How to Write Science Fiction That Earns Its Ideas

The Best Science Fiction Extrapolates One Thing and Holds Everything Else Constant

Le Guin talked about this as a technique: change one variable, see what happens to everything else. The Left Hand of Darkness has one society, no fixed gender. Everything else about the world is basically what we know. Politics, religion, climate, human longing, bureaucratic pettiness. The constraint is what creates the shock of recognition. If everything is different, nothing is strange.

This works the same way a controlled experiment works in a lab. You don't change twelve variables at once and then try to figure out what caused the result. You isolate one thing. You hold everything else steady. Then the effect of that one change becomes visible in a way it couldn't be if the entire landscape were alien.

I'm not sure why so many writers feel the need to change everything. Maybe it feels like more work, more invention, and therefore more impressive. But the opposite is true. A world that differs from ours in one fundamental way is harder to build than a world that differs in fifty ways, because you have to think through every ripple, and you can't hide behind spectacle.

Hard Science Is a Starting Point, Not a Destination

Andy Weir's The Martian spent years as a self-published serial on his blog, where engineering-literate readers would correct his math in the comments and he'd revise the next chapter. The chemistry is real. The orbital mechanics check out. But the book works because the science is in service of a specific human situation: a man alone, with finite resources, refusing to die. Take the personality out and it's a textbook.

Octavia Butler was, by her own admission, bad at science. She wrote some of the most psychologically precise speculative fiction of the 20th century. The Oankali in her Xenogenesis trilogy aren't scientifically plausible in any rigorous sense. But the way humans respond to them, the way they negotiate with beings who are simultaneously offering salvation and taking away autonomy, that reads like a case study in colonial psychology.

The science in science fiction serves the same purpose that gravity serves in architecture. It's the constraint the structure has to work within. Some architects make gravity the point. Some barely acknowledge it. Both can produce buildings that make you stop walking.

Your Protagonist’s Relationship with Technology Reveals Who They Are

Kim Stanley Robinson's Mars trilogy works in part because his three main characters disagree about terraforming. Ann Clayborne wants to preserve Mars as it is. Sax Russell wants to make it livable. Hiroko Ai wants to grow something new that belongs to neither Earth nor old Mars. That disagreement isn't a plot device. It's a disagreement about what "progress" means, what humans owe to worlds they encounter, whether beauty that predates human consciousness has value.

Their relationship to the technology of planetary engineering tells you everything about who they are before Robinson writes a single line of backstory. You know Ann before you know her childhood. The technology is a mirror, and each character sees something different in it.

This is useful even if you're writing near-future fiction with no terraforming in sight. A character who turns their phone off at dinner is telling you something. A character who sleeps next to theirs is telling you something else. The technology doesn't have to be speculative to function as a reveal. It just has to be something the character has a relationship with, something they've chosen or refused or accepted without examining. I don't know what to make of the fact that so few writing guides mention this. Character worksheets will ask about fears, desires, childhood wounds, favorite colors. Almost none of them ask: how does this person relate to the tools that shape their daily life? In science fiction, that question is the whole game.

---

The thing I keep coming back to is that none of these ideas are about inventing something new. They're about looking more carefully at what's already in front of you, at the moment you're living in, at the one question you actually want to answer. Science fiction just gives you permission to follow that question further than realism allows.

Which might be why a daily writing practice matters more for science fiction than for most genres. You need the habit of noticing before you can have something worth extrapolating. A single observation, caught on a Tuesday morning before the coffee kicks in, can become the variable you change while holding everything else constant.

That's what we send writers every morning. One lens to look through before you open the draft.

Get tomorrow's reflection free.