Sanderson's First Law

[EUOL]

New Essay Up

http://www.brandonsanderson.com/article.php?id=40

Finally wrote out my ideas on magic formally.  TWG people, would you like to post this as an article?  I'm not sure if writing help would fit with the current atmosphere of the site, which has focused a lot more on reviews and less on articles.  However, if you want it, I'll submit it.

[42]

So,  now that you have that written down, does that mean we'll be hearing your "Brandonism" less?

[Spriggan]

Submit it, the only reason we don't have articles is people never do them anymore.

[Peter Ahlstrom]

I have often marveled to myself at how fascinated I can be by exceptions to made-up rules.

I think I really noticed this first with CJ Cherryh's space books, both the Chanur books and Tripoint. In her universe she sets out rules saying that humans (and hani) need drugs to make it through the jump into hyperspace (they tranq up right before hitting the button, then wake up as soon as the ship comes out of hyperspace--or whatever hyperspace is called in her books, I can't remember at the moment). And then she proceeds to make several characters who can move about freely during hyperspace, doing convenient things like setting the weapons controls to fire right after dropping into realspace. I eat it up. Yet it's a completely artificial exception to a completely artificial rule--the only reason the rule exists at all is because Cherryh said so. Why do I find it so cool that people exist who can break this rule, when I only found out about the rule a few books ago?

I'm not saying she's using a bad storytelling technique or anything like that. It works; I think the way she builds up the rule system and slowly gets you to accept it, then breaks the rules dramatically makes for a very compelling read. I find it interesting that my mind works that way though, that I so easily accept the rules as described and lived by the characters that when the rules get broken (on purpose) it's thrilling to me.

Of course, when writers (apparently) firmly establish rules and then break them withOUT there appearing to be a rhyme or reason to it--making it seem like they just forgot the rules existed rather than that they planned the rule break from the start--a book can really lose me.

[DavidB]

I'm not sure if I agree with this essay. It seems to me that when the heroes use a special ability (like magic) to get out of trouble, there are three discrete possibilities, and not (as the essay contends) a continuum. The possibilities are:

• 1. The reader knew beforehand that the hero had that specific special ability.
  
  For example, if you already established that your hero can turn people into frogs, then it's okay for the hero to get out of trouble by turning a bad guy into a frog. In this case, it doesn't matter whether you've spent pages and pages establishing the rules of magic and how turning people into frogs works because a person's body is a physical manifestation of their ethereal selves and magic lets you mess with stuff on the ethereal plane and blah dee blah, or whether you just said "hey, this guy can magically turn people into frogs" back in chapter one. The ethereal planes explanation might help to flesh out the setting, or it might just get in the way, but either way, once the power is established, your hero is allowed to turn people into frogs to get out of trouble.
  
  I think this is the most common type of magic. Frodo knows the ring turns you invisible, so he can turn invisible in the Prancing Pony. Spiderman can climb walls and shoot webs, because he's Spiderman (and mutant DNA aside, this is basically just the author saying "hey, this guy can shoot webs" back in chapter one). Harry Potter learned a spell to cast a patro-whatever to get rid of the nightmare-monster-thingies, so when the monster-thingies show up, it's okay for him to cast the patro-whatever to get out of trouble.
  

• 2. The reader didn't know that the hero had that specific ability, but could have figured it out.
  
  If you had laid out the rules of magic with physical manifestations and ethereal planes as I described above, but didn't actually say that this meant the hero could turn people into frogs, then the hero could still turn bad guys into frogs to get out of trouble, because the reader (and maybe the hero) could figure out that the system of magic allowed for it.
  
  This is the only situation where you need rules of magic. It also takes some setting up, so it's probably the least common type of magic, though I guess Brandon likes to use it once or twice per novel. Raoden figuring out that drawing a line to turn Elantris into a magic symbol would make the magic work again is an example of this, but his subsequent use of magic to teleport and shoot at bad guys is an example of Type 1 magic, since we already knew the magic could do that stuff.
  

• 3. It's deus ex machina, baby.
  
  Huh. Turns out the hero could make those bad guys into frogs all along. Guess we got all worried for nothing.
  
  This is more common than we would like. (Of course, it's not deus ex machina if the magic isn't used to get the heroes out of trouble. Harry Potter can do any old spell in class, where it doesn't matter, but he can only use the spells that we've already seen to get out of trouble.)

So, having a well-developed magic system can help flesh out your setting, but it doesn't affect your ability to get the heroes out of trouble with magic, as long as you're using Type 1 magic (or Type 3, I guess). If you use Type 2 magic, then you do have to have a well-developed magic system, but there's a discrete cutoff -- the magic system has to be just developed enough to make it clear that the hero can do whatever it is that he actually does.  Any further development is just setting-building.

