Brandon Sanderon’s Three Laws of Magic


Some may assume that only fantasy settings involve magic, but SF can have it, too. There’s a tendency to assume worlds have one or the other, which begs the question of what happens when planet-hopping SF characters land on one without technology, but magic instead? Regardless, world builders will know if they need or want magic for a world. If so, read on. But before we invent spells or systems of magic, we should understand our options and some theories.

Principles of Good Magic Systems

Before creating a magic system, we should consider some guiding principles.

Sanderson’s Three Laws

Author Brandon Sanderson once proposed three laws of magic that we’ll examine for perspective. Any quotes in this section are directly from his website,, unless otherwise stated, with the rest paraphrased.

Sanderson’s first law is “an author’s ability to solve conflict with magic is directly proportional to how well the reader understands said magic.” Try not to give characters abilities unless we’ve already shown they possess them. Determine how problems can be solved without magic or within the existing magic system, before altering it. And if we do change it, don’t give them abilities that perfectly solve their problem. Breaking this law can make magic too convenient.

Sanderson’s second law is: “limitations are greater than powers.” What characters can’t do forces them to stretch and forces our story in other directions. This adds dimension, depth, and tension, so we gain more with limits than we do by granting powers that just resolve everything.

Sanderson’s third law is: “expand what you already have before you add something new.” We can make our system too complicated if we’re adding new elements rather than refining what we have. We also run the risk of contradiction when the goal for our characters might be achievable with an expansion of our current system.

Soft Magic Vs Hard Magic

Sanderson also discussed what he calls soft vs. hard magic. In his view, soft magic is not rigidly defined (if defined at all), allowing for authors to “preserve the sense of wonder in their books.” By not explaining how magic works, we feel a sense of amazement because we don’t understand how it works or what is possible. It is mysterious and exciting, but also dangerous and unpredictable. It adds tension. Authors who use such systems don’t let magic solve problems for their characters, who seem like a small part of a vast world that can overwhelm them with its power.

By contrast, in hard magic, “the author explicitly describes the rules of magic.” We identify what it can do and what it can’t. Therefore, when a character needs to do something, we aren’t surprised by their ability (or lack thereof). Though it’s a comic book character, his example of Spiderman is a good one: we already know his abilities and simply accept it when he uses one of them, but he doesn’t acquire new abilities on the spur of the moment. He has to use the tools we’ve already defined.

Sanderson also talks about a middle ground, where some things are defined but others aren’t, allowing for flexibility and that feeling of mystery that soft magic allows, while also providing the structure and understanding that hard magic gives.