrument or a type of booze) versus when you should use something from the real world. It depends on why you’re doing it, and whether the arguments in favor outweigh those against. But the one thing I can say with conviction is, think about why you’re doing it. Don’t just wander through your story sometimes borrowing and sometimes inventing without putting any thought into the choices you make. Broaden your mental horizons with some research — and then when you make something up, use what you learned from that research to make your creation feel as vivid and concrete as the real thing.

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: Inventing Trees — 20 Comments

  1. The invented ones are the products of magic. If every rannet or whitebark comes with the proviso that you don’t want to sleep under one, and you don’t want to be near the whitebark at night because it turns into a woman with a lethal touch, the rules will be clear enough.

  2. Pingback: New Worlds Theory Post: Inventing Trees - Swan Tower

  3. Excellent points, all.

    Sometimes we invent trees because we underestimate just how weird trees can be. (flowers on their trunk? yep. two leaves and no trunk? yep)…or we’re unsure if its really a tree in anything but name (Mountain Lobelias, Hawaiian Silverswords, …Bananas?)

    • Amusingly, the Silversword was the magical macguffin plant in the fantasy story I made up when I was nine . . . and I’m pretty sure I’d never heard of the real kind. (Nor do I know what was so special about the Silversword in my story, as the idea never went anywhere at all.)

  4. It also helps if the writer knows what microclimate supports the kind of plant or animal they want. Pines don’t do well in tropics. Predators need prey. Everything/everyone needs water. Many writers and would-be writers forget this kind of thing.

    • This is why I tend to lean so hard on copying a real-world environment: because I honestly don’t know much on that kind of thing, which means it’s easier for me to just yoink all the flora and fauna of a real location than try and assemble my own ecology out of bits.

  5. I’d say for me the dividing line is between the rannet grove and the rannet grove in autumn: if you worldbuild anything, you’d better look at it in depth and follow it to its logical conclusions. (You can’t fell those damn things. Gods have mercy if you’re gathering firewood and picking up the wrong kind of stick. My cousin nearly lost an eye to that.)

    I’m perfectly happy mixing rannets and oaks because so much of writing is translation: if something has the appearance and characteristics and fills the general niche of oaks, I’m happy to call it an oak, even if it’s not quercus; I spend a lot of time wandering around Kew, reading labels, and marvelling how wrong I was.

    When it comes to inventions like musical instruments, on the other hand, I *want* to see more variety. There are just so many ways you can hit things, pluck things, or blow air into and around things to make a noise, and every culture is going to arrive at different solutions. I _want_ a window into your culture.

    Last but not least, there’s smeerps, and there are coneys: sometimes an alternate name for a bird, a plant, an animal, conveys something that calling it a rabbit does not.

    • if you worldbuild anything, you’d better look at it in depth and follow it to its logical conclusions

      Yeah, that’s what I mean about the problem with laziness. Lots of the authors I see doing this kind of thing don’t worldbuild around the made-up tree; they drop in a name so they won’t have to worldbuild around its charactersitics.

      sometimes an alternate name for a bird, a plant, an animal, conveys something that calling it a rabbit does not.

      I’d say alternate names and made-up names are slightly different approaches. As you say, one of them conveys something (regional association, or historical, or ethnic subgroup, or whatever), while the other is devoid of association and is just there to try and make the thing sound more “fantastical.”

  6. Of course, there’s a difference between “making up a tree” (which, as Our Gracious Hostess points out, is usually more a result of ignorance than anything else) and “naming a direct-terran-analog tree as part of creating a culture.” That’s what always bugged me about Gunn’s disdain for smeerps — it’s such a snide putdown of language differences and more than a little bit Anglo-imperialistic. (Especially since there’s actually a difference between a “hare” and a “rabbit,” and who knows what filled that niche in Australia before the Europeans arrived…)

    • That’s what always bugged me about Gunn’s disdain for smeerps — it’s such a snide putdown of language differences and more than a little bit Anglo-imperialistic. (Especially since there’s actually a difference between a “hare” and a “rabbit,” and who knows what filled that niche in Australia before the Europeans arrived…)

      I noted in a comment above that I see a difference between using terms that call forth some kind of association, and those which have no association whatsoever and are only a cheap attempt to make the world seem more fantastical. If you call Vachellia xanthophloea a fever tree, you’re going to evoke English colonial experiences in East Africa; if you call it olerai, kimwea, murera, or mwelele you’re going to evoke the cultures of that area, at least for the people who recognize those names; if you call it a muggim, you’re not going to evoke anything for anybody. You can go to the effort of describing the tree so vividly and making it such a part of your characters’ landscape that the reader eventually sees that word as more than a random collection of phonemes, but most authors don’t go to that effort. But avoiding “muggim” or “smeerp” is a different affair than avoiding “olerai” or “mwelele.”