[dreamking47]

It might be useful in some subsequent essay to define what you mean by "magic."  As I read your explanation in this essay, you're essentially creating the natural sciences of imaginary worlds.

MattD

[Nessa]

I posted it. I also took the liberty of a few minor edits.

[vadia]

given the essay it would be more accurate to say

Sanderson’s First Law of Magics: An author’s ability to solve conflict with magic is DIRECTLY PROPORTIONAL to how well the reader feels that he understands said magic.

And yes, I do believe in gender neutral words -- "he" just happens to be one of them

[Peter Ahlstrom]

It might be useful in some subsequent essay to define what you mean by "magic." As I read your explanation in this essay, you're essentially creating the natural sciences of imaginary worlds.

Yes, that's what he's doing. For example, BioChroma from Warbreaker--the name sounds scientific, because that's what it is. And Mistborn's magic systems are Allomancy, Feruchemy, and Hemalurgy--only one of those has a magic-related name. I think the Mistborn books specifically don't call it magic...there may be some quote that's relevant.

As far as we're concerned, they're magic, but people are scientific beings. Modern science grew out of a desire to understand the workings of the natural world, and many things we would consider magic nowadays like alchemy were studied in centuries past by the same people who developed scientific theories. If something like Allomancy or BioChroma existed in our world, scientists would study it and work out its details.

Plus, consider Clarke's Third Law, that any science (or actually, technology) sufficiently advanced is indistinguishable from magic, and the corollary, that any magic sufficiently unsophistocated is virtually indistinguishable from science.

[EUOL]

DavidB,

One thing I'd point out--this essay is about using magic systems in a GOOD way to further your plot.  That means option three, in my opinion, is not an option.  Deus ex machina may exist in good books, but it's NOT in my opinion ever a good plotting structure, or a good way to use magic.

We do have a continuum--you just left out some options.  One option is to not use your magic to get characters out of their problems at all.  Another option is to use magic only to solve minor conflicts, but to avoid using it in major situations. Another option would be to have everything go wrong if the characters try to use magic to get them out of situations.

Your option two is what I would place on the continuum on the side of hard magic, but not all the way to the end.  Your option one is what I would put all the way to the right side of the line.  The options I outlined above would all be closer to the soft side of the continuum. 

[DavidB]

...option three, in my opinion, is not an option.  Deus ex machina may exist in good books, but it's NOT in my opinion ever a good plotting structure, or a good way to use magic.

No argument here; I completely agree.

We do have a continuum--you just left out some options.  One option is to not use your magic to get characters out of their problems at all.  Another option is to use magic only to solve minor conflicts, but to avoid using it in major situations. Another option would be to have everything go wrong if the characters try to use magic to get them out of situations.

Er...what I was trying to do was characterize individual situations that characters solve with magic. If magic isn't used, or if it doesn't solve the problem, then it doesn't count as a "situation that characters solve with magic", and when you're just looking at one situation at a time, it doesn't really matter how major or minor the conflict is.

My point was that, as long as you're in what I called a type 1 situation, then there's no correlation between how well the reader understands how the magic works, and the author's ability to solve the problem with magic -- in contrast to your Law of Magics. As long as the reader knows that a character has a certain specific ability, then whether there are pages devoted to exploring the rules of magic, or whether that ability is simply stated makes no difference as far as the plot is concerned.

Obviously, I'm not saying that developing rules of magic is bad -- but I am saying that it's not a virtue in and of itself, either. Developing the rules of magic can help flesh out the setting, but if you choose not to develop the rules of magic, you can still use it as much as you like to solve problems, by making sure that all the problems that are solved are type 1 problems. This isn't really much of a limitation because most situations where magic is used in fantasy are of type 1.

The only time when developing the rules of magic directly affects the plot is if you have a type 2 situation. In those cases, the Law of Magics might apply...but I think books with type 2 situations are relatively rare.

EDIT: I should've said this to begin with, but thanks for posting the essay and prompting me to think about this stuff!

[dreamking47]

I am reminded of the scene in Lord of the Rings when the hobbits ask Galadriel if the cloaks they've been given are "magic"...

If something like Allomancy or BioChroma existed in our world, scientists would study it and work out its details.

Yes, and then Michael Crichton would write a techno-thriller about it. 

My point was that, to many people, magic by definition involves the supernatural, that which is not understandable by the scientific inquiry of those who experience it.  Brandon's disagreement at the convention panel was not really a disagreement about what good magic should contain, but rather about what magic is.  Thus it would be good to formally define his understanding of it, so that we can have that shared definition in mind when reading this and his other essays.

MattD

[EUOL]

Er...what I was trying to do was characterize individual situations that characters solve with magic. If magic isn't used, or if it doesn't solve the problem, then it doesn't count as a "situation that characters solve with magic", and when you're just looking at one situation at a time, it doesn't really matter how major or minor the conflict is.

My point was that, as long as you're in what I called a type 1 situation, then there's no correlation between how well the reader understands how the magic works, and the author's ability to solve the problem with magic -- in contrast to your Law of Magics. As long as the reader knows that a character has a certain specific ability, then whether there are pages devoted to exploring the rules of magic, or whether that ability is simply stated makes no difference as far as the plot is concerned.

DavidB,

I understand 1) that you're not arguing against developing magic and 2) that you're not ungrateful or anything like this.  Discussion is good, and your arguments are interesting.

That said, I think we're approaching this from different viewpoints.  Let me discuss the second point you've mentioned before the first.

No correlation, you ask?  For it to be a type 1, as you've outlined, the reader has to understand exactly what the character can do.  If it's spiderman, they have to understand EXACTLY what his powers are.  That's what I'm talking about by 'explaining' the magic.  In Spiderman, the magic is this:  Spiderman can shoot webs.  Spiderman can walk on walls.  Spiderman is extra strong, and has a sense of when danger is going to arrive.  All of those powers are extremely hard magic.

Therefore, Spiderman can use his magic all the time to solve problems, for the reader understands the magic system quite well.  True, it's simple--but I'm not talking about simplicity vs complexity here. 

A softer magic system is one where the rules are more vague.  Take, for instance, Spiderman's 'Spidy sense'.  This is a far softer magic system, since it's not very well defined what its limits are.  It's still not completely soft, since it has obvious rules, but they're more vague. 

Still, all in all, you can say "What are spiderman's powers" and a person familiar with him can reply.  Contrast this to Gandalf.  What are Gandalf's powers?  You can probably name a few, but you can't--like Spiderman--pin down a list of all of them. 

As for the first point (that of it doesn't matter how minor the conflict is) I have to disagree.  I look at this from a writer's standpoint, and narratively, not all climaxes are created equal.  If a character's dog is in danger it's not generally going to be the same as if his son is in danger.  And, the emotional reaction of the reader is what we're working to not undermine here.  If he comes up with a contrived way to save the dog, but then has a very solid, thoughtful, and emotionally poignant way of saving the son, it will be much better a story than if it were the other way around.

[DavidB]

Thanks for the quick response! I think I get it now....

My mistake was in assuming that if a character solves a problem using a magic ability, and the reader didn't know and couldn't figure out beforehand that the character has that ability, then it's automatically deus ex machina. This is wrong; there's another common possibility:

• 2.5) If the reader doesn't know and can't figure out that the character has that ability, but can reasonably guess, then the author might or might not have committed deus ex machina.

For example, if Gandalf Jr. levitated an dog in chapter one, and then in chapter ten gets away from a rampaging elephant by levitating it, this might or might not be deus ex machina, because the reader only knew he could levitate dogs, but some readers might assume that that meant he could levitate elephants as well, while others might assume that elephants would be too heavy for him to levitate.

In this case, Sanderson’s First Law of Magics applies and can be restated as: The more completely the reader understands what a character is capable of doing, the less likely the author is to commit deus ex machina when that character uses his abilities to get out of trouble.

I initially assumed (as Matt apparently did) that Brandon was saying we should be creating natural sciences of imaginary worlds if we wanted to use magic a lot. But this completely misses the point. If Q hands James Bond a watch and says that pushing the button on the watch will let Bond becomes invisible for up to thirty seconds, then in terms of this discussion, we would say that the reader now completely understands the magic, even though there's no explanation of how the watch makes Bond invisible.

By the way, regarding whether it matters how minor the conflict is, we're looking at this from different viewpoints again. I agree that if you use deus ex machina to save the dog, you're likely to end up with a better book than if you used deus ex machina to save the child. What I meant was, the fact that the dog isn't very important doesn't by itself somehow make it not deus ex machina to invent a new ability in order for your character to save it.

[EUOL]

David,

This thing we just discussed needs to be part of the essay, and so I'll include an explanation of what it means to me to have a reader 'understand' a magic system.  There was vagueness written into the essay that I didn't even know was there!  Thanks a ton for taking the time to argue with me. 

In a related note, you may be interested to hear that I've decided to start Midius's viewpoint in Dragonsteel (once I can get back to it) with Hoid already dead, as a certain person suggested in their comments.  (I almost did this the first time I wrote the chapter, and it turns out I should have done that in the first place.)

Keep this up, and I might just have to start sending you ALL of my books beforehand to see what you have to say about them.