    • Yes, there are differences (and I think the niche was split between small roos, wombats, and maybe bilbies)…but if you use hares or maras simply because you don’t want to say “rabbit,” then isn’t that also bad?
      (though if you mention bilbies(sp) or inventing beeleapers, and all the rest of the flora and fauna is from somewhere else in the world, I’ll be confused – or thinking you’re building up to an ecological confrontation, like in paragraph 7 above)

      On the other hand, if someone mentions Devils Claws, I’d be fine if they didn’t talk about physicality of supernatural evil, belief systems of times past, etc – though I wouldn’t drop the book if they did talk about it.

      So yeah, its a balancing act.

      • but if you use hares or maras simply because you don’t want to say “rabbit,” then isn’t that also bad?

        If that’s your only reason, then yes, it’s bad — because you’re subbing in a different word without regard for its specific meaning and associations. You wind up sounding like Dogberry in Much Ado About Nothing, trying to prove how clever you are and only showing your foolishness. 😛

        though if you mention bilbies(sp) or inventing beeleapers, and all the rest of the flora and fauna is from somewhere else in the world, I’ll be confused – or thinking you’re building up to an ecological confrontation, like in paragraph 7 above

        I’m sort of split on this one. There’s no reason a secondary world couldn’t have a mix of creatures from different regions in the real world; you could easily put horses next to buffalo without first supplying the equivalent of European colonization, or (as I mentioned when talking about food) have tomatoes in your faux-medieval-English society. But the more distinctively regional the names for things are (at least as perceived by Anglophone readers), the more it’s likely to stick out . . .

        • Sadly, I wasn’t thinking of secondary worlds. (with secondaries, it makes perfect sense)
          Or have one of those Victorian/Edwardian periods of exploration without an Age of Exploration — the former is why we have gardens and flowerbeds full of plants from a tiny cluster of mountains in southeastern China.

          Yeah, names might stick the reader’s attention…though, some things get renamed to sound more familiar*, while others get renamed for exotic soundingness (or to sound more tempting…Iceland vs Greenland)…though that would probably be backstory, to be touched on or hinted at, rather than spelled out.

          * = watching _Time Team_, I had to keep reminding myself “when they say the Romans and Normans ate corn, they don’t mean on the cob.” 😀

  7. One word: mallorns. A fully invented elvish tree but one that CAN flourish in our own world (if Middle Earth is the analogue of our own world) because one replaced the Party Tree in the Shire. But TOlkien invented mallorns. I don’t think they’re meant to be an analogue of anything. They’re not tree-smerps, they’re themselves…

    • A fine example of the author making the trees feel real for many readers. But I don’t think it’s coincidence that Tolkien had a well-developed eye for nature in the first place; the things he made up drew on that knowledge.

  8. Elizabeth Moon did a good job of this in The Deed of Paksenarrion I find – familiar, but with just enough invented items (mostly food related I’m noticing) to add the layer of unfamiliarity.

    • Haven’t read her work, but that’s good to know! “The layer of unfamiliarity” is one of the things that can make speculative fiction so enjoyable, when it’s done right.

  9. In SF it’s often the human ecosystem that is invasive. Barrayar, Steerswoman, Vlad Taltos… the latter is fantasy but with hints of SF, plus alien genetic engineers, so you’ve got a mix of human life, native life, and some Jenoine’s science fair project (like most of the people.)

    The you’ve got Gene Wolfe and Anne McCaffrey, using names I think meant to imply “this is like a horse but not exactly”, destriers and riderbeasts or whatnot.

    Seidensticker has an essay claiming that Japanese art was attentive to species details far earlier than European art, that a Japanese tree or bird is likely to be an identifiable species from centuries ago whereas the European tree was usually foliage. He also notes the Tale of Genji describing seasons fairly closely (though not always ‘naturalistically’ — everyone dies in autumn), whereas in Austen or Dickens season and weather only come up with driving the plot (time for a picnic, or to get sick). This despite Austen heroines being much more physically active outdoors than Genji characters.

    • Interesting idea about Japan vs. Europe! I know my general impression of Japanese art is that it was deeply engaged with the specifics of the natural world from the Heian period onward; being able to reference not just a season but particular manifestations of that season for their symbolic connotations is, for example, vital to writing poetry